bump version to 10.0.2-4

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

.gitignore

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

Makefile

@@ -1,60 +1,89 @@
-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 = $(PACKAGE)-dbgsym_$(DEB_VERSION_UPSTREAM_REVISION)_$(DEB_BUILD_ARCH).deb
 DEBS = $(DEB) $(DEB_DBG)
 
 all: $(DEBS)
 
 .PHONY: submodule
 submodule:
-	test -f "${SRCDIR}/configure" || git submodule update --init --recursive
+ifeq ($(shell test -f "$(SRCDIR)/configure" && echo 1 || echo 0), 0)
+	git submodule update --init --recursive
+	cd $(SRCDIR); meson subprojects download
+endif
 
-$(BUILDDIR): keycodemapdb | submodule
+PC_BIOS_FW_PURGE_LIST_IN = \
+	hppa-firmware.img \
+	hppa-firmware64.img \
+	openbios-ppc \
+	openbios-sparc32 \
+	openbios-sparc64 \
+	palcode-clipper \
+	s390-ccw.img \
+	s390-netboot.img \
+	u-boot.e500 \
+	.*[a-zA-Z0-9]\.dtb \
+	.*[a-zA-Z0-9]\.dts \
+	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): submodule
 	# 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 $(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
+	rm -rf $@.tmp $@
+	cp -a $(SRCDIR) $@.tmp
+	cp -a debian $@.tmp/debian
+	rm -rf $@.tmp/roms/edk2 # packaged separately
+	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 $(BUILDDIR); dpkg-buildpackage -b -us -uc -j
+$(DEBS) &: $(BUILDDIR)
+	cd $(BUILDDIR); dpkg-buildpackage -b -us -uc
 	lintian $(DEBS)
 
-.PHONY: update
-update:
-	cd $(SRCDIR) && git submodule deinit ui/keycodemapdb || true
-	rm -rf $(SRCDIR)/ui/keycodemapdb
-	mkdir $(SRCDIR)/ui/keycodemapdb
-	cd $(SRCDIR) && git submodule update --init ui/keycodemapdb
-	rm -rf keycodemapdb
-	mkdir keycodemapdb
-	cp -R $(SRCDIR)/ui/keycodemapdb/* keycodemapdb/
-	git add keycodemapdb
+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: upload
+upload: UPLOAD_DIST ?= $(DEB_DISTRIBUTION)
 upload: $(DEBS)
-	tar cf - ${DEBS} | ssh repoman@repo.proxmox.com upload --product pve --dist bullseye
+	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

@@ -1,3 +1,382 @@
+pve-qemu-kvm (10.0.2-4) trixie; urgency=medium
+
+  * savevm-async: reuse migration blocker check for snapshots/hibernation to
+    avoid crashing a VM when on these actions if its configuration does not
+    support them.
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 17 Jul 2025 23:08:46 +0200
+
+pve-qemu-kvm (10.0.2-3) trixie; urgency=medium
+
+  * add backup/zeroinit/track-alloc blockdev patches
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 03 Jul 2025 12:02:03 +0200
+
+pve-qemu-kvm (10.0.2-2) trixie; urgency=medium
+
+  * drop support for accessing Gluster based storage directly due to its
+    effective end of support and maintenance. The last upstream release
+    happened over 2.5 years ago and there's currently no one providing
+    enterprise support or security updates. Further, upstream QEMU will remove
+    the integration in one of the next releases, so use the upcomming PVE 9
+    major release to provide a clean cut.
+    User can either stay on Proxmox VE 8 until its end-of-life (probably end
+    of June 2026), or mount GlusterFS "manually" (e.g., /etc/fstab) and add it
+    as directory storage to Proxmox VE.
+    We recommend moving to other actively maintained storage technology
+    altogether though.
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 25 Jun 2025 19:23:30 +0200
+
+pve-qemu-kvm (10.0.2-1) trixie; urgency=medium
+
+  * update QEMU and downstream patches for 10.0.2 release.
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 17 Jun 2025 08:38:51 +0200
+
+pve-qemu-kvm (9.2.0-5) bookworm; urgency=medium
+
+  * pve backup: backup-access api: simplify bitmap logic
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 04 Apr 2025 16:15:58 +0200
+
+pve-qemu-kvm (9.2.0-4) bookworm; urgency=medium
+
+  * various async snapshot improvements, inclduing using a dedicated IO thread
+    for the state file when doing a live snapshot. That should reduce load on
+    the main thread and for it to get stuck on IO, i.e. same benefits as using
+    a dedicated IO thread for regular drives. This is particularly interesting
+    when the VM state storage is a network storage like NFS. It should also
+    address #6262.
+
+  * pve backup: implement basic features and API in preperation for external
+    backup provider storage plugins.
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 03 Apr 2025 17:00:34 +0200
+
+pve-qemu-kvm (9.2.0-3) bookworm; urgency=medium
+
+  * revert changes to the High Precision Event Timer (HPET) to fix performance
+    regression
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 26 Mar 2025 09:56:01 +0100
+
+pve-qemu-kvm (9.2.0-2) bookworm; urgency=medium
+
+  * fix assertion failure when migrating a VM with multiple disks on a
+    replicated ZFS.
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 24 Feb 2025 17:33:34 +0100
+
+pve-qemu-kvm (9.2.0-1) bookworm; urgency=medium
+
+  * update submodule and patches to QEMU 9.2.0
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 04 Feb 2025 08:49:20 +0100
+
+pve-qemu-kvm (9.1.2-3) bookworm; urgency=medium
+
+  * async snapshot: explicitly specify raw format when loading the VM state
+    file
+
+  * vma create: rework CLI parameters for passing disk to a more structured
+    style and use that to allow explicitly specifying the format
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 24 Jan 2025 16:12:34 +0100
+
+pve-qemu-kvm (9.1.2-2) bookworm; urgency=medium
+
+  * adapt machine version deprecation for Proxmox VE release and support
+    cycle.
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 17 Jan 2025 16:34:06 +0100
+
+pve-qemu-kvm (9.1.2-1) bookworm; urgency=medium
+
+  * update submodule and patches to QEMU 9.1.2
+
+  * improve error handling and edge cases with fleecing backups.
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 11 Dec 2024 16:47:21 +0100
+
+pve-qemu-kvm (9.0.2-4) bookworm; urgency=medium
+
+  * async snapshot: ensure any dynamic vCPU-throttling applied for
+    auto-converge gets always disabled again after finishing the snapshot.
+
+ -- Proxmox Support Team <support@proxmox.com>  Sun, 10 Nov 2024 11:23:09 +0100
+
+pve-qemu-kvm (9.0.2-3) bookworm; urgency=medium
+
+  * pick up fix for VirtIO PCI regressions
+
+  * pick up stable fixes for 9.0, including fixes for VirtIO-net, ARM and
+    x86(_64) emulation, CVEs to harden NBD server against malicious clients,
+    as well as a few others (VNC, physmem, Intel IOMMU, ...).
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 06 Sep 2024 16:21:42 +0200
+
+pve-qemu-kvm (9.0.2-2) bookworm; urgency=medium
+
+  * actually update submodule to QEMU 9.0.2. The previous release was still
+    based on 9.0.0 by mistake.
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 07 Aug 2024 10:16:01 +0200
+
+pve-qemu-kvm (9.0.2-1) bookworm; urgency=medium
+
+  * update submodule and patches to QEMU 9.0.2. While our version had most
+    stable fixes included already, there are new fixes for VirtIO and VGA
+    display screen blanking (#4786)
+
+  * backport fix for a regression with the LSI-53c895a controller and one for
+    the boot order getting ignored for USB storage
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 29 Jul 2024 18:59:40 +0200
+
+pve-qemu-kvm (9.0.0-6) bookworm; urgency=medium
+
+  * fix a regression in the zeroinit block driver that prevented importing and
+    cloning disks to RBD storages which are not using the krbd setting
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 08 Jul 2024 16:11:15 +0200
+
+pve-qemu-kvm (9.0.0-5) bookworm; urgency=medium
+
+  * backport fix for CVE-2024-4467 to prevent malicious qcow2 image files from
+    already causing bad effects if being queried via 'qemu-img info'. For
+    Proxmox VE, this is an additional safe guard, as currently it directly
+    creates and manages the qcow2 images used by VMs and does not allow
+    unprivileged users to import them
+
+  * fix #4726: code cleanup: avoid superfluous check in vma backup code
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 03 Jul 2024 13:13:35 +0200
+
+pve-qemu-kvm (9.0.0-4) bookworm; urgency=medium
+
+  * fix crash after saving a snapshot without including VM state when a VirtIO
+    block device with iothread is configured.
+
+  * fix edge case in error handling when opening a block device from PBS fails
+
+  * minor code cleanup in backup code
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 01 Jul 2024 11:26:11 +0200
+
+pve-qemu-kvm (9.0.0-3) bookworm; urgency=medium
+
+  * fix crash when doing resize after hotplugging a disk using io_uring
+
+  * fix some minor issues in software CPU emulation (i.e. non-KVM) for ARM and
+    x86(_64)
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 29 May 2024 15:55:44 +0200
+
+pve-qemu-kvm (9.0.0-2) bookworm; urgency=medium
+
+  * fix #5409: backup: fix copy-before-write timeout
+
+  * backup: improve error when copy-before-write fails for fleecing
+
+  * fix forwards and backwards migration with VirtIO-GPU display
+
+  * fix a regression in pflash device introduced in 8.2
+
+  * revert a commit for VirtIO PCI devices that turned out to cause more
+    potential security issues than what it fixed
+
+  * move compatibility flags for a new VirtIO-net feature to the correct
+    machine type. The feature was introduced in QEMU 8.2, but the
+    compatibility flags got added to machine version 8.0 instead of 8.1. This
+    breaks backwards migration with machine version 8.1 from a 8.2/9.0 binary
+    to an 8.1 binary, in cases where the guest kernel enables the feature
+    (e.g. Ubuntu 23.10).
+    While that breaks migration with machine version 8.1 from an unpatched to
+    a patched binary, Proxmox VE only ever had 8.2 on the test repository and
+    9.0 not yet in any public repository.
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 17 May 2024 17:04:52 +0200
+
+pve-qemu-kvm (9.0.0-1) bookworm; urgency=medium
+
+  * update submodule and patches to QEMU 9.0.0
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 29 Apr 2024 10:51:37 +0200
+
+pve-qemu-kvm (8.2.2-1) bookworm; urgency=medium
+
+  * update submodule and patches to QEMU 8.2.2
+
+ -- Proxmox Support Team <support@proxmox.com>  Sat, 27 Apr 2024 12:44:30 +0200
+
+pve-qemu-kvm (8.1.5-5) bookworm; urgency=medium
+
+  * implement support for backup fleecing
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 11 Apr 2024 17:46:48 +0200
+
+pve-qemu-kvm (8.1.5-4) bookworm; urgency=medium
+
+  * fix live-import for certain kinds of VMDK images that rely on padding
+
+  * backup: avoid bubbling up first error if it's an ECANCELED one, as those
+    are often a result of cancling the job due to running into an actual
+    issue.
+
+  * backup: factor out & clean up gathering device info into helper
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 12 Mar 2024 14:08:40 +0100
+
+pve-qemu-kvm (8.1.5-3) bookworm; urgency=medium
+
+  * backport fix for potential deadlock during QMP stop command if the VM has
+    disks attached through VirtIO-Block and IO-Thread enabled
+
+  * fix #4507: add patch to automatically increase NOFILE soft limit
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 21 Feb 2024 20:11:23 +0100
+
+pve-qemu-kvm (8.1.5-2) bookworm; urgency=medium
+
+  * work around for a situation where guest IO might get stuck, if the VM is
+    configure  with iothread and VirtIO block/SCSI
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 02 Feb 2024 19:41:27 +0100
+
+pve-qemu-kvm (8.1.5-1) bookworm; urgency=medium
+
+  * update to 8.1.5 stable release, including more relevant fixes like:
+    - virtio-net: correctly copy vnet header when flushing TX
+    - hw/pflash: implement update buffer for block writes
+    - Fixes to i386 emulation and ARM emulation.
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 02 Feb 2024 19:08:13 +0100
+
+pve-qemu-kvm (8.1.2-6) bookworm; urgency=medium
+
+  * revert attempted fix to avoid rare issue with stuck guest IO when using
+    iothread, because it caused a much more common issue with iothreads
+    consuming too much CPU
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 15 Dec 2023 14:22:06 +0100
+
+pve-qemu-kvm (8.1.2-5) bookworm; urgency=medium
+
+  * backport workaround for stuck guest IO with iothread and VirtIO block/SCSI
+    in some rare edge cases
+
+  * backport fix for potential deadlock when issuing the "resize" QMP command
+    for a disk that is using iothread
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 11 Dec 2023 16:58:27 +0100
+
+pve-qemu-kvm (8.1.2-4) bookworm; urgency=medium
+
+  * fix vnc clipboard in the host to guest direction
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 22 Nov 2023 14:28:21 +0100
+
+pve-qemu-kvm (8.1.2-3) bookworm; urgency=medium
+
+  * fix #5054: backport fix for software reset with SATA, avoiding breakage
+    with, e.g., some FreeBSD VMs
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 20 Nov 2023 10:24:50 +0100
+
+pve-qemu-kvm (8.1.2-2) bookworm; urgency=medium
+
+  * revert "x86: acpi: workaround Windows not handling name references in
+    Package properly" as that seems to have broken networking (and possibly
+    other things) one some localized variants of Windows (e.g., the German
+    versions).
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 17 Nov 2023 11:55:23 +0100
+
+pve-qemu-kvm (8.1.2-1) bookworm; urgency=medium
+
+  * update submodule and patches to QEMU 8.1.2
+
+  * use QEMU's keycode-map-db again instead of our static copy from QEMU 6.0
+
+  * disable graph locking, newly introduced in the 8.1 release, as it has
+    still various deadlock issuess, e.g., during canceling backup jobs.
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 24 Oct 2023 13:42:45 +0200
+
+pve-qemu-kvm (8.0.2-7) bookworm; urgency=medium
+
+  * fix #2874: SATA: avoid unsolicited write to sector 0 during reset
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 04 Oct 2023 08:33:35 +0200
+
+pve-qemu-kvm (8.0.2-6) bookworm; urgency=medium
+
+  * fix #1534: vma: add extract-filter for disk images allowing users to pass
+    a comma separated list of the disks they want to extract from an archive.
+
+  * backup: create jobs in a drained section to avoid subtle bugs where
+    something interferes with the block-copy-state bitmap on initialization
+
+  * backup: drop experimental, and since a while also fully broken, directory
+    backup format (BACKUP_FORMAT_DIR). This format was never exposed via the
+    Proxmox VE API, but only available via QMP, as its broken since QEMU 8 and
+    we got zero reports about that, it's safe to assume that there are no
+    public users, so just remove it completely.
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 06 Sep 2023 17:03:59 +0200
+
+pve-qemu-kvm (8.0.2-5) bookworm; urgency=medium
+
+  * improve memory footprint after backup by not keeping as much memory
+    resident.
+
+  * fix file descriptor leak for vhost (used by default by vNICs).
+
+ -- Proxmox Support Team <support@proxmox.com>  Wed, 16 Aug 2023 11:52:24 +0200
+
+pve-qemu-kvm (8.0.2-4) bookworm; urgency=medium
+
+  * fix resume for snapshot and hibernate in combination with iothread and
+    dirty bitmap
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 28 Jul 2023 12:58:22 +0200
+
+pve-qemu-kvm (8.0.2-3) bookworm; urgency=medium
+
+  * fix regression in QEMU 8.0 for drive mirror with bitmap
+
+ -- Proxmox Support Team <support@proxmox.com>  Thu, 15 Jun 2023 13:57:46 +0200
+
+pve-qemu-kvm (8.0.2-2) bookworm; urgency=medium
+
+  * drop custom get_link_status QMP command, was never really used.
+
+  * drop custom & deprecated drive snapshot QMP commands, we use a better
+    alternative since a while.
+
+ -- Proxmox Support Team <support@proxmox.com>  Fri, 09 Jun 2023 07:57:56 +0200
+
+pve-qemu-kvm (8.0.2-1) bookworm; urgency=medium
+
+  * update to QEMU stable release 8.0.2
+
+  * update patches for avoiding issues with DMA reentrancy to current,
+    slightly optimized version.
+
+ -- Proxmox Support Team <support@proxmox.com>  Tue, 06 Jun 2023 16:34:50 +0200
+
+pve-qemu-kvm (8.0.0-1) bookworm; urgency=medium
+
+  * update to QEMU stable release 8.0.0
+
+  * re-build for Proxmox VE 8 / Debian 12 Bookworm
+
+  * adapt to the local virtiofsd C variant being dropped, it has been
+    rewritten in Rust and is now hosted in a separate source repository.
+
+ -- Proxmox Support Team <support@proxmox.com>  Mon, 22 May 2023 13:45:49 +0200
+
 pve-qemu-kvm (7.2.0-8) bullseye; urgency=medium
 
   * backport fix for ACPI CPU hotplug issue with TCG

debian/compat

@@ -1 +0,0 @@
-10

debian/control

@@ -2,9 +2,8 @@ Source: pve-qemu-kvm
 Section: admin
 Priority: optional
 Maintainer: Proxmox Support Team <support@proxmox.com>
-Build-Depends: autotools-dev,
+Build-Depends: debhelper-compat (= 13),
                check,
-               debhelper (>= 9),
                libacl1-dev,
                libaio-dev,
                libattr1-dev,
@@ -13,7 +12,6 @@ Build-Depends: autotools-dev,
                libepoxy-dev,
                libfdt-dev,
                libgbm-dev,
-               libglusterfs-dev (>= 5.2-2),
                libgnutls28-dev,
                libiscsi-dev (>= 1.12.0),
                libjpeg-dev,
@@ -21,7 +19,7 @@ Build-Depends: autotools-dev,
                libnuma-dev,
                libpci-dev,
                libpixman-1-dev,
-               libproxmox-backup-qemu0-dev (>= 1.3.0-1),
+               libproxmox-backup-qemu0-dev (>= 1.3.0),
                librbd-dev (>= 0.48),
                libsdl1.2-dev,
                libseccomp-dev,
@@ -30,7 +28,7 @@ Build-Depends: autotools-dev,
                libspice-server-dev (>= 0.14.0~),
                libsystemd-dev,
                liburing-dev,
-               libusb-1.0-0-dev (>= 1.0.17-1),
+               libusb-1.0-0-dev (>= 1.0.17),
                libusbredirparser-dev (>= 0.6-2),
                libvirglrenderer-dev,
                libzstd-dev,
@@ -38,9 +36,8 @@ Build-Depends: autotools-dev,
                python3-minimal,
                python3-sphinx,
                python3-sphinx-rtd-theme,
+               python3-venv,
                quilt,
-               texi2html,
-               texinfo,
                uuid-dev,
                xfslibs-dev,
 Standards-Version: 3.7.2
@@ -49,13 +46,6 @@ Package: pve-qemu-kvm
 Architecture: any
 Depends: ceph-common (>= 0.48),
          iproute2,
-         libgfapi0 | glusterfs-common (>= 5.6),
-         libgfchangelog0 | glusterfs-common (>= 5.6),
-         libgfdb0 | glusterfs-common (>= 5.6),
-         libgfrpc0 | glusterfs-common (>= 5.6),
-         libgfxdr0 | glusterfs-common (>= 5.6),
-         libglusterfs-dev | glusterfs-common (>= 5.6),
-         libglusterfs0 | glusterfs-common (>= 5.6),
          libiscsi4 (>= 1.12.0) | libiscsi7,
          libjpeg62-turbo,
          libspice-server1 (>= 0.14.0~),
@@ -64,8 +54,8 @@ Depends: ceph-common (>= 0.48),
          libuuid1,
          ${misc:Depends},
          ${shlibs:Depends},
-Recommends: numactl
-Suggests: libgl1
+Recommends: numactl,
+Suggests: libgl1,
 Conflicts: kvm,
            pve-kvm,
            pve-qemu-kvm-2.6.18,
@@ -73,23 +63,17 @@ Conflicts: kvm,
            qemu-kvm,
            qemu-system-arm,
            qemu-system-common,
-           qemu-system-x86,
            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,
           qemu-system-x86,
           qemu-utils,
+Breaks: qemu-server (<= 8.0.6)
 Description: Full virtualization on x86 hardware
  Using KVM, one can run multiple virtual PCs, each running unmodified Linux or
  Windows images. Each virtual machine has private virtualized hardware: a
  network card, disk, graphics adapter, etc.
-
-Package: pve-qemu-kvm-dbg
-Architecture: any
-Section: debug
-Depends: pve-qemu-kvm (= ${binary:Version})
-Description: pve qemu debugging symbols
- This package contains the debugging symbols for pve-qemu-kvm.

debian/parse-machines.pl

@@ -24,4 +24,5 @@ while (<STDIN>) {
 
 die "no QEMU machine types detected from STDIN input" if scalar (@$machines) <= 0;
 
-print to_json($machines, { utf8 => 1 }) or die "$!\n";
+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

@@ -27,16 +27,18 @@ 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>
+[FE: rebased for 9.1.2]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- block/mirror.c                         | 98 +++++++++++++++++++++-----
- blockdev.c                             | 39 +++++++++-
+ block/mirror.c                         | 99 ++++++++++++++++++++------
+ blockdev.c                             | 38 +++++++++-
  include/block/block_int-global-state.h |  4 +-
- qapi/block-core.json                   | 29 ++++++--
+ qapi/block-core.json                   | 25 ++++++-
  tests/unit/test-block-iothread.c       |  4 +-
- 5 files changed, 145 insertions(+), 29 deletions(-)
+ 5 files changed, 142 insertions(+), 28 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index 251adc5ae0..8ead5f77a0 100644
+index a53582f17b..fafca1360e 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
 @@ -51,7 +51,7 @@ typedef struct MirrorBlockJob {
@@ -48,7 +50,7 @@ index 251adc5ae0..8ead5f77a0 100644
      BlockMirrorBackingMode backing_mode;
      /* Whether the target image requires explicit zero-initialization */
      bool zero_target;
-@@ -65,6 +65,8 @@ typedef struct MirrorBlockJob {
+@@ -73,6 +73,8 @@ typedef struct MirrorBlockJob {
      size_t buf_size;
      int64_t bdev_length;
      unsigned long *cow_bitmap;
@@ -57,9 +59,9 @@ index 251adc5ae0..8ead5f77a0 100644
      BdrvDirtyBitmap *dirty_bitmap;
      BdrvDirtyBitmapIter *dbi;
      uint8_t *buf;
-@@ -699,7 +701,8 @@ static int mirror_exit_common(Job *job)
-     bdrv_child_refresh_perms(mirror_top_bs, mirror_top_bs->backing,
+@@ -723,7 +725,8 @@ static int mirror_exit_common(Job *job)
                               &error_abort);
+ 
      if (!abort && s->backing_mode == MIRROR_SOURCE_BACKING_CHAIN) {
 -        BlockDriverState *backing = s->is_none_mode ? src : s->base;
 +        BlockDriverState *backing;
@@ -67,7 +69,7 @@ index 251adc5ae0..8ead5f77a0 100644
          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)
+@@ -824,6 +827,16 @@ static void mirror_abort(Job *job)
      assert(ret == 0);
  }
  
@@ -84,7 +86,7 @@ index 251adc5ae0..8ead5f77a0 100644
  static void coroutine_fn mirror_throttle(MirrorBlockJob *s)
  {
      int64_t now = qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
-@@ -977,7 +990,8 @@ static int coroutine_fn mirror_run(Job *job, Error **errp)
+@@ -1020,7 +1033,8 @@ 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);
@@ -94,7 +96,7 @@ index 251adc5ae0..8ead5f77a0 100644
          ret = mirror_dirty_init(s);
          if (ret < 0 || job_is_cancelled(&s->common.job)) {
              goto immediate_exit;
-@@ -1224,6 +1238,7 @@ static const BlockJobDriver mirror_job_driver = {
+@@ -1309,6 +1323,7 @@ static const BlockJobDriver mirror_job_driver = {
          .run                    = mirror_run,
          .prepare                = mirror_prepare,
          .abort                  = mirror_abort,
@@ -102,7 +104,7 @@ index 251adc5ae0..8ead5f77a0 100644
          .pause                  = mirror_pause,
          .complete               = mirror_complete,
          .cancel                 = mirror_cancel,
-@@ -1240,6 +1255,7 @@ static const BlockJobDriver commit_active_job_driver = {
+@@ -1327,6 +1342,7 @@ static const BlockJobDriver commit_active_job_driver = {
          .run                    = mirror_run,
          .prepare                = mirror_prepare,
          .abort                  = mirror_abort,
@@ -110,7 +112,7 @@ index 251adc5ae0..8ead5f77a0 100644
          .pause                  = mirror_pause,
          .complete               = mirror_complete,
          .cancel                 = commit_active_cancel,
-@@ -1627,7 +1643,10 @@ static BlockJob *mirror_start_job(
+@@ -1719,7 +1735,10 @@ static BlockJob *mirror_start_job(
                               BlockCompletionFunc *cb,
                               void *opaque,
                               const BlockJobDriver *driver,
@@ -121,10 +123,10 @@ index 251adc5ae0..8ead5f77a0 100644
 +                             BlockDriverState *base,
                               bool auto_complete, const char *filter_node_name,
                               bool is_mirror, MirrorCopyMode copy_mode,
-                              Error **errp)
-@@ -1639,10 +1658,39 @@ static BlockJob *mirror_start_job(
-     uint64_t target_perms, target_shared_perms;
-     int ret;
+                              bool base_ro,
+@@ -1734,10 +1753,39 @@ static BlockJob *mirror_start_job(
+ 
+     GLOBAL_STATE_CODE();
  
 -    if (granularity == 0) {
 -        granularity = bdrv_get_default_bitmap_granularity(target);
@@ -164,7 +166,7 @@ index 251adc5ae0..8ead5f77a0 100644
      assert(is_power_of_2(granularity));
  
      if (buf_size < 0) {
-@@ -1774,7 +1822,9 @@ static BlockJob *mirror_start_job(
+@@ -1878,7 +1926,9 @@ static BlockJob *mirror_start_job(
      s->replaces = g_strdup(replaces);
      s->on_source_error = on_source_error;
      s->on_target_error = on_target_error;
@@ -174,10 +176,10 @@ index 251adc5ae0..8ead5f77a0 100644
 +    s->bitmap_mode = bitmap_mode;
      s->backing_mode = backing_mode;
      s->zero_target = zero_target;
-     s->copy_mode = copy_mode;
-@@ -1795,6 +1845,18 @@ static BlockJob *mirror_start_job(
-         bdrv_disable_dirty_bitmap(s->dirty_bitmap);
-     }
+     qatomic_set(&s->copy_mode, copy_mode);
+@@ -1904,6 +1954,18 @@ static BlockJob *mirror_start_job(
+      */
+     bdrv_disable_dirty_bitmap(s->dirty_bitmap);
  
 +    if (s->sync_bitmap) {
 +        bdrv_dirty_bitmap_set_busy(s->sync_bitmap, true);
@@ -191,10 +193,10 @@ index 251adc5ae0..8ead5f77a0 100644
 +        }
 +    }
 +
+     bdrv_graph_wrlock();
      ret = block_job_add_bdrv(&s->common, "source", bs, 0,
                               BLK_PERM_WRITE_UNCHANGED | BLK_PERM_WRITE |
-                              BLK_PERM_CONSISTENT_READ,
-@@ -1872,6 +1934,9 @@ fail:
+@@ -1986,6 +2048,9 @@ fail:
          if (s->dirty_bitmap) {
              bdrv_release_dirty_bitmap(s->dirty_bitmap);
          }
@@ -204,7 +206,7 @@ index 251adc5ae0..8ead5f77a0 100644
          job_early_fail(&s->common.job);
      }
  
-@@ -1889,31 +1954,25 @@ void mirror_start(const char *job_id, BlockDriverState *bs,
+@@ -2008,35 +2073,28 @@ 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,
@@ -229,19 +231,23 @@ index 251adc5ae0..8ead5f77a0 100644
 -                   MirrorSyncMode_str(mode));
 -        return;
 -    }
+-
+     bdrv_graph_rdlock_main_loop();
 -    is_none_mode = mode == MIRROR_SYNC_MODE_NONE;
      base = mode == MIRROR_SYNC_MODE_TOP ? bdrv_backing_chain_next(bs) : NULL;
+     bdrv_graph_rdunlock_main_loop();
+ 
      mirror_start_job(job_id, bs, creation_flags, target, replaces,
                       speed, granularity, buf_size, backing_mode, zero_target,
                       on_source_error, on_target_error, unmap, NULL, NULL,
 -                     &mirror_job_driver, is_none_mode, base, false,
--                     filter_node_name, true, copy_mode, errp);
+-                     filter_node_name, true, copy_mode, false, errp);
 +                     &mirror_job_driver, mode, bitmap, bitmap_mode, base,
-+                     false, filter_node_name, true, copy_mode, errp);
++                     false, filter_node_name, true, copy_mode, false, errp);
  }
  
  BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
-@@ -1940,7 +1999,8 @@ BlockJob *commit_active_start(const char *job_id, BlockDriverState *bs,
+@@ -2063,7 +2121,8 @@ 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,
@@ -249,36 +255,35 @@ index 251adc5ae0..8ead5f77a0 100644
 +                     &commit_active_job_driver, MIRROR_SYNC_MODE_FULL,
 +                     NULL, 0, base, auto_complete,
                       filter_node_name, false, MIRROR_COPY_MODE_BACKGROUND,
-                      errp);
+                      base_read_only, errp);
      if (!job) {
 diff --git a/blockdev.c b/blockdev.c
-index 3f1dec6242..2ee30323cb 100644
+index 1d1f27cfff..ec45bbaa52 100644
 --- a/blockdev.c
 +++ b/blockdev.c
-@@ -2946,6 +2946,10 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -2797,6 +2797,9 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
                                     BlockDriverState *target,
-                                    bool has_replaces, const char *replaces,
+                                    const char *replaces,
                                     enum MirrorSyncMode sync,
-+                                   bool has_bitmap,
 +                                   const char *bitmap_name,
 +                                   bool has_bitmap_mode,
 +                                   BitmapSyncMode bitmap_mode,
                                     BlockMirrorBackingMode backing_mode,
                                     bool zero_target,
                                     bool has_speed, int64_t speed,
-@@ -2965,6 +2969,7 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -2815,6 +2818,7 @@ 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;
-@@ -3019,6 +3024,29 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+     GLOBAL_STATE_CODE();
+     GRAPH_RDLOCK_GUARD_MAINLOOP();
+@@ -2869,6 +2873,29 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
          sync = MIRROR_SYNC_MODE_FULL;
      }
  
-+    if (has_bitmap) {
++    if (bitmap_name) {
 +        if (granularity) {
 +            error_setg(errp, "Granularity and bitmap cannot both be set");
 +            return;
@@ -301,53 +306,53 @@ index 3f1dec6242..2ee30323cb 100644
 +        }
 +    }
 +
-     if (!has_replaces) {
+     if (!replaces) {
          /* We want to mirror from @bs, but keep implicit filters on top */
          unfiltered_bs = bdrv_skip_implicit_filters(bs);
-@@ -3065,8 +3093,8 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -2910,8 +2937,8 @@ 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,
--                 has_replaces ? replaces : NULL, job_flags,
+-                 replaces, job_flags,
 -                 speed, granularity, buf_size, sync, backing_mode, zero_target,
-+                 has_replaces ? replaces : NULL, job_flags, speed, granularity,
-+                 buf_size, sync, bitmap, bitmap_mode, backing_mode, zero_target,
++                 replaces, job_flags, speed, granularity, buf_size, sync,
++                 bitmap, bitmap_mode, backing_mode, zero_target,
                   on_source_error, on_target_error, unmap, filter_node_name,
                   copy_mode, errp);
  }
-@@ -3211,6 +3239,8 @@ void qmp_drive_mirror(DriveMirror *arg, Error **errp)
+@@ -3055,6 +3082,8 @@ 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,
-+                           arg->has_bitmap, arg->bitmap,
+     blockdev_mirror_common(arg->job_id, bs, target_bs,
+                            arg->replaces, arg->sync,
++                           arg->bitmap,
 +                           arg->has_bitmap_mode, arg->bitmap_mode,
                             backing_mode, zero_target,
                             arg->has_speed, arg->speed,
                             arg->has_granularity, arg->granularity,
-@@ -3232,6 +3262,8 @@ void qmp_blockdev_mirror(bool has_job_id, const char *job_id,
+@@ -3074,6 +3103,8 @@ void qmp_blockdev_mirror(const char *job_id,
                           const char *device, const char *target,
-                          bool has_replaces, const char *replaces,
+                          const char *replaces,
                           MirrorSyncMode sync,
-+                         bool has_bitmap, const char *bitmap,
++                         const char *bitmap,
 +                         bool has_bitmap_mode, BitmapSyncMode bitmap_mode,
                           bool has_speed, int64_t speed,
                           bool has_granularity, uint32_t granularity,
                           bool has_buf_size, int64_t buf_size,
-@@ -3281,7 +3313,8 @@ void qmp_blockdev_mirror(bool has_job_id, const char *job_id,
+@@ -3114,7 +3145,8 @@ void qmp_blockdev_mirror(const char *job_id,
      }
  
-     blockdev_mirror_common(has_job_id ? job_id : NULL, bs, target_bs,
--                           has_replaces, replaces, sync, backing_mode,
-+                           has_replaces, replaces, sync, has_bitmap,
+     blockdev_mirror_common(job_id, bs, target_bs,
+-                           replaces, sync, backing_mode,
++                           replaces, sync,
 +                           bitmap, has_bitmap_mode, bitmap_mode, backing_mode,
                             zero_target, has_speed, speed,
                             has_granularity, granularity,
                             has_buf_size, buf_size,
 diff --git a/include/block/block_int-global-state.h b/include/block/block_int-global-state.h
-index b49f4eb35b..9d744db618 100644
+index eb2d92a226..f0c642b194 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,
+@@ -158,7 +158,9 @@ 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,
@@ -359,31 +364,26 @@ index b49f4eb35b..9d744db618 100644
                    BlockdevOnError on_source_error,
                    BlockdevOnError on_target_error,
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 95ac4fa634..7daaf545be 100644
+index b1937780e1..0e5f148d30 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -2000,10 +2000,19 @@
- #        (all the disk, only the sectors allocated in the topmost image, or
- #        only new I/O).
+@@ -2182,6 +2182,15 @@
+ #     destination (all the disk, only the sectors allocated in the
+ #     topmost image, or only new I/O).
  #
-+# @bitmap: The name of a bitmap to use for sync=bitmap mode. This argument must
-+#          be present for bitmap mode and absent otherwise. The bitmap's
-+#          granularity is used instead of @granularity (since 4.1).
++# @bitmap: The name of a bitmap to use for sync=bitmap mode. This
++#     argument must be present for bitmap mode and absent otherwise.
++#     The bitmap's granularity is used instead of @granularity (Since
++#     4.1).
 +#
-+# @bitmap-mode: Specifies the type of data the bitmap should contain after
-+#               the operation concludes. Must be present if sync is "bitmap".
-+#               Must NOT be present otherwise. (Since 4.1)
++# @bitmap-mode: Specifies the type of data the bitmap should contain
++#     after the operation concludes. Must be present if sync is
++#     "bitmap". Must NOT be present otherwise. (Since 4.1)
 +#
- # @granularity: granularity of the dirty bitmap, default is 64K
- #               if the image format doesn't have clusters, 4K if the clusters
- #               are smaller than that, else the cluster size.  Must be a
--#               power of 2 between 512 and 64M (since 1.4).
-+#               power of 2 between 512 and 64M. Must not be specified if
-+#               @bitmap is present (since 1.4).
- #
- # @buf-size: maximum amount of data in flight from source to
- #            target (since 1.4).
-@@ -2043,7 +2052,9 @@
+ # @granularity: granularity of the dirty bitmap, default is 64K if the
+ #     image format doesn't have clusters, 4K if the clusters are
+ #     smaller than that, else the cluster size.  Must be a power of 2
+@@ -2224,7 +2233,9 @@
  { 'struct': 'DriveMirror',
    'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
              '*format': 'str', '*node-name': 'str', '*replaces': 'str',
@@ -394,28 +394,23 @@ index 95ac4fa634..7daaf545be 100644
              '*speed': 'int', '*granularity': 'uint32',
              '*buf-size': 'int', '*on-source-error': 'BlockdevOnError',
              '*on-target-error': 'BlockdevOnError',
-@@ -2322,10 +2333,19 @@
- #        (all the disk, only the sectors allocated in the topmost image, or
- #        only new I/O).
+@@ -2503,6 +2514,15 @@
+ #     destination (all the disk, only the sectors allocated in the
+ #     topmost image, or only new I/O).
  #
-+# @bitmap: The name of a bitmap to use for sync=bitmap mode. This argument must
-+#          be present for bitmap mode and absent otherwise. The bitmap's
-+#          granularity is used instead of @granularity (since 4.1).
++# @bitmap: The name of a bitmap to use for sync=bitmap mode. This
++#     argument must be present for bitmap mode and absent otherwise.
++#     The bitmap's granularity is used instead of @granularity (since
++#     4.1).
 +#
-+# @bitmap-mode: Specifies the type of data the bitmap should contain after
-+#               the operation concludes. Must be present if sync is "bitmap".
-+#               Must NOT be present otherwise. (Since 4.1)
++# @bitmap-mode: Specifies the type of data the bitmap should contain
++#     after the operation concludes. Must be present if sync is
++#     "bitmap". Must NOT be present otherwise. (Since 4.1)
 +#
- # @granularity: granularity of the dirty bitmap, default is 64K
- #               if the image format doesn't have clusters, 4K if the clusters
- #               are smaller than that, else the cluster size.  Must be a
--#               power of 2 between 512 and 64M
-+#               power of 2 between 512 and 64M . Must not be specified if
-+#               @bitmap is present.
- #
- # @buf-size: maximum amount of data in flight from source to
- #            target
-@@ -2375,7 +2395,8 @@
+ # @granularity: granularity of the dirty bitmap, default is 64K if the
+ #     image format doesn't have clusters, 4K if the clusters are
+ #     smaller than that, else the cluster size.  Must be a power of 2
+@@ -2551,7 +2571,8 @@
  { 'command': 'blockdev-mirror',
    'data': { '*job-id': 'str', 'device': 'str', 'target': 'str',
              '*replaces': 'str',
@@ -426,7 +421,7 @@ index 95ac4fa634..7daaf545be 100644
              '*buf-size': 'int', '*on-source-error': 'BlockdevOnError',
              '*on-target-error': 'BlockdevOnError',
 diff --git a/tests/unit/test-block-iothread.c b/tests/unit/test-block-iothread.c
-index 8ca5adec5e..dae80e5a5f 100644
+index 2b358eaaa8..2a149fe021 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)
@@ -439,4 +434,4 @@ index 8ca5adec5e..dae80e5a5f 100644
 +                 false, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
                   false, "filter_node", MIRROR_COPY_MODE_BACKGROUND,
                   &error_abort);
-     WITH_JOB_LOCK_GUARD() {
+ 

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

@@ -24,10 +24,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 18 insertions(+), 6 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index 8ead5f77a0..35c1b8f25d 100644
+index fafca1360e..05e738bcce 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
-@@ -676,8 +676,6 @@ static int mirror_exit_common(Job *job)
+@@ -694,8 +694,6 @@ static int mirror_exit_common(Job *job)
          bdrv_unfreeze_backing_chain(mirror_top_bs, target_bs);
      }
  
@@ -36,9 +36,9 @@ index 8ead5f77a0..35c1b8f25d 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);
-@@ -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);
+@@ -805,6 +803,18 @@ static int mirror_exit_common(Job *job)
+     bdrv_drained_end(target_bs);
+     bdrv_unref(target_bs);
  
 +    if (s->sync_bitmap) {
 +        if (s->bitmap_mode == BITMAP_SYNC_MODE_ALWAYS ||
@@ -55,7 +55,7 @@ index 8ead5f77a0..35c1b8f25d 100644
      bs_opaque->job = NULL;
  
      bdrv_drained_end(src);
-@@ -1668,10 +1678,6 @@ static BlockJob *mirror_start_job(
+@@ -1763,10 +1773,6 @@ static BlockJob *mirror_start_job(
                         " sync mode",
                         MirrorSyncMode_str(sync_mode));
              return NULL;
@@ -66,7 +66,7 @@ index 8ead5f77a0..35c1b8f25d 100644
          }
      } else if (bitmap) {
          error_setg(errp,
-@@ -1688,6 +1694,12 @@ static BlockJob *mirror_start_job(
+@@ -1783,6 +1789,12 @@ 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

@@ -16,10 +16,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 3 insertions(+)
 
 diff --git a/blockdev.c b/blockdev.c
-index 2ee30323cb..dd1c2cdef7 100644
+index ec45bbaa52..9fab7ec554 100644
 --- a/blockdev.c
 +++ b/blockdev.c
-@@ -3045,6 +3045,9 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -2894,6 +2894,9 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
          if (bdrv_dirty_bitmap_check(bitmap, BDRV_BITMAP_ALLOW_RO, errp)) {
              return;
          }
@@ -28,4 +28,4 @@ index 2ee30323cb..dd1c2cdef7 100644
 +        return;
      }
  
-     if (!has_replaces) {
+     if (!replaces) {

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

@@ -16,10 +16,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 4 insertions(+), 7 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index 35c1b8f25d..4969c6833c 100644
+index 05e738bcce..2a2a227f3b 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
-@@ -782,8 +782,8 @@ static int mirror_exit_common(Job *job)
+@@ -809,8 +809,8 @@ 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);
@@ -30,7 +30,7 @@ index 35c1b8f25d..4969c6833c 100644
          }
      }
      bdrv_release_dirty_bitmap(s->dirty_bitmap);
-@@ -1862,11 +1862,8 @@ static BlockJob *mirror_start_job(
+@@ -1971,11 +1971,8 @@ static BlockJob *mirror_start_job(
      }
  
      if (s->sync_mode == MIRROR_SYNC_MODE_BITMAP) {
@@ -43,4 +43,4 @@ index 35c1b8f25d..4969c6833c 100644
 +                                         NULL, true);
      }
  
-     ret = block_job_add_bdrv(&s->common, "source", bs, 0,
+     bdrv_graph_wrlock();

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

@@ -12,6 +12,8 @@ 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>
+[FE: rebase for 8.2.2]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  block/mirror.c             | 28 +++------------
  blockdev.c                 | 29 +++++++++++++++
@@ -19,12 +21,12 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  3 files changed, 70 insertions(+), 59 deletions(-)
 
 diff --git a/block/mirror.c b/block/mirror.c
-index 4969c6833c..cf85ae1074 100644
+index 2a2a227f3b..87c0856979 100644
 --- a/block/mirror.c
 +++ b/block/mirror.c
-@@ -1668,31 +1668,13 @@ static BlockJob *mirror_start_job(
-     uint64_t target_perms, target_shared_perms;
-     int ret;
+@@ -1763,31 +1763,13 @@ static BlockJob *mirror_start_job(
+ 
+     GLOBAL_STATE_CODE();
  
 -    if (sync_mode == MIRROR_SYNC_MODE_INCREMENTAL) {
 -        error_setg(errp, "Sync mode '%s' not supported",
@@ -60,17 +62,17 @@ index 4969c6833c..cf85ae1074 100644
  
          if (bitmap_mode != BITMAP_SYNC_MODE_NEVER) {
 diff --git a/blockdev.c b/blockdev.c
-index dd1c2cdef7..756e980889 100644
+index 9fab7ec554..158ac9314b 100644
 --- a/blockdev.c
 +++ b/blockdev.c
-@@ -3024,7 +3024,36 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
+@@ -2873,7 +2873,36 @@ static void blockdev_mirror_common(const char *job_id, BlockDriverState *bs,
          sync = MIRROR_SYNC_MODE_FULL;
      }
  
 +    if ((sync == MIRROR_SYNC_MODE_BITMAP) ||
 +        (sync == MIRROR_SYNC_MODE_INCREMENTAL)) {
 +        /* done before desugaring 'incremental' to print the right message */
-+        if (!has_bitmap) {
++        if (!bitmap_name) {
 +            error_setg(errp, "Must provide a valid bitmap name for "
 +                       "'%s' sync mode", MirrorSyncMode_str(sync));
 +            return;
@@ -91,7 +93,7 @@ index dd1c2cdef7..756e980889 100644
 +        bitmap_mode = BITMAP_SYNC_MODE_ON_SUCCESS;
 +    }
 +
-     if (has_bitmap) {
+     if (bitmap_name) {
 +        if (sync != MIRROR_SYNC_MODE_BITMAP) {
 +            error_setg(errp, "Sync mode '%s' not supported with bitmap.",
 +                       MirrorSyncMode_str(sync));

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

@@ -48,7 +48,7 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  6 files changed, 59 insertions(+), 5 deletions(-)
 
 diff --git a/include/monitor/monitor.h b/include/monitor/monitor.h
-index 737e750670..38804b8595 100644
+index c3740ec616..7f38ce6b8b 100644
 --- a/include/monitor/monitor.h
 +++ b/include/monitor/monitor.h
 @@ -16,6 +16,7 @@ extern QemuOptsList qemu_mon_opts;
@@ -60,10 +60,10 @@ index 737e750670..38804b8595 100644
  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
+index 5676eb334e..4c452a6aeb 100644
 --- a/monitor/monitor-internal.h
 +++ b/monitor/monitor-internal.h
-@@ -152,6 +152,13 @@ typedef struct {
+@@ -151,6 +151,13 @@ typedef struct {
      QemuMutex qmp_queue_lock;
      /* Input queue that holds all the parsed QMP requests */
      GQueue *qmp_requests;
@@ -78,10 +78,10 @@ index a2cdbbf646..b531bd50e7 100644
  
  /**
 diff --git a/monitor/monitor.c b/monitor/monitor.c
-index 86949024f6..c306cadcf4 100644
+index c5a5d30877..07775784d4 100644
 --- a/monitor/monitor.c
 +++ b/monitor/monitor.c
-@@ -135,6 +135,21 @@ bool monitor_cur_is_qmp(void)
+@@ -116,6 +116,21 @@ bool monitor_cur_is_qmp(void)
      return cur_mon && monitor_is_qmp(cur_mon);
  }
  
@@ -104,10 +104,10 @@ index 86949024f6..c306cadcf4 100644
   * 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 092c527b6f..6b8cfcf6d8 100644
+index 2f46cf9e49..f093e256e9 100644
 --- a/monitor/qmp.c
 +++ b/monitor/qmp.c
-@@ -141,6 +141,8 @@ static void monitor_qmp_dispatch(MonitorQMP *mon, QObject *req)
+@@ -165,6 +165,8 @@ static void monitor_qmp_dispatch(MonitorQMP *mon, QObject *req)
      QDict *rsp;
      QDict *error;
  
@@ -116,7 +116,7 @@ index 092c527b6f..6b8cfcf6d8 100644
      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)
+@@ -180,7 +182,17 @@ static void monitor_qmp_dispatch(MonitorQMP *mon, QObject *req)
          }
      }
  
@@ -135,7 +135,7 @@ index 092c527b6f..6b8cfcf6d8 100644
      qobject_unref(rsp);
  }
  
-@@ -444,6 +456,7 @@ static void monitor_qmp_event(void *opaque, QEMUChrEvent event)
+@@ -461,6 +473,7 @@ static void monitor_qmp_event(void *opaque, QEMUChrEvent event)
  
      switch (event) {
      case CHR_EVENT_OPENED:
@@ -144,7 +144,7 @@ index 092c527b6f..6b8cfcf6d8 100644
          monitor_qmp_caps_reset(mon);
          data = qmp_greeting(mon);
 diff --git a/qapi/qmp-dispatch.c b/qapi/qmp-dispatch.c
-index 0990873ec8..e605003771 100644
+index e569224eae..eb03782e91 100644
 --- a/qapi/qmp-dispatch.c
 +++ b/qapi/qmp-dispatch.c
 @@ -117,16 +117,28 @@ typedef struct QmpDispatchBH {
@@ -180,16 +180,16 @@ index 0990873ec8..e605003771 100644
      aio_co_wake(data->co);
  }
  
-@@ -231,6 +243,7 @@ QDict *qmp_dispatch(const QmpCommandList *cmds, QObject *request,
+@@ -253,6 +265,7 @@ QDict *coroutine_mixed_fn qmp_dispatch(const QmpCommandList *cmds, QObject *requ
              .ret        = &ret,
              .errp       = &err,
              .co         = qemu_coroutine_self(),
 +            .conn_nr    = monitor_get_connection_nr(cur_mon),
          };
-         aio_bh_schedule_oneshot(qemu_get_aio_context(), do_qmp_dispatch_bh,
+         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
+index 1894cdfe1f..d74d0459f0 100644
 --- a/stubs/monitor-core.c
 +++ b/stubs/monitor-core.c
 @@ -12,6 +12,11 @@ Monitor *monitor_set_cur(Coroutine *co, Monitor *mon)
@@ -201,6 +201,6 @@ index afa477aae6..d3ff124bf3 100644
 +    return -1;
 +}
 +
- void monitor_init_qmp(Chardev *chr, bool pretty, Error **errp)
+ void qapi_event_emit(QAPIEvent event, QDict *qdict)
  {
  }

debian/patches/extra/0002-ide-avoid-potential-deadlock-when-draining-during-tr.patch

@@ -55,10 +55,10 @@ Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
  1 file changed, 6 insertions(+), 6 deletions(-)
 
 diff --git a/hw/ide/core.c b/hw/ide/core.c
-index 39afdc0006..b67c1885a8 100644
+index b14983ec54..41c543e627 100644
 --- a/hw/ide/core.c
 +++ b/hw/ide/core.c
-@@ -443,7 +443,7 @@ static void ide_trim_bh_cb(void *opaque)
+@@ -456,7 +456,7 @@ static void ide_trim_bh_cb(void *opaque)
      iocb->bh = NULL;
      qemu_aio_unref(iocb);
  
@@ -67,7 +67,7 @@ index 39afdc0006..b67c1885a8 100644
      blk_dec_in_flight(blk);
  }
  
-@@ -503,6 +503,8 @@ static void ide_issue_trim_cb(void *opaque, int ret)
+@@ -516,6 +516,8 @@ static void ide_issue_trim_cb(void *opaque, int ret)
  done:
      iocb->aiocb = NULL;
      if (iocb->bh) {
@@ -76,8 +76,8 @@ index 39afdc0006..b67c1885a8 100644
          replay_bh_schedule_event(iocb->bh);
      }
  }
-@@ -514,9 +516,6 @@ BlockAIOCB *ide_issue_trim(
-     IDEState *s = opaque;
+@@ -528,9 +530,6 @@ 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() */
@@ -85,8 +85,8 @@ index 39afdc0006..b67c1885a8 100644
 -
      iocb = blk_aio_get(&trim_aiocb_info, s->blk, cb, cb_opaque);
      iocb->s = s;
-     iocb->bh = qemu_bh_new(ide_trim_bh_cb, iocb);
-@@ -739,8 +738,9 @@ void ide_cancel_dma_sync(IDEState *s)
+     iocb->bh = qemu_bh_new_guarded(ide_trim_bh_cb, iocb,
+@@ -754,8 +753,9 @@ void ide_cancel_dma_sync(IDEState *s)
       */
      if (s->bus->dma->aiocb) {
          trace_ide_cancel_dma_sync_remaining();

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

@@ -1,42 +0,0 @@
-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 5115221efe..5f7f6ca981 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -2460,10 +2460,11 @@ 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-virtio-mem-Fix-the-bitmap-index-of-the-section-offse.patch

@@ -1,44 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Chenyi Qiang <chenyi.qiang@intel.com>
-Date: Fri, 16 Dec 2022 14:22:31 +0800
-Subject: [PATCH] virtio-mem: Fix the bitmap index of the section offset
-
-vmem->bitmap indexes the memory region of the virtio-mem backend at a
-granularity of block_size. To calculate the index of target section offset,
-the block_size should be divided instead of the bitmap_size.
-
-Fixes: 2044969f0b ("virtio-mem: Implement RamDiscardManager interface")
-Signed-off-by: Chenyi Qiang <chenyi.qiang@intel.com>
-Message-Id: <20221216062231.11181-1-chenyi.qiang@intel.com>
-Reviewed-by: David Hildenbrand <david@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Cc: qemu-stable@nongnu.org
-Signed-off-by: David Hildenbrand <david@redhat.com>
-(cherry-picked from commit b11cf32e07a2f7ff0d171b89497381a04c9d07e0)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/virtio/virtio-mem.c | 4 ++--
- 1 file changed, 2 insertions(+), 2 deletions(-)
-
-diff --git a/hw/virtio/virtio-mem.c b/hw/virtio/virtio-mem.c
-index ed170def48..e19ee817fe 100644
---- a/hw/virtio/virtio-mem.c
-+++ b/hw/virtio/virtio-mem.c
-@@ -235,7 +235,7 @@ static int virtio_mem_for_each_plugged_section(const VirtIOMEM *vmem,
-     uint64_t offset, size;
-     int ret = 0;
- 
--    first_bit = s->offset_within_region / vmem->bitmap_size;
-+    first_bit = s->offset_within_region / vmem->block_size;
-     first_bit = find_next_bit(vmem->bitmap, vmem->bitmap_size, first_bit);
-     while (first_bit < vmem->bitmap_size) {
-         MemoryRegionSection tmp = *s;
-@@ -267,7 +267,7 @@ static int virtio_mem_for_each_unplugged_section(const VirtIOMEM *vmem,
-     uint64_t offset, size;
-     int ret = 0;
- 
--    first_bit = s->offset_within_region / vmem->bitmap_size;
-+    first_bit = s->offset_within_region / vmem->block_size;
-     first_bit = find_next_zero_bit(vmem->bitmap, vmem->bitmap_size, first_bit);
-     while (first_bit < vmem->bitmap_size) {
-         MemoryRegionSection tmp = *s;

debian/patches/extra/0004-virtio-mem-Fix-the-iterator-variable-in-a-vmem-rdl_l.patch

@@ -1,36 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Chenyi Qiang <chenyi.qiang@intel.com>
-Date: Wed, 28 Dec 2022 17:03:12 +0800
-Subject: [PATCH] virtio-mem: Fix the iterator variable in a vmem->rdl_list
- loop
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-It should be the variable rdl2 to revert the already-notified listeners.
-
-Fixes: 2044969f0b ("virtio-mem: Implement RamDiscardManager interface")
-Signed-off-by: Chenyi Qiang <chenyi.qiang@intel.com>
-Message-Id: <20221228090312.17276-1-chenyi.qiang@intel.com>
-Cc: qemu-stable@nongnu.org
-Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
-Signed-off-by: David Hildenbrand <david@redhat.com>
-(cherry-picked from commit 29f1b328e3b767cba2661920a8470738469b9e36)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/virtio/virtio-mem.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/hw/virtio/virtio-mem.c b/hw/virtio/virtio-mem.c
-index e19ee817fe..56db586c89 100644
---- a/hw/virtio/virtio-mem.c
-+++ b/hw/virtio/virtio-mem.c
-@@ -341,7 +341,7 @@ static int virtio_mem_notify_plug(VirtIOMEM *vmem, uint64_t offset,
-     if (ret) {
-         /* Notify all already-notified listeners. */
-         QLIST_FOREACH(rdl2, &vmem->rdl_list, next) {
--            MemoryRegionSection tmp = *rdl->section;
-+            MemoryRegionSection tmp = *rdl2->section;
- 
-             if (rdl2 == rdl) {
-                 break;

debian/patches/extra/0005-vhost-fix-vq-dirty-bitmap-syncing-when-vIOMMU-is-ena.patch

@@ -1,141 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Jason Wang <jasowang@redhat.com>
-Date: Fri, 16 Dec 2022 11:35:52 +0800
-Subject: [PATCH] vhost: fix vq dirty bitmap syncing when vIOMMU is enabled
-
-When vIOMMU is enabled, the vq->used_phys is actually the IOVA not
-GPA. So we need to translate it to GPA before the syncing otherwise we
-may hit the following crash since IOVA could be out of the scope of
-the GPA log size. This could be noted when using virtio-IOMMU with
-vhost using 1G memory.
-
-Fixes: c471ad0e9bd46 ("vhost_net: device IOTLB support")
-Cc: qemu-stable@nongnu.org
-Tested-by: Lei Yang <leiyang@redhat.com>
-Reported-by: Yalan Zhang <yalzhang@redhat.com>
-Signed-off-by: Jason Wang <jasowang@redhat.com>
-Message-Id: <20221216033552.77087-1-jasowang@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 345cc1cbcbce2bab00abc2b88338d7d89c702d6b)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost.c | 84 ++++++++++++++++++++++++++++++++++++-----------
- 1 file changed, 64 insertions(+), 20 deletions(-)
-
-diff --git a/hw/virtio/vhost.c b/hw/virtio/vhost.c
-index 7fb008bc9e..fdcd1a8fdf 100644
---- a/hw/virtio/vhost.c
-+++ b/hw/virtio/vhost.c
-@@ -20,6 +20,7 @@
- #include "qemu/range.h"
- #include "qemu/error-report.h"
- #include "qemu/memfd.h"
-+#include "qemu/log.h"
- #include "standard-headers/linux/vhost_types.h"
- #include "hw/virtio/virtio-bus.h"
- #include "hw/virtio/virtio-access.h"
-@@ -106,6 +107,24 @@ static void vhost_dev_sync_region(struct vhost_dev *dev,
-     }
- }
- 
-+static bool vhost_dev_has_iommu(struct vhost_dev *dev)
-+{
-+    VirtIODevice *vdev = dev->vdev;
-+
-+    /*
-+     * For vhost, VIRTIO_F_IOMMU_PLATFORM means the backend support
-+     * incremental memory mapping API via IOTLB API. For platform that
-+     * does not have IOMMU, there's no need to enable this feature
-+     * which may cause unnecessary IOTLB miss/update transactions.
-+     */
-+    if (vdev) {
-+        return virtio_bus_device_iommu_enabled(vdev) &&
-+            virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
-+    } else {
-+        return false;
-+    }
-+}
-+
- static int vhost_sync_dirty_bitmap(struct vhost_dev *dev,
-                                    MemoryRegionSection *section,
-                                    hwaddr first,
-@@ -137,8 +156,51 @@ static int vhost_sync_dirty_bitmap(struct vhost_dev *dev,
-             continue;
-         }
- 
--        vhost_dev_sync_region(dev, section, start_addr, end_addr, vq->used_phys,
--                              range_get_last(vq->used_phys, vq->used_size));
-+        if (vhost_dev_has_iommu(dev)) {
-+            IOMMUTLBEntry iotlb;
-+            hwaddr used_phys = vq->used_phys, used_size = vq->used_size;
-+            hwaddr phys, s, offset;
-+
-+            while (used_size) {
-+                rcu_read_lock();
-+                iotlb = address_space_get_iotlb_entry(dev->vdev->dma_as,
-+                                                      used_phys,
-+                                                      true,
-+                                                      MEMTXATTRS_UNSPECIFIED);
-+                rcu_read_unlock();
-+
-+                if (!iotlb.target_as) {
-+                    qemu_log_mask(LOG_GUEST_ERROR, "translation "
-+                                  "failure for used_iova %"PRIx64"\n",
-+                                  used_phys);
-+                    return -EINVAL;
-+                }
-+
-+                offset = used_phys & iotlb.addr_mask;
-+                phys = iotlb.translated_addr + offset;
-+
-+                /*
-+                 * Distance from start of used ring until last byte of
-+                 * IOMMU page.
-+                 */
-+                s = iotlb.addr_mask - offset;
-+                /*
-+                 * Size of used ring, or of the part of it until end
-+                 * of IOMMU page. To avoid zero result, do the adding
-+                 * outside of MIN().
-+                 */
-+                s = MIN(s, used_size - 1) + 1;
-+
-+                vhost_dev_sync_region(dev, section, start_addr, end_addr, phys,
-+                                      range_get_last(phys, s));
-+                used_size -= s;
-+                used_phys += s;
-+            }
-+        } else {
-+            vhost_dev_sync_region(dev, section, start_addr,
-+                                  end_addr, vq->used_phys,
-+                                  range_get_last(vq->used_phys, vq->used_size));
-+        }
-     }
-     return 0;
- }
-@@ -306,24 +368,6 @@ static inline void vhost_dev_log_resize(struct vhost_dev *dev, uint64_t size)
-     dev->log_size = size;
- }
- 
--static bool vhost_dev_has_iommu(struct vhost_dev *dev)
--{
--    VirtIODevice *vdev = dev->vdev;
--
--    /*
--     * For vhost, VIRTIO_F_IOMMU_PLATFORM means the backend support
--     * incremental memory mapping API via IOTLB API. For platform that
--     * does not have IOMMU, there's no need to enable this feature
--     * which may cause unnecessary IOTLB miss/update transactions.
--     */
--    if (vdev) {
--        return virtio_bus_device_iommu_enabled(vdev) &&
--            virtio_host_has_feature(vdev, VIRTIO_F_IOMMU_PLATFORM);
--    } else {
--        return false;
--    }
--}
--
- static void *vhost_memory_map(struct vhost_dev *dev, hwaddr addr,
-                               hwaddr *plen, bool is_write)
- {

debian/patches/extra/0006-virtio-rng-pci-fix-migration-compat-for-vectors.patch

@@ -1,42 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
-Date: Mon, 9 Jan 2023 10:58:09 +0000
-Subject: [PATCH] virtio-rng-pci: fix migration compat for vectors
-
-Fixup the migration compatibility for existing machine types
-so that they do not enable msi-x.
-
-Symptom:
-
-(qemu) qemu: get_pci_config_device: Bad config data: i=0x34 read: 84 device: 98 cmask: ff wmask: 0 w1cmask:0
-qemu: Failed to load PCIDevice:config
-qemu: Failed to load virtio-rng:virtio
-qemu: error while loading state for instance 0x0 of device '0000:00:03.0/virtio-rng'
-qemu: load of migration failed: Invalid argument
-
-Note: This fix will break migration from 7.2->7.2-fixed with this patch
-
-bz: https://bugzilla.redhat.com/show_bug.cgi?id=2155749
-Fixes: 9ea02e8f1 ("virtio-rng-pci: Allow setting nvectors, so we can use MSI-X")
-
-Reviewed-by: Thomas Huth <thuth@redhat.com>
-Acked-by: David Daney <david.daney@fungible.com>
-Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
-(picked-up from https://lists.nongnu.org/archive/html/qemu-devel/2023-01/msg01319.html)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/core/machine.c | 1 +
- 1 file changed, 1 insertion(+)
-
-diff --git a/hw/core/machine.c b/hw/core/machine.c
-index 8d34caa31d..77a0a131d1 100644
---- a/hw/core/machine.c
-+++ b/hw/core/machine.c
-@@ -42,6 +42,7 @@
- 
- GlobalProperty hw_compat_7_1[] = {
-     { "virtio-device", "queue_reset", "false" },
-+    { "virtio-rng-pci", "vectors", "0" },
- };
- const size_t hw_compat_7_1_len = G_N_ELEMENTS(hw_compat_7_1);
- 

debian/patches/extra/0007-block-fix-detect-zeroes-with-BDRV_REQ_REGISTERED_BUF.patch

@@ -1,36 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Hajnoczi <stefanha@redhat.com>
-Date: Thu, 26 Jan 2023 15:13:58 -0500
-Subject: [PATCH] block: fix detect-zeroes= with BDRV_REQ_REGISTERED_BUF
-
-When a write request is converted into a write zeroes request by the
-detect-zeroes= feature, it is no longer associated with an I/O buffer.
-The BDRV_REQ_REGISTERED_BUF flag doesn't make sense without an I/O
-buffer and must be cleared because bdrv_co_do_pwrite_zeroes() fails with
--EINVAL when it's set.
-
-Fiona Ebner <f.ebner@proxmox.com> bisected and diagnosed this QEMU 7.2
-regression where writes containing zeroes to a blockdev with
-discard=unmap,detect-zeroes=unmap fail.
-
-Buglink: https://gitlab.com/qemu-project/qemu/-/issues/1404
-Fixes: e8b6535533be ("block: add BDRV_REQ_REGISTERED_BUF request flag")
-Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
----
- block/io.c | 3 +++
- 1 file changed, 3 insertions(+)
-
-diff --git a/block/io.c b/block/io.c
-index b9424024f9..bbaa0d1b2d 100644
---- a/block/io.c
-+++ b/block/io.c
-@@ -2087,6 +2087,9 @@ static int coroutine_fn bdrv_aligned_pwritev(BdrvChild *child,
-         if (bs->detect_zeroes == BLOCKDEV_DETECT_ZEROES_OPTIONS_UNMAP) {
-             flags |= BDRV_REQ_MAY_UNMAP;
-         }
-+
-+        /* Can't use optimization hint with bufferless zero write */
-+        flags &= ~BDRV_REQ_REGISTERED_BUF;
-     }
- 
-     if (ret < 0) {

debian/patches/extra/0008-memory-prevent-dma-reentracy-issues.patch

@@ -1,118 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Alexander Bulekov <alxndr@bu.edu>
-Date: Sat, 4 Feb 2023 23:07:34 -0500
-Subject: [PATCH] memory: prevent dma-reentracy issues
-
-Add a flag to the DeviceState, when a device is engaged in PIO/MMIO/DMA.
-This flag is set/checked prior to calling a device's MemoryRegion
-handlers, and set when device code initiates DMA.  The purpose of this
-flag is to prevent two types of DMA-based reentrancy issues:
-
-1.) mmio -> dma -> mmio case
-2.) bh -> dma write -> mmio case
-
-These issues have led to problems such as stack-exhaustion and
-use-after-frees.
-
-Summary of the problem from Peter Maydell:
-https://lore.kernel.org/qemu-devel/CAFEAcA_23vc7hE3iaM-JVA6W38LK4hJoWae5KcknhPRD5fPBZA@mail.gmail.com
-
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/62
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/540
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/541
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/556
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/557
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/827
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1282
-
-Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
-Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
-Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
-Acked-by: Peter Xu <peterx@redhat.com>
-(picked-up from https://lists.nongnu.org/archive/html/qemu-devel/2023-02/msg01142.html)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- include/hw/qdev-core.h |  7 +++++++
- softmmu/memory.c       | 17 +++++++++++++++++
- softmmu/trace-events   |  1 +
- 3 files changed, 25 insertions(+)
-
-diff --git a/include/hw/qdev-core.h b/include/hw/qdev-core.h
-index 785dd5a56e..886f6bb79e 100644
---- a/include/hw/qdev-core.h
-+++ b/include/hw/qdev-core.h
-@@ -162,6 +162,10 @@ struct NamedClockList {
-     QLIST_ENTRY(NamedClockList) node;
- };
- 
-+typedef struct {
-+    bool engaged_in_io;
-+} MemReentrancyGuard;
-+
- /**
-  * DeviceState:
-  * @realized: Indicates whether the device has been fully constructed.
-@@ -194,6 +198,9 @@ struct DeviceState {
-     int alias_required_for_version;
-     ResettableState reset;
-     GSList *unplug_blockers;
-+
-+    /* Is the device currently in mmio/pio/dma? Used to prevent re-entrancy */
-+    MemReentrancyGuard mem_reentrancy_guard;
- };
- 
- struct DeviceListener {
-diff --git a/softmmu/memory.c b/softmmu/memory.c
-index bc0be3f62c..7dcb3347aa 100644
---- a/softmmu/memory.c
-+++ b/softmmu/memory.c
-@@ -533,6 +533,7 @@ static MemTxResult access_with_adjusted_size(hwaddr addr,
-     uint64_t access_mask;
-     unsigned access_size;
-     unsigned i;
-+    DeviceState *dev = NULL;
-     MemTxResult r = MEMTX_OK;
- 
-     if (!access_size_min) {
-@@ -542,6 +543,19 @@ static MemTxResult access_with_adjusted_size(hwaddr addr,
-         access_size_max = 4;
-     }
- 
-+    /* Do not allow more than one simultanous access to a device's IO Regions */
-+    if (mr->owner &&
-+        !mr->ram_device && !mr->ram && !mr->rom_device && !mr->readonly) {
-+        dev = (DeviceState *) object_dynamic_cast(mr->owner, TYPE_DEVICE);
-+        if (dev) {
-+            if (dev->mem_reentrancy_guard.engaged_in_io) {
-+                trace_memory_region_reentrant_io(get_cpu_index(), mr, addr, size);
-+                return MEMTX_ERROR;
-+            }
-+            dev->mem_reentrancy_guard.engaged_in_io = true;
-+        }
-+    }
-+
-     /* FIXME: support unaligned access? */
-     access_size = MAX(MIN(size, access_size_max), access_size_min);
-     access_mask = MAKE_64BIT_MASK(0, access_size * 8);
-@@ -556,6 +570,9 @@ static MemTxResult access_with_adjusted_size(hwaddr addr,
-                         access_mask, attrs);
-         }
-     }
-+    if (dev) {
-+        dev->mem_reentrancy_guard.engaged_in_io = false;
-+    }
-     return r;
- }
- 
-diff --git a/softmmu/trace-events b/softmmu/trace-events
-index 22606dc27b..62d04ea9a7 100644
---- a/softmmu/trace-events
-+++ b/softmmu/trace-events
-@@ -13,6 +13,7 @@ memory_region_ops_read(int cpu_index, void *mr, uint64_t addr, uint64_t value, u
- memory_region_ops_write(int cpu_index, void *mr, uint64_t addr, uint64_t value, unsigned size, const char *name) "cpu %d mr %p addr 0x%"PRIx64" value 0x%"PRIx64" size %u name '%s'"
- memory_region_subpage_read(int cpu_index, void *mr, uint64_t offset, uint64_t value, unsigned size) "cpu %d mr %p offset 0x%"PRIx64" value 0x%"PRIx64" size %u"
- memory_region_subpage_write(int cpu_index, void *mr, uint64_t offset, uint64_t value, unsigned size) "cpu %d mr %p offset 0x%"PRIx64" value 0x%"PRIx64" size %u"
-+memory_region_reentrant_io(int cpu_index, void *mr, uint64_t offset, unsigned size) "cpu %d mr %p offset 0x%"PRIx64" size %u"
- memory_region_ram_device_read(int cpu_index, void *mr, uint64_t addr, uint64_t value, unsigned size) "cpu %d mr %p addr 0x%"PRIx64" value 0x%"PRIx64" size %u"
- memory_region_ram_device_write(int cpu_index, void *mr, uint64_t addr, uint64_t value, unsigned size) "cpu %d mr %p addr 0x%"PRIx64" value 0x%"PRIx64" size %u"
- memory_region_sync_dirty(const char *mr, const char *listener, int global) "mr '%s' listener '%s' synced (global=%d)"

debian/patches/extra/0009-block-iscsi-fix-double-free-on-BUSY-or-similar-statu.patch

@@ -1,32 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Paolo Bonzini <pbonzini@redhat.com>
-Date: Tue, 10 Jan 2023 17:36:33 +0100
-Subject: [PATCH] block/iscsi: fix double-free on BUSY or similar statuses
-
-Commit 8c460269aa77 ("iscsi: base all handling of check condition on
-scsi_sense_to_errno", 2019-07-15) removed a "goto out" so that the
-same coroutine is re-entered twice; once from iscsi_co_generic_cb,
-once from the timer callback iscsi_retry_timer_expired.  This can
-cause a crash.
-
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1378
-Reported-by: Grzegorz Zdanowski <https://gitlab.com/kiler129>
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit 5080152e2ef6cde7aa692e29880c62bd54acb750)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- block/iscsi.c | 1 +
- 1 file changed, 1 insertion(+)
-
-diff --git a/block/iscsi.c b/block/iscsi.c
-index 3ed4a50c0d..89cd032c3a 100644
---- a/block/iscsi.c
-+++ b/block/iscsi.c
-@@ -268,6 +268,7 @@ iscsi_co_generic_cb(struct iscsi_context *iscsi, int status,
-                 timer_mod(&iTask->retry_timer,
-                           qemu_clock_get_ms(QEMU_CLOCK_REALTIME) + retry_time);
-                 iTask->do_retry = 1;
-+                return;
-             } else if (status == SCSI_STATUS_CHECK_CONDITION) {
-                 int error = iscsi_translate_sense(&task->sense);
-                 if (error == EAGAIN) {

debian/patches/extra/0010-scsi-megasas-Internal-cdbs-have-16-byte-length.patch

@@ -1,69 +0,0 @@
-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/0012-hw-nvme-fix-missing-endian-conversions-for-doorbell-.patch

@@ -1,67 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Klaus Jensen <k.jensen@samsung.com>
-Date: Wed, 8 Mar 2023 19:57:12 +0300
-Subject: [PATCH] hw/nvme: fix missing endian conversions for doorbell buffers
-
-The eventidx and doorbell value are not handling endianness correctly.
-Fix this.
-
-Fixes: 3f7fe8de3d49 ("hw/nvme: Implement shadow doorbell buffer support")
-Cc: qemu-stable@nongnu.org
-Reported-by: Guenter Roeck <linux@roeck-us.net>
-Reviewed-by: Keith Busch <kbusch@kernel.org>
-Signed-off-by: Klaus Jensen <k.jensen@samsung.com>
-(cherry picked from commit 2fda0726e5149e032acfa5fe442db56cd6433c4c)
-Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
-Conflicts: hw/nvme/ctrl.c
-(picked up from qemu-stable mailing list)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/nvme/ctrl.c | 22 ++++++++++++++++------
- 1 file changed, 16 insertions(+), 6 deletions(-)
-
-diff --git a/hw/nvme/ctrl.c b/hw/nvme/ctrl.c
-index e54276dc1d..98d8e34109 100644
---- a/hw/nvme/ctrl.c
-+++ b/hw/nvme/ctrl.c
-@@ -1333,8 +1333,12 @@ static inline void nvme_blk_write(BlockBackend *blk, int64_t offset,
- 
- static void nvme_update_cq_head(NvmeCQueue *cq)
- {
--    pci_dma_read(&cq->ctrl->parent_obj, cq->db_addr, &cq->head,
--            sizeof(cq->head));
-+    uint32_t v;
-+
-+    pci_dma_read(&cq->ctrl->parent_obj, cq->db_addr, &v, sizeof(v));
-+
-+    cq->head = le32_to_cpu(v);
-+
-     trace_pci_nvme_shadow_doorbell_cq(cq->cqid, cq->head);
- }
- 
-@@ -6141,15 +6145,21 @@ static uint16_t nvme_admin_cmd(NvmeCtrl *n, NvmeRequest *req)
- 
- static void nvme_update_sq_eventidx(const NvmeSQueue *sq)
- {
--    pci_dma_write(&sq->ctrl->parent_obj, sq->ei_addr, &sq->tail,
--                  sizeof(sq->tail));
-+    uint32_t v = cpu_to_le32(sq->tail);
-+
-+    pci_dma_write(&sq->ctrl->parent_obj, sq->ei_addr, &v, sizeof(v));
-+
-     trace_pci_nvme_eventidx_sq(sq->sqid, sq->tail);
- }
- 
- static void nvme_update_sq_tail(NvmeSQueue *sq)
- {
--    pci_dma_read(&sq->ctrl->parent_obj, sq->db_addr, &sq->tail,
--                 sizeof(sq->tail));
-+    uint32_t v;
-+
-+    pci_dma_read(&sq->ctrl->parent_obj, sq->db_addr, &v, sizeof(v));
-+
-+    sq->tail = le32_to_cpu(v);
-+
-     trace_pci_nvme_shadow_doorbell_sq(sq->sqid, sq->tail);
- }
- 

debian/patches/extra/0013-hw-smbios-fix-field-corruption-in-type-4-table.patch

@@ -1,50 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Julia Suvorova <jusual@redhat.com>
-Date: Thu, 23 Feb 2023 13:57:47 +0100
-Subject: [PATCH] hw/smbios: fix field corruption in type 4 table
-
-Since table type 4 of SMBIOS version 2.6 is shorter than 3.0, the
-strings which follow immediately after the struct fields have been
-overwritten by unconditional filling of later fields such as core_count2.
-Make these fields dependent on the SMBIOS version.
-
-Fixes: 05e27d74c7 ("hw/smbios: add core_count2 to smbios table type 4")
-Resolves: https://bugzilla.redhat.com/show_bug.cgi?id=2169904
-
-Signed-off-by: Julia Suvorova <jusual@redhat.com>
-Message-Id: <20230223125747.254914-1-jusual@redhat.com>
-Reviewed-by: Igor Mammedov <imammedo@redhat.com>
-Reviewed-by: Ani Sinha <ani@anisinha.ca>
-Reviewed-by: Igor Mammedov <imammedo@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 60d09b8dc7dd4256d664ad680795cb1327805b2b)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/smbios/smbios.c | 8 +++++---
- 1 file changed, 5 insertions(+), 3 deletions(-)
-
-diff --git a/hw/smbios/smbios.c b/hw/smbios/smbios.c
-index b4243de735..66a020999b 100644
---- a/hw/smbios/smbios.c
-+++ b/hw/smbios/smbios.c
-@@ -749,14 +749,16 @@ static void smbios_build_type_4_table(MachineState *ms, unsigned instance)
-     t->core_count = (ms->smp.cores > 255) ? 0xFF : ms->smp.cores;
-     t->core_enabled = t->core_count;
- 
--    t->core_count2 = t->core_enabled2 = cpu_to_le16(ms->smp.cores);
--
-     t->thread_count = (ms->smp.threads > 255) ? 0xFF : ms->smp.threads;
--    t->thread_count2 = cpu_to_le16(ms->smp.threads);
- 
-     t->processor_characteristics = cpu_to_le16(0x02); /* Unknown */
-     t->processor_family2 = cpu_to_le16(0x01); /* Other */
- 
-+    if (tbl_len == SMBIOS_TYPE_4_LEN_V30) {
-+        t->core_count2 = t->core_enabled2 = cpu_to_le16(ms->smp.cores);
-+        t->thread_count2 = cpu_to_le16(ms->smp.threads);
-+    }
-+
-     SMBIOS_BUILD_TABLE_POST;
-     smbios_type4_count++;
- }

debian/patches/extra/0014-virtio-rng-pci-fix-transitional-migration-compat-for.patch

@@ -1,35 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: "Dr. David Alan Gilbert" <dgilbert@redhat.com>
-Date: Tue, 7 Feb 2023 17:49:44 +0000
-Subject: [PATCH] virtio-rng-pci: fix transitional migration compat for vectors
-
-In bad9c5a516 ("virtio-rng-pci: fix migration compat for vectors") I
-fixed the virtio-rng-pci migration compatibility, but it was discovered
-that we also need to fix the other aliases of the device for the
-transitional cases.
-
-Fixes: 9ea02e8f1 ('virtio-rng-pci: Allow setting nvectors, so we can use MSI-X')
-bz: https://bugzilla.redhat.com/show_bug.cgi?id=2162569
-Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
-Message-Id: <20230207174944.138255-1-dgilbert@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 62bdb8871512076841f4464f7e26efdc7783f78d)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/core/machine.c | 2 ++
- 1 file changed, 2 insertions(+)
-
-diff --git a/hw/core/machine.c b/hw/core/machine.c
-index cd84579591..4297315984 100644
---- a/hw/core/machine.c
-+++ b/hw/core/machine.c
-@@ -43,6 +43,8 @@
- GlobalProperty hw_compat_7_1[] = {
-     { "virtio-device", "queue_reset", "false" },
-     { "virtio-rng-pci", "vectors", "0" },
-+    { "virtio-rng-pci-transitional", "vectors", "0" },
-+    { "virtio-rng-pci-non-transitional", "vectors", "0" },
- };
- const size_t hw_compat_7_1_len = G_N_ELEMENTS(hw_compat_7_1);
- 

debian/patches/extra/0015-hw-timer-hpet-Fix-expiration-time-overflow.patch

@@ -1,80 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Akihiko Odaki <akihiko.odaki@daynix.com>
-Date: Tue, 31 Jan 2023 12:00:37 +0900
-Subject: [PATCH] hw/timer/hpet: Fix expiration time overflow
-
-The expiration time provided for timer_mod() can overflow if a
-ridiculously large value is set to the comparator register. The
-resulting value can represent a past time after rounded, forcing the
-timer to fire immediately. If the timer is configured as periodic, it
-will rearm the timer again, and form an endless loop.
-
-Check if the expiration value will overflow, and if it will, stop the
-timer instead of rearming the timer with the overflowed time.
-
-This bug was found by Alexander Bulekov when fuzzing igb, a new
-network device emulation:
-https://patchew.org/QEMU/20230129053316.1071513-1-alxndr@bu.edu/
-
-The fixed test case is:
-fuzz/crash_2d7036941dcda1ad4380bb8a9174ed0c949bcefd
-
-Fixes: 16b29ae180 ("Add HPET emulation to qemu (Beth Kon)")
-Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
-Acked-by: Michael S. Tsirkin <mst@redhat.com>
-Message-Id: <20230131030037.18856-1-akihiko.odaki@daynix.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 37d2bcbc2a4e9c2e9061bec72a32c7e49b9f81ec)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/timer/hpet.c | 19 +++++++++++++------
- 1 file changed, 13 insertions(+), 6 deletions(-)
-
-diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
-index 9520471be2..5f88ffdef8 100644
---- a/hw/timer/hpet.c
-+++ b/hw/timer/hpet.c
-@@ -352,6 +352,16 @@ static const VMStateDescription vmstate_hpet = {
-     }
- };
- 
-+static void hpet_arm(HPETTimer *t, uint64_t ticks)
-+{
-+    if (ticks < ns_to_ticks(INT64_MAX / 2)) {
-+        timer_mod(t->qemu_timer,
-+                  qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + ticks_to_ns(ticks));
-+    } else {
-+        timer_del(t->qemu_timer);
-+    }
-+}
-+
- /*
-  * timer expiration callback
-  */
-@@ -374,13 +384,11 @@ static void hpet_timer(void *opaque)
-             }
-         }
-         diff = hpet_calculate_diff(t, cur_tick);
--        timer_mod(t->qemu_timer,
--                       qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + (int64_t)ticks_to_ns(diff));
-+        hpet_arm(t, diff);
-     } else if (t->config & HPET_TN_32BIT && !timer_is_periodic(t)) {
-         if (t->wrap_flag) {
-             diff = hpet_calculate_diff(t, cur_tick);
--            timer_mod(t->qemu_timer, qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) +
--                           (int64_t)ticks_to_ns(diff));
-+            hpet_arm(t, diff);
-             t->wrap_flag = 0;
-         }
-     }
-@@ -407,8 +415,7 @@ static void hpet_set_timer(HPETTimer *t)
-             t->wrap_flag = 1;
-         }
-     }
--    timer_mod(t->qemu_timer,
--                   qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + (int64_t)ticks_to_ns(diff));
-+    hpet_arm(t, diff);
- }
- 
- static void hpet_del_timer(HPETTimer *t)

debian/patches/extra/0016-vdpa-stop-all-svq-on-device-deletion.patch

@@ -1,71 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Eugenio=20P=C3=A9rez?= <eperezma@redhat.com>
-Date: Thu, 9 Feb 2023 18:00:04 +0100
-Subject: [PATCH] vdpa: stop all svq on device deletion
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Not stopping them leave the device in a bad state when virtio-net
-fronted device is unplugged with device_del monitor command.
-
-This is not triggable in regular poweroff or qemu forces shutdown
-because cleanup is called right after vhost_vdpa_dev_start(false).  But
-devices hot unplug does not call vdpa device cleanups.  This lead to all
-the vhost_vdpa devices without stop the SVQ but the last.
-
-Fix it and clean the code, making it symmetric with
-vhost_vdpa_svqs_start.
-
-Fixes: dff4426fa656 ("vhost: Add Shadow VirtQueue kick forwarding capabilities")
-Reported-by: Lei Yang <leiyang@redhat.com>
-Signed-off-by: Eugenio Pérez <eperezma@redhat.com>
-Message-Id: <20230209170004.899472-1-eperezma@redhat.com>
-Tested-by: Laurent Vivier <lvivier@redhat.com>
-Acked-by: Jason Wang <jasowang@redhat.com>
-(cherry-picked from commit 2e1a9de96b487cf818a22d681cad8d3f5d18dcca)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost-vdpa.c | 17 ++---------------
- 1 file changed, 2 insertions(+), 15 deletions(-)
-
-diff --git a/hw/virtio/vhost-vdpa.c b/hw/virtio/vhost-vdpa.c
-index 7468e44b87..03c78d25d8 100644
---- a/hw/virtio/vhost-vdpa.c
-+++ b/hw/virtio/vhost-vdpa.c
-@@ -707,26 +707,11 @@ static int vhost_vdpa_get_device_id(struct vhost_dev *dev,
-     return ret;
- }
- 
--static void vhost_vdpa_reset_svq(struct vhost_vdpa *v)
--{
--    if (!v->shadow_vqs_enabled) {
--        return;
--    }
--
--    for (unsigned i = 0; i < v->shadow_vqs->len; ++i) {
--        VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, i);
--        vhost_svq_stop(svq);
--    }
--}
--
- static int vhost_vdpa_reset_device(struct vhost_dev *dev)
- {
--    struct vhost_vdpa *v = dev->opaque;
-     int ret;
-     uint8_t status = 0;
- 
--    vhost_vdpa_reset_svq(v);
--
-     ret = vhost_vdpa_call(dev, VHOST_VDPA_SET_STATUS, &status);
-     trace_vhost_vdpa_reset_device(dev, status);
-     return ret;
-@@ -1088,6 +1073,8 @@ static void vhost_vdpa_svqs_stop(struct vhost_dev *dev)
- 
-     for (unsigned i = 0; i < v->shadow_vqs->len; ++i) {
-         VhostShadowVirtqueue *svq = g_ptr_array_index(v->shadow_vqs, i);
-+
-+        vhost_svq_stop(svq);
-         vhost_vdpa_svq_unmap_rings(dev, svq);
-     }
- }

debian/patches/extra/0017-vhost-avoid-a-potential-use-of-an-uninitialized-vari.patch

@@ -1,132 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: =?UTF-8?q?Carlos=20L=C3=B3pez?= <clopez@suse.de>
-Date: Mon, 13 Feb 2023 09:57:47 +0100
-Subject: [PATCH] vhost: avoid a potential use of an uninitialized variable in
- vhost_svq_poll()
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-In vhost_svq_poll(), if vhost_svq_get_buf() fails due to a device
-providing invalid descriptors, len is left uninitialized and returned
-to the caller, potentally leaking stack data or causing undefined
-behavior.
-
-Fix this by initializing len to 0.
-
-Found with GCC 13 and -fanalyzer (abridged):
-
-../hw/virtio/vhost-shadow-virtqueue.c: In function ‘vhost_svq_poll’:
-../hw/virtio/vhost-shadow-virtqueue.c:538:12: warning: use of uninitialized value ‘len’ [CWE-457] [-Wanalyzer-use-of-uninitialized-value]
-  538 |     return len;
-      |            ^~~
-  ‘vhost_svq_poll’: events 1-4
-    |
-    |  522 | size_t vhost_svq_poll(VhostShadowVirtqueue *svq)
-    |      |        ^~~~~~~~~~~~~~
-    |      |        |
-    |      |        (1) entry to ‘vhost_svq_poll’
-    |......
-    |  525 |     uint32_t len;
-    |      |              ~~~
-    |      |              |
-    |      |              (2) region created on stack here
-    |      |              (3) capacity: 4 bytes
-    |......
-    |  528 |         if (vhost_svq_more_used(svq)) {
-    |      |             ~
-    |      |             |
-    |      |             (4) inlined call to ‘vhost_svq_more_used’ from ‘vhost_svq_poll’
-
-    (...)
-
-    |  528 |         if (vhost_svq_more_used(svq)) {
-    |      |            ^~~~~~~~~~~~~~~~~~~~~~~~~
-    |      |            ||
-    |      |            |(8) ...to here
-    |      |            (7) following ‘true’ branch...
-    |......
-    |  537 |     vhost_svq_get_buf(svq, &len);
-    |      |     ~~~~~~~~~~~~~~~~~~~~~~~~~~~~
-    |      |     |
-    |      |     (9) calling ‘vhost_svq_get_buf’ from ‘vhost_svq_poll’
-    |
-    +--> ‘vhost_svq_get_buf’: events 10-11
-           |
-           |  416 | static VirtQueueElement *vhost_svq_get_buf(VhostShadowVirtqueue *svq,
-           |      |                          ^~~~~~~~~~~~~~~~~
-           |      |                          |
-           |      |                          (10) entry to ‘vhost_svq_get_buf’
-           |......
-           |  423 |     if (!vhost_svq_more_used(svq)) {
-           |      |          ~
-           |      |          |
-           |      |          (11) inlined call to ‘vhost_svq_more_used’ from ‘vhost_svq_get_buf’
-           |
-
-           (...)
-
-           |
-         ‘vhost_svq_get_buf’: event 14
-           |
-           |  423 |     if (!vhost_svq_more_used(svq)) {
-           |      |        ^
-           |      |        |
-           |      |        (14) following ‘false’ branch...
-           |
-         ‘vhost_svq_get_buf’: event 15
-           |
-           |cc1:
-           | (15): ...to here
-           |
-    <------+
-    |
-  ‘vhost_svq_poll’: events 16-17
-    |
-    |  537 |     vhost_svq_get_buf(svq, &len);
-    |      |     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~
-    |      |     |
-    |      |     (16) returning to ‘vhost_svq_poll’ from ‘vhost_svq_get_buf’
-    |  538 |     return len;
-    |      |            ~~~
-    |      |            |
-    |      |            (17) use of uninitialized value ‘len’ here
-
-Note by  Laurent Vivier <lvivier@redhat.com>:
-
-    The return value is only used to detect an error:
-
-    vhost_svq_poll
-        vhost_vdpa_net_cvq_add
-            vhost_vdpa_net_load_cmd
-                vhost_vdpa_net_load_mac
-                  -> a negative return is only used to detect error
-                vhost_vdpa_net_load_mq
-                  -> a negative return is only used to detect error
-            vhost_vdpa_net_handle_ctrl_avail
-              -> a negative return is only used to detect error
-
-Fixes: d368c0b052ad ("vhost: Do not depend on !NULL VirtQueueElement on vhost_svq_flush")
-Signed-off-by: Carlos López <clopez@suse.de>
-Message-Id: <20230213085747.19956-1-clopez@suse.de>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit e4dd39c699b7d63a06f686ec06ded8adbee989c1)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/virtio/vhost-shadow-virtqueue.c | 2 +-
- 1 file changed, 1 insertion(+), 1 deletion(-)
-
-diff --git a/hw/virtio/vhost-shadow-virtqueue.c b/hw/virtio/vhost-shadow-virtqueue.c
-index 5bd14cad96..a723073747 100644
---- a/hw/virtio/vhost-shadow-virtqueue.c
-+++ b/hw/virtio/vhost-shadow-virtqueue.c
-@@ -522,7 +522,7 @@ static void vhost_svq_flush(VhostShadowVirtqueue *svq,
- size_t vhost_svq_poll(VhostShadowVirtqueue *svq)
- {
-     int64_t start_us = g_get_monotonic_time();
--    uint32_t len;
-+    uint32_t len = 0;
- 
-     do {
-         if (vhost_svq_more_used(svq)) {

debian/patches/extra/0018-chardev-char-socket-set-s-listener-NULL-in-char_sock.patch

@@ -1,70 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Yajun Wu <yajunw@nvidia.com>
-Date: Tue, 14 Feb 2023 10:14:30 +0800
-Subject: [PATCH] chardev/char-socket: set s->listener = NULL in
- char_socket_finalize
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-After live migration with virtio block device, qemu crash at:
-
-	#0  0x000055914f46f795 in object_dynamic_cast_assert (obj=0x559151b7b090, typename=0x55914f80fbc4 "qio-channel", file=0x55914f80fb90 "/images/testvfe/sw/qemu.gerrit/include/io/channel.h", line=30, func=0x55914f80fcb8 <__func__.17257> "QIO_CHANNEL") at ../qom/object.c:872
-	#1  0x000055914f480d68 in QIO_CHANNEL (obj=0x559151b7b090) at /images/testvfe/sw/qemu.gerrit/include/io/channel.h:29
-	#2  0x000055914f4812f8 in qio_net_listener_set_client_func_full (listener=0x559151b7a720, func=0x55914f580b97 <tcp_chr_accept>, data=0x5591519f4ea0, notify=0x0, context=0x0) at ../io/net-listener.c:166
-	#3  0x000055914f580059 in tcp_chr_update_read_handler (chr=0x5591519f4ea0) at ../chardev/char-socket.c:637
-	#4  0x000055914f583dca in qemu_chr_be_update_read_handlers (s=0x5591519f4ea0, context=0x0) at ../chardev/char.c:226
-	#5  0x000055914f57b7c9 in qemu_chr_fe_set_handlers_full (b=0x559152bf23a0, fd_can_read=0x0, fd_read=0x0, fd_event=0x0, be_change=0x0, opaque=0x0, context=0x0, set_open=false, sync_state=true) at ../chardev/char-fe.c:279
-	#6  0x000055914f57b86d in qemu_chr_fe_set_handlers (b=0x559152bf23a0, fd_can_read=0x0, fd_read=0x0, fd_event=0x0, be_change=0x0, opaque=0x0, context=0x0, set_open=false) at ../chardev/char-fe.c:304
-	#7  0x000055914f378caf in vhost_user_async_close (d=0x559152bf21a0, chardev=0x559152bf23a0, vhost=0x559152bf2420, cb=0x55914f2fb8c1 <vhost_user_blk_disconnect>) at ../hw/virtio/vhost-user.c:2725
-	#8  0x000055914f2fba40 in vhost_user_blk_event (opaque=0x559152bf21a0, event=CHR_EVENT_CLOSED) at ../hw/block/vhost-user-blk.c:395
-	#9  0x000055914f58388c in chr_be_event (s=0x5591519f4ea0, event=CHR_EVENT_CLOSED) at ../chardev/char.c:61
-	#10 0x000055914f583905 in qemu_chr_be_event (s=0x5591519f4ea0, event=CHR_EVENT_CLOSED) at ../chardev/char.c:81
-	#11 0x000055914f581275 in char_socket_finalize (obj=0x5591519f4ea0) at ../chardev/char-socket.c:1083
-	#12 0x000055914f46f073 in object_deinit (obj=0x5591519f4ea0, type=0x5591519055c0) at ../qom/object.c:680
-	#13 0x000055914f46f0e5 in object_finalize (data=0x5591519f4ea0) at ../qom/object.c:694
-	#14 0x000055914f46ff06 in object_unref (objptr=0x5591519f4ea0) at ../qom/object.c:1202
-	#15 0x000055914f4715a4 in object_finalize_child_property (obj=0x559151b76c50, name=0x559151b7b250 "char3", opaque=0x5591519f4ea0) at ../qom/object.c:1747
-	#16 0x000055914f46ee86 in object_property_del_all (obj=0x559151b76c50) at ../qom/object.c:632
-	#17 0x000055914f46f0d2 in object_finalize (data=0x559151b76c50) at ../qom/object.c:693
-	#18 0x000055914f46ff06 in object_unref (objptr=0x559151b76c50) at ../qom/object.c:1202
-	#19 0x000055914f4715a4 in object_finalize_child_property (obj=0x559151b6b560, name=0x559151b76630 "chardevs", opaque=0x559151b76c50) at ../qom/object.c:1747
-	#20 0x000055914f46ef67 in object_property_del_child (obj=0x559151b6b560, child=0x559151b76c50) at ../qom/object.c:654
-	#21 0x000055914f46f042 in object_unparent (obj=0x559151b76c50) at ../qom/object.c:673
-	#22 0x000055914f58632a in qemu_chr_cleanup () at ../chardev/char.c:1189
-	#23 0x000055914f16c66c in qemu_cleanup () at ../softmmu/runstate.c:830
-	#24 0x000055914eee7b9e in qemu_default_main () at ../softmmu/main.c:38
-	#25 0x000055914eee7bcc in main (argc=86, argv=0x7ffc97cb8d88) at ../softmmu/main.c:48
-
-In char_socket_finalize after s->listener freed, event callback function
-vhost_user_blk_event will be called to handle CHR_EVENT_CLOSED.
-vhost_user_blk_event is calling qio_net_listener_set_client_func_full which
-is still using s->listener.
-
-Setting s->listener = NULL after object_unref(OBJECT(s->listener)) can
-solve this issue.
-
-Signed-off-by: Yajun Wu <yajunw@nvidia.com>
-Acked-by: Jiri Pirko <jiri@nvidia.com>
-Message-Id: <20230214021430.3638579-1-yajunw@nvidia.com>
-Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit b8a7f51f59e28d5a8e0c07ed3919cc9695560ed2)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- chardev/char-socket.c | 1 +
- 1 file changed, 1 insertion(+)
-
-diff --git a/chardev/char-socket.c b/chardev/char-socket.c
-index 879564aa8a..b00efb1482 100644
---- a/chardev/char-socket.c
-+++ b/chardev/char-socket.c
-@@ -1065,6 +1065,7 @@ static void char_socket_finalize(Object *obj)
-         qio_net_listener_set_client_func_full(s->listener, NULL, NULL,
-                                               NULL, chr->gcontext);
-         object_unref(OBJECT(s->listener));
-+        s->listener = NULL;
-     }
-     if (s->tls_creds) {
-         object_unref(OBJECT(s->tls_creds));

debian/patches/extra/0019-intel-iommu-fail-MAP-notifier-without-caching-mode.patch

@@ -1,41 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Jason Wang <jasowang@redhat.com>
-Date: Thu, 23 Feb 2023 14:59:20 +0800
-Subject: [PATCH] intel-iommu: fail MAP notifier without caching mode
-
-Without caching mode, MAP notifier won't work correctly since guest
-won't send IOTLB update event when it establishes new mappings in the
-I/O page tables. Let's fail the IOMMU notifiers early instead of
-misbehaving silently.
-
-Reviewed-by: Eric Auger <eric.auger@redhat.com>
-Tested-by: Viktor Prutyanov <viktor@daynix.com>
-Signed-off-by: Jason Wang <jasowang@redhat.com>
-Message-Id: <20230223065924.42503-2-jasowang@redhat.com>
-Reviewed-by: Peter Xu <peterx@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit b8d78277c091f26fdd64f239bc8bb7e55d74cecf)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/i386/intel_iommu.c | 7 +++++++
- 1 file changed, 7 insertions(+)
-
-diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
-index a08ee85edf..9143376677 100644
---- a/hw/i386/intel_iommu.c
-+++ b/hw/i386/intel_iommu.c
-@@ -3186,6 +3186,13 @@ static int vtd_iommu_notify_flag_changed(IOMMUMemoryRegion *iommu,
-                          "Snoop Control with vhost or VFIO is not supported");
-         return -ENOTSUP;
-     }
-+    if (!s->caching_mode && (new & IOMMU_NOTIFIER_MAP)) {
-+        error_setg_errno(errp, ENOTSUP,
-+                         "device %02x.%02x.%x requires caching mode",
-+                         pci_bus_num(vtd_as->bus), PCI_SLOT(vtd_as->devfn),
-+                         PCI_FUNC(vtd_as->devfn));
-+        return -ENOTSUP;
-+    }
- 
-     /* Update per-address-space notifier flags */
-     vtd_as->notifier_flags = new;

debian/patches/extra/0020-intel-iommu-fail-DEVIOTLB_UNMAP-without-dt-mode.patch

@@ -1,50 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Jason Wang <jasowang@redhat.com>
-Date: Thu, 23 Feb 2023 14:59:21 +0800
-Subject: [PATCH] intel-iommu: fail DEVIOTLB_UNMAP without dt mode
-
-Without dt mode, device IOTLB notifier won't work since guest won't
-send device IOTLB invalidation descriptor in this case. Let's fail
-early instead of misbehaving silently.
-
-Reviewed-by: Laurent Vivier <lvivier@redhat.com>
-Tested-by: Laurent Vivier <lvivier@redhat.com>
-Tested-by: Viktor Prutyanov <viktor@daynix.com>
-Buglink: https://bugzilla.redhat.com/2156876
-Signed-off-by: Jason Wang <jasowang@redhat.com>
-Message-Id: <20230223065924.42503-3-jasowang@redhat.com>
-Reviewed-by: Peter Xu <peterx@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit 09adb0e021207b60a0c51a68939b4539d98d3ef3)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/i386/intel_iommu.c | 8 ++++++++
- 1 file changed, 8 insertions(+)
-
-diff --git a/hw/i386/intel_iommu.c b/hw/i386/intel_iommu.c
-index 9143376677..d025ef2873 100644
---- a/hw/i386/intel_iommu.c
-+++ b/hw/i386/intel_iommu.c
-@@ -3179,6 +3179,7 @@ static int vtd_iommu_notify_flag_changed(IOMMUMemoryRegion *iommu,
- {
-     VTDAddressSpace *vtd_as = container_of(iommu, VTDAddressSpace, iommu);
-     IntelIOMMUState *s = vtd_as->iommu_state;
-+    X86IOMMUState *x86_iommu = X86_IOMMU_DEVICE(s);
- 
-     /* TODO: add support for VFIO and vhost users */
-     if (s->snoop_control) {
-@@ -3193,6 +3194,13 @@ static int vtd_iommu_notify_flag_changed(IOMMUMemoryRegion *iommu,
-                          PCI_FUNC(vtd_as->devfn));
-         return -ENOTSUP;
-     }
-+    if (!x86_iommu->dt_supported && (new & IOMMU_NOTIFIER_DEVIOTLB_UNMAP)) {
-+        error_setg_errno(errp, ENOTSUP,
-+                         "device %02x.%02x.%x requires device IOTLB mode",
-+                         pci_bus_num(vtd_as->bus), PCI_SLOT(vtd_as->devfn),
-+                         PCI_FUNC(vtd_as->devfn));
-+        return -ENOTSUP;
-+    }
- 
-     /* Update per-address-space notifier flags */
-     vtd_as->notifier_flags = new;

debian/patches/extra/0021-memory-Allow-disabling-re-entrancy-checking-per-MR.patch

@@ -1,38 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Alexander Bulekov <alxndr@bu.edu>
-Date: Mon, 13 Mar 2023 04:24:16 -0400
-Subject: [PATCH] memory: Allow disabling re-entrancy checking per-MR
-
-Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
----
- include/exec/memory.h | 3 +++
- softmmu/memory.c      | 2 +-
- 2 files changed, 4 insertions(+), 1 deletion(-)
-
-diff --git a/include/exec/memory.h b/include/exec/memory.h
-index 91f8a2395a..d7268d9f39 100644
---- a/include/exec/memory.h
-+++ b/include/exec/memory.h
-@@ -765,6 +765,9 @@ struct MemoryRegion {
-     unsigned ioeventfd_nb;
-     MemoryRegionIoeventfd *ioeventfds;
-     RamDiscardManager *rdm; /* Only for RAM */
-+
-+    /* For devices designed to perform re-entrant IO into their own IO MRs */
-+    bool disable_reentrancy_guard;
- };
- 
- struct IOMMUMemoryRegion {
-diff --git a/softmmu/memory.c b/softmmu/memory.c
-index 7dcb3347aa..2b46714191 100644
---- a/softmmu/memory.c
-+++ b/softmmu/memory.c
-@@ -544,7 +544,7 @@ static MemTxResult access_with_adjusted_size(hwaddr addr,
-     }
- 
-     /* Do not allow more than one simultanous access to a device's IO Regions */
--    if (mr->owner &&
-+    if (mr->owner && !mr->disable_reentrancy_guard &&
-         !mr->ram_device && !mr->ram && !mr->rom_device && !mr->readonly) {
-         dev = (DeviceState *) object_dynamic_cast(mr->owner, TYPE_DEVICE);
-         if (dev) {

debian/patches/extra/0022-lsi53c895a-disable-reentrancy-detection-for-script-R.patch

@@ -1,33 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Alexander Bulekov <alxndr@bu.edu>
-Date: Mon, 13 Mar 2023 04:24:17 -0400
-Subject: [PATCH] lsi53c895a: disable reentrancy detection for script RAM
-
-As the code is designed to use the memory APIs to access the script ram,
-disable reentrancy checks for the pseudo-RAM ram_io MemoryRegion.
-
-In the future, ram_io may be converted from an IO to a proper RAM MemoryRegion.
-
-Reported-by: Fiona Ebner <f.ebner@proxmox.com>
-Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
----
- hw/scsi/lsi53c895a.c | 6 ++++++
- 1 file changed, 6 insertions(+)
-
-diff --git a/hw/scsi/lsi53c895a.c b/hw/scsi/lsi53c895a.c
-index 50979640c3..894b9311ac 100644
---- a/hw/scsi/lsi53c895a.c
-+++ b/hw/scsi/lsi53c895a.c
-@@ -2302,6 +2302,12 @@ static void lsi_scsi_realize(PCIDevice *dev, Error **errp)
-     memory_region_init_io(&s->io_io, OBJECT(s), &lsi_io_ops, s,
-                           "lsi-io", 256);
- 
-+    /*
-+     * Since we use the address-space API to interact with ram_io, disable the
-+     * re-entrancy guard.
-+     */
-+    s->ram_io.disable_reentrancy_guard = true;
-+
-     address_space_init(&s->pci_io_as, pci_address_space_io(dev), "lsi-pci-io");
-     qdev_init_gpio_out(d, &s->ext_irq, 1);
- 

debian/patches/extra/0023-acpi-cpuhp-fix-guest-visible-maximum-access-size-to-.patch

@@ -1,166 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Laszlo Ersek <lersek@redhat.com>
-Date: Thu, 5 Jan 2023 17:18:04 +0100
-Subject: [PATCH] acpi: cpuhp: fix guest-visible maximum access size to the
- legacy reg block
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-The modern ACPI CPU hotplug interface was introduced in the following
-series (aa1dd39ca307..679dd1a957df), released in v2.7.0:
-
-  1  abd49bc2ed2f docs: update ACPI CPU hotplug spec with new protocol
-  2  16bcab97eb9f pc: piix4/ich9: add 'cpu-hotplug-legacy' property
-  3  5e1b5d93887b acpi: cpuhp: add CPU devices AML with _STA method
-  4  ac35f13ba8f8 pc: acpi: introduce AcpiDeviceIfClass.madt_cpu hook
-  5  d2238cb6781d acpi: cpuhp: implement hot-add parts of CPU hotplug
-                  interface
-  6  8872c25a26cc acpi: cpuhp: implement hot-remove parts of CPU hotplug
-                  interface
-  7  76623d00ae57 acpi: cpuhp: add cpu._OST handling
-  8  679dd1a957df pc: use new CPU hotplug interface since 2.7 machine type
-
-Before patch#1, "docs/specs/acpi_cpu_hotplug.txt" only specified 1-byte
-accesses for the hotplug register block.  Patch#1 preserved the same
-restriction for the legacy register block, but:
-
-- it specified DWORD accesses for some of the modern registers,
-
-- in particular, the switch from the legacy block to the modern block
-  would require a DWORD write to the *legacy* block.
-
-The latter functionality was then implemented in cpu_status_write()
-[hw/acpi/cpu_hotplug.c], in patch#8.
-
-Unfortunately, all DWORD accesses depended on a dormant bug: the one
-introduced in earlier commit a014ed07bd5a ("memory: accept mismatching
-sizes in memory_region_access_valid", 2013-05-29); first released in
-v1.6.0.  Due to commit a014ed07bd5a, the DWORD accesses to the *legacy*
-CPU hotplug register block would work in spite of the above series *not*
-relaxing "valid.max_access_size = 1" in "hw/acpi/cpu_hotplug.c":
-
-> static const MemoryRegionOps AcpiCpuHotplug_ops = {
->     .read = cpu_status_read,
->     .write = cpu_status_write,
->     .endianness = DEVICE_LITTLE_ENDIAN,
->     .valid = {
->         .min_access_size = 1,
->         .max_access_size = 1,
->     },
-> };
-
-Later, in commits e6d0c3ce6895 ("acpi: cpuhp: introduce 'Command data 2'
-field", 2020-01-22) and ae340aa3d256 ("acpi: cpuhp: spec: add typical
-usecases", 2020-01-22), first released in v5.0.0, the modern CPU hotplug
-interface (including the documentation) was extended with another DWORD
-*read* access, namely to the "Command data 2" register, which would be
-important for the guest to confirm whether it managed to switch the
-register block from legacy to modern.
-
-This functionality too silently depended on the bug from commit
-a014ed07bd5a.
-
-In commit 5d971f9e6725 ('memory: Revert "memory: accept mismatching sizes
-in memory_region_access_valid"', 2020-06-26), first released in v5.1.0,
-the bug from commit a014ed07bd5a was fixed (the commit was reverted).
-That swiftly exposed the bug in "AcpiCpuHotplug_ops", still present from
-the v2.7.0 series quoted at the top -- namely the fact that
-"valid.max_access_size = 1" didn't match what the guest was supposed to
-do, according to the spec ("docs/specs/acpi_cpu_hotplug.txt").
-
-The symptom is that the "modern interface negotiation protocol"
-described in commit ae340aa3d256:
-
-> +      Use following steps to detect and enable modern CPU hotplug interface:
-> +        1. Store 0x0 to the 'CPU selector' register,
-> +           attempting to switch to modern mode
-> +        2. Store 0x0 to the 'CPU selector' register,
-> +           to ensure valid selector value
-> +        3. Store 0x0 to the 'Command field' register,
-> +        4. Read the 'Command data 2' register.
-> +           If read value is 0x0, the modern interface is enabled.
-> +           Otherwise legacy or no CPU hotplug interface available
-
-falls apart for the guest: steps 1 and 2 are lost, because they are DWORD
-writes; so no switching happens.  Step 3 (a single-byte write) is not
-lost, but it has no effect; see the condition in cpu_status_write() in
-patch#8.  And step 4 *misleads* the guest into thinking that the switch
-worked: the DWORD read is lost again -- it returns zero to the guest
-without ever reaching the device model, so the guest never learns the
-switch didn't work.
-
-This means that guest behavior centered on the "Command data 2" register
-worked *only* in the v5.0.0 release; it got effectively regressed in
-v5.1.0.
-
-To make things *even more* complicated, the breakage was (and remains, as
-of today) visible with TCG acceleration only.  Commit 5d971f9e6725 makes
-no difference with KVM acceleration -- the DWORD accesses still work,
-despite "valid.max_access_size = 1".
-
-As commit 5d971f9e6725 suggests, fix the problem by raising
-"valid.max_access_size" to 4 -- the spec now clearly instructs the guest
-to perform DWORD accesses to the legacy register block too, for enabling
-(and verifying!) the modern block.  In order to keep compatibility for the
-device model implementation though, set "impl.max_access_size = 1", so
-that wide accesses be split before they reach the legacy read/write
-handlers, like they always have been on KVM, and like they were on TCG
-before 5d971f9e6725 (v5.1.0).
-
-Tested with:
-
-- OVMF IA32 + qemu-system-i386, CPU hotplug/hot-unplug with SMM,
-  intermixed with ACPI S3 suspend/resume, using KVM accel
-  (regression-test);
-
-- OVMF IA32X64 + qemu-system-x86_64, CPU hotplug/hot-unplug with SMM,
-  intermixed with ACPI S3 suspend/resume, using KVM accel
-  (regression-test);
-
-- OVMF IA32 + qemu-system-i386, SMM enabled, using TCG accel; verified the
-  register block switch and the present/possible CPU counting through the
-  modern hotplug interface, during OVMF boot (bugfix test);
-
-- I do not have any testcase (guest payload) for regression-testing CPU
-  hotplug through the *legacy* CPU hotplug register block.
-
-Cc: "Michael S. Tsirkin" <mst@redhat.com>
-Cc: Ani Sinha <ani@anisinha.ca>
-Cc: Ard Biesheuvel <ardb@kernel.org>
-Cc: Igor Mammedov <imammedo@redhat.com>
-Cc: Paolo Bonzini <pbonzini@redhat.com>
-Cc: Peter Maydell <peter.maydell@linaro.org>
-Cc: Philippe Mathieu-Daudé <philmd@linaro.org>
-Cc: qemu-stable@nongnu.org
-Ref: "IO port write width clamping differs between TCG and KVM"
-Link: http://mid.mail-archive.com/aaedee84-d3ed-a4f9-21e7-d221a28d1683@redhat.com
-Link: https://lists.gnu.org/archive/html/qemu-devel/2023-01/msg00199.html
-Reported-by: Ard Biesheuvel <ardb@kernel.org>
-Signed-off-by: Laszlo Ersek <lersek@redhat.com>
-Tested-by: Ard Biesheuvel <ardb@kernel.org>
-Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
-Tested-by: Igor Mammedov <imammedo@redhat.com>
-Message-Id: <20230105161804.82486-1-lersek@redhat.com>
-Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
-Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
-(cherry-picked from commit dab30fbef3896bb652a09d46c37d3f55657cbcbb)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- hw/acpi/cpu_hotplug.c | 3 +++
- 1 file changed, 3 insertions(+)
-
-diff --git a/hw/acpi/cpu_hotplug.c b/hw/acpi/cpu_hotplug.c
-index 53654f8638..ff14c3f410 100644
---- a/hw/acpi/cpu_hotplug.c
-+++ b/hw/acpi/cpu_hotplug.c
-@@ -52,6 +52,9 @@ static const MemoryRegionOps AcpiCpuHotplug_ops = {
-     .endianness = DEVICE_LITTLE_ENDIAN,
-     .valid = {
-         .min_access_size = 1,
-+        .max_access_size = 4,
-+    },
-+    .impl = {
-         .max_access_size = 1,
-     },
- };

debian/patches/extra/0024-tests-tcg-i386-Introduce-and-use-reg_t-consistently.patch

@@ -1,286 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Richard Henderson <richard.henderson@linaro.org>
-Date: Sat, 14 Jan 2023 13:05:41 -1000
-Subject: [PATCH] tests/tcg/i386: Introduce and use reg_t consistently
-MIME-Version: 1.0
-Content-Type: text/plain; charset=UTF-8
-Content-Transfer-Encoding: 8bit
-
-Define reg_t based on the actual register width.
-Define the inlines using that type.  This will allow
-input registers to 32-bit insns to be set to 64-bit
-values on x86-64, which allows testing various edge cases.
-
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
-Message-Id: <20230114230542.3116013-2-richard.henderson@linaro.org>
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit 5d62d6649cd367b5b4a3676e7514d2f9ca86cb03)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- tests/tcg/i386/test-i386-bmi2.c | 182 ++++++++++++++++----------------
- 1 file changed, 93 insertions(+), 89 deletions(-)
-
-diff --git a/tests/tcg/i386/test-i386-bmi2.c b/tests/tcg/i386/test-i386-bmi2.c
-index 5fadf47510..3c3ef85513 100644
---- a/tests/tcg/i386/test-i386-bmi2.c
-+++ b/tests/tcg/i386/test-i386-bmi2.c
-@@ -3,34 +3,40 @@
- #include <stdint.h>
- #include <stdio.h>
- 
-+#ifdef __x86_64
-+typedef uint64_t reg_t;
-+#else
-+typedef uint32_t reg_t;
-+#endif
-+
- #define insn1q(name, arg0)                                                           \
--static inline uint64_t name##q(uint64_t arg0)                                        \
-+static inline reg_t name##q(reg_t arg0)                                              \
- {                                                                                    \
--    uint64_t result64;                                                               \
-+    reg_t result64;                                                                  \
-     asm volatile (#name "q   %1, %0" : "=r"(result64) : "rm"(arg0));                 \
-     return result64;                                                                 \
- }
- 
- #define insn1l(name, arg0)                                                           \
--static inline uint32_t name##l(uint32_t arg0)                                        \
-+static inline reg_t name##l(reg_t arg0)                                              \
- {                                                                                    \
--    uint32_t result32;                                                               \
-+    reg_t result32;                                                                  \
-     asm volatile (#name "l   %k1, %k0" : "=r"(result32) : "rm"(arg0));               \
-     return result32;                                                                 \
- }
- 
- #define insn2q(name, arg0, c0, arg1, c1)                                             \
--static inline uint64_t name##q(uint64_t arg0, uint64_t arg1)                         \
-+static inline reg_t name##q(reg_t arg0, reg_t arg1)                                  \
- {                                                                                    \
--    uint64_t result64;                                                               \
-+    reg_t result64;                                                                  \
-     asm volatile (#name "q   %2, %1, %0" : "=r"(result64) : c0(arg0), c1(arg1));     \
-     return result64;                                                                 \
- }
- 
- #define insn2l(name, arg0, c0, arg1, c1)                                             \
--static inline uint32_t name##l(uint32_t arg0, uint32_t arg1)                         \
-+static inline reg_t name##l(reg_t arg0, reg_t arg1)                                  \
- {                                                                                    \
--    uint32_t result32;                                                               \
-+    reg_t result32;                                                                  \
-     asm volatile (#name "l   %k2, %k1, %k0" : "=r"(result32) : c0(arg0), c1(arg1));  \
-     return result32;                                                                 \
- }
-@@ -65,130 +71,128 @@ insn1l(blsr, src)
- int main(int argc, char *argv[]) {
-     uint64_t ehlo = 0x202020204f4c4845ull;
-     uint64_t mask = 0xa080800302020001ull;
--    uint32_t result32;
-+    reg_t result;
- 
- #ifdef __x86_64
--    uint64_t result64;
--
-     /* 64 bits */
--    result64 = andnq(mask, ehlo);
--    assert(result64 == 0x002020204d4c4844);
-+    result = andnq(mask, ehlo);
-+    assert(result == 0x002020204d4c4844);
- 
--    result64 = pextq(ehlo, mask);
--    assert(result64 == 133);
-+    result = pextq(ehlo, mask);
-+    assert(result == 133);
- 
--    result64 = pdepq(result64, mask);
--    assert(result64 == (ehlo & mask));
-+    result = pdepq(result, mask);
-+    assert(result == (ehlo & mask));
- 
--    result64 = pextq(-1ull, mask);
--    assert(result64 == 511); /* mask has 9 bits set */
-+    result = pextq(-1ull, mask);
-+    assert(result == 511); /* mask has 9 bits set */
- 
--    result64 = pdepq(-1ull, mask);
--    assert(result64 == mask);
-+    result = pdepq(-1ull, mask);
-+    assert(result == mask);
- 
--    result64 = bextrq(mask, 0x3f00);
--    assert(result64 == (mask & ~INT64_MIN));
-+    result = bextrq(mask, 0x3f00);
-+    assert(result == (mask & ~INT64_MIN));
- 
--    result64 = bextrq(mask, 0x1038);
--    assert(result64 == 0xa0);
-+    result = bextrq(mask, 0x1038);
-+    assert(result == 0xa0);
- 
--    result64 = bextrq(mask, 0x10f8);
--    assert(result64 == 0);
-+    result = bextrq(mask, 0x10f8);
-+    assert(result == 0);
- 
--    result64 = blsiq(0x30);
--    assert(result64 == 0x10);
-+    result = blsiq(0x30);
-+    assert(result == 0x10);
- 
--    result64 = blsiq(0x30ull << 32);
--    assert(result64 == 0x10ull << 32);
-+    result = blsiq(0x30ull << 32);
-+    assert(result == 0x10ull << 32);
- 
--    result64 = blsmskq(0x30);
--    assert(result64 == 0x1f);
-+    result = blsmskq(0x30);
-+    assert(result == 0x1f);
- 
--    result64 = blsrq(0x30);
--    assert(result64 == 0x20);
-+    result = blsrq(0x30);
-+    assert(result == 0x20);
- 
--    result64 = blsrq(0x30ull << 32);
--    assert(result64 == 0x20ull << 32);
-+    result = blsrq(0x30ull << 32);
-+    assert(result == 0x20ull << 32);
- 
--    result64 = bzhiq(mask, 0x3f);
--    assert(result64 == (mask & ~INT64_MIN));
-+    result = bzhiq(mask, 0x3f);
-+    assert(result == (mask & ~INT64_MIN));
- 
--    result64 = bzhiq(mask, 0x1f);
--    assert(result64 == (mask & ~(-1 << 30)));
-+    result = bzhiq(mask, 0x1f);
-+    assert(result == (mask & ~(-1 << 30)));
- 
--    result64 = rorxq(0x2132435465768798, 8);
--    assert(result64 == 0x9821324354657687);
-+    result = rorxq(0x2132435465768798, 8);
-+    assert(result == 0x9821324354657687);
- 
--    result64 = sarxq(0xffeeddccbbaa9988, 8);
--    assert(result64 == 0xffffeeddccbbaa99);
-+    result = sarxq(0xffeeddccbbaa9988, 8);
-+    assert(result == 0xffffeeddccbbaa99);
- 
--    result64 = sarxq(0x77eeddccbbaa9988, 8 | 64);
--    assert(result64 == 0x0077eeddccbbaa99);
-+    result = sarxq(0x77eeddccbbaa9988, 8 | 64);
-+    assert(result == 0x0077eeddccbbaa99);
- 
--    result64 = shrxq(0xffeeddccbbaa9988, 8);
--    assert(result64 == 0x00ffeeddccbbaa99);
-+    result = shrxq(0xffeeddccbbaa9988, 8);
-+    assert(result == 0x00ffeeddccbbaa99);
- 
--    result64 = shrxq(0x77eeddccbbaa9988, 8 | 192);
--    assert(result64 == 0x0077eeddccbbaa99);
-+    result = shrxq(0x77eeddccbbaa9988, 8 | 192);
-+    assert(result == 0x0077eeddccbbaa99);
- 
--    result64 = shlxq(0xffeeddccbbaa9988, 8);
--    assert(result64 == 0xeeddccbbaa998800);
-+    result = shlxq(0xffeeddccbbaa9988, 8);
-+    assert(result == 0xeeddccbbaa998800);
- #endif
- 
-     /* 32 bits */
--    result32 = andnl(mask, ehlo);
--    assert(result32 == 0x04d4c4844);
-+    result = andnl(mask, ehlo);
-+    assert(result == 0x04d4c4844);
- 
--    result32 = pextl((uint32_t) ehlo, mask);
--    assert(result32 == 5);
-+    result = pextl((uint32_t) ehlo, mask);
-+    assert(result == 5);
- 
--    result32 = pdepl(result32, mask);
--    assert(result32 == (uint32_t)(ehlo & mask));
-+    result = pdepl(result, mask);
-+    assert(result == (uint32_t)(ehlo & mask));
- 
--    result32 = pextl(-1u, mask);
--    assert(result32 == 7); /* mask has 3 bits set */
-+    result = pextl(-1u, mask);
-+    assert(result == 7); /* mask has 3 bits set */
- 
--    result32 = pdepl(-1u, mask);
--    assert(result32 == (uint32_t)mask);
-+    result = pdepl(-1u, mask);
-+    assert(result == (uint32_t)mask);
- 
--    result32 = bextrl(mask, 0x1f00);
--    assert(result32 == (mask & ~INT32_MIN));
-+    result = bextrl(mask, 0x1f00);
-+    assert(result == (mask & ~INT32_MIN));
- 
--    result32 = bextrl(ehlo, 0x1018);
--    assert(result32 == 0x4f);
-+    result = bextrl(ehlo, 0x1018);
-+    assert(result == 0x4f);
- 
--    result32 = bextrl(mask, 0x1038);
--    assert(result32 == 0);
-+    result = bextrl(mask, 0x1038);
-+    assert(result == 0);
- 
--    result32 = blsil(0xffff);
--    assert(result32 == 1);
-+    result = blsil(0xffff);
-+    assert(result == 1);
- 
--    result32 = blsmskl(0x300);
--    assert(result32 == 0x1ff);
-+    result = blsmskl(0x300);
-+    assert(result == 0x1ff);
- 
--    result32 = blsrl(0xffc);
--    assert(result32 == 0xff8);
-+    result = blsrl(0xffc);
-+    assert(result == 0xff8);
- 
--    result32 = bzhil(mask, 0xf);
--    assert(result32 == 1);
-+    result = bzhil(mask, 0xf);
-+    assert(result == 1);
- 
--    result32 = rorxl(0x65768798, 8);
--    assert(result32 == 0x98657687);
-+    result = rorxl(0x65768798, 8);
-+    assert(result == 0x98657687);
- 
--    result32 = sarxl(0xffeeddcc, 8);
--    assert(result32 == 0xffffeedd);
-+    result = sarxl(0xffeeddcc, 8);
-+    assert(result == 0xffffeedd);
- 
--    result32 = sarxl(0x77eeddcc, 8 | 32);
--    assert(result32 == 0x0077eedd);
-+    result = sarxl(0x77eeddcc, 8 | 32);
-+    assert(result == 0x0077eedd);
- 
--    result32 = shrxl(0xffeeddcc, 8);
--    assert(result32 == 0x00ffeedd);
-+    result = shrxl(0xffeeddcc, 8);
-+    assert(result == 0x00ffeedd);
- 
--    result32 = shrxl(0x77eeddcc, 8 | 128);
--    assert(result32 == 0x0077eedd);
-+    result = shrxl(0x77eeddcc, 8 | 128);
-+    assert(result == 0x0077eedd);
- 
--    result32 = shlxl(0xffeeddcc, 8);
--    assert(result32 == 0xeeddcc00);
-+    result = shlxl(0xffeeddcc, 8);
-+    assert(result == 0xeeddcc00);
- 
-     return 0;
- }

debian/patches/extra/0025-target-i386-Fix-BEXTR-instruction.patch

@@ -1,97 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Richard Henderson <richard.henderson@linaro.org>
-Date: Sat, 14 Jan 2023 13:05:42 -1000
-Subject: [PATCH] target/i386: Fix BEXTR instruction
-
-There were two problems here: not limiting the input to operand bits,
-and not correctly handling large extraction length.
-
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1372
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Message-Id: <20230114230542.3116013-3-richard.henderson@linaro.org>
-Cc: qemu-stable@nongnu.org
-Fixes: 1d0b926150e5 ("target/i386: move scalar 0F 38 and 0F 3A instruction to new decoder", 2022-10-18)
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit b14c0098975264ed03144f145bca0179a6763a07)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- target/i386/tcg/emit.c.inc      | 22 +++++++++++-----------
- tests/tcg/i386/test-i386-bmi2.c | 12 ++++++++++++
- 2 files changed, 23 insertions(+), 11 deletions(-)
-
-diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc
-index 7037ff91c6..99f6ba6e19 100644
---- a/target/i386/tcg/emit.c.inc
-+++ b/target/i386/tcg/emit.c.inc
-@@ -1078,30 +1078,30 @@ static void gen_ANDN(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
- static void gen_BEXTR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
- {
-     MemOp ot = decode->op[0].ot;
--    TCGv bound, zero;
-+    TCGv bound = tcg_constant_tl(ot == MO_64 ? 63 : 31);
-+    TCGv zero = tcg_constant_tl(0);
-+    TCGv mone = tcg_constant_tl(-1);
- 
-     /*
-      * Extract START, and shift the operand.
-      * Shifts larger than operand size get zeros.
-      */
-     tcg_gen_ext8u_tl(s->A0, s->T1);
-+    if (TARGET_LONG_BITS == 64 && ot == MO_32) {
-+        tcg_gen_ext32u_tl(s->T0, s->T0);
-+    }
-     tcg_gen_shr_tl(s->T0, s->T0, s->A0);
- 
--    bound = tcg_constant_tl(ot == MO_64 ? 63 : 31);
--    zero = tcg_constant_tl(0);
-     tcg_gen_movcond_tl(TCG_COND_LEU, s->T0, s->A0, bound, s->T0, zero);
- 
-     /*
--     * Extract the LEN into a mask.  Lengths larger than
--     * operand size get all ones.
-+     * Extract the LEN into an inverse mask.  Lengths larger than
-+     * operand size get all zeros, length 0 gets all ones.
-      */
-     tcg_gen_extract_tl(s->A0, s->T1, 8, 8);
--    tcg_gen_movcond_tl(TCG_COND_LEU, s->A0, s->A0, bound, s->A0, bound);
--
--    tcg_gen_movi_tl(s->T1, 1);
--    tcg_gen_shl_tl(s->T1, s->T1, s->A0);
--    tcg_gen_subi_tl(s->T1, s->T1, 1);
--    tcg_gen_and_tl(s->T0, s->T0, s->T1);
-+    tcg_gen_shl_tl(s->T1, mone, s->A0);
-+    tcg_gen_movcond_tl(TCG_COND_LEU, s->T1, s->A0, bound, s->T1, zero);
-+    tcg_gen_andc_tl(s->T0, s->T0, s->T1);
- 
-     gen_op_update1_cc(s);
-     set_cc_op(s, CC_OP_LOGICB + ot);
-diff --git a/tests/tcg/i386/test-i386-bmi2.c b/tests/tcg/i386/test-i386-bmi2.c
-index 3c3ef85513..982d4abda4 100644
---- a/tests/tcg/i386/test-i386-bmi2.c
-+++ b/tests/tcg/i386/test-i386-bmi2.c
-@@ -99,6 +99,9 @@ int main(int argc, char *argv[]) {
-     result = bextrq(mask, 0x10f8);
-     assert(result == 0);
- 
-+    result = bextrq(0xfedcba9876543210ull, 0x7f00);
-+    assert(result == 0xfedcba9876543210ull);
-+
-     result = blsiq(0x30);
-     assert(result == 0x10);
- 
-@@ -164,6 +167,15 @@ int main(int argc, char *argv[]) {
-     result = bextrl(mask, 0x1038);
-     assert(result == 0);
- 
-+    result = bextrl((reg_t)0x8f635a775ad3b9b4ull, 0x3018);
-+    assert(result == 0x5a);
-+
-+    result = bextrl((reg_t)0xfedcba9876543210ull, 0x7f00);
-+    assert(result == 0x76543210u);
-+
-+    result = bextrl(-1, 0);
-+    assert(result == 0);
-+
-     result = blsil(0xffff);
-     assert(result == 1);
- 

debian/patches/extra/0026-target-i386-Fix-C-flag-for-BLSI-BLSMSK-BLSR.patch

@@ -1,47 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Richard Henderson <richard.henderson@linaro.org>
-Date: Sat, 14 Jan 2023 08:06:01 -1000
-Subject: [PATCH] target/i386: Fix C flag for BLSI, BLSMSK, BLSR
-
-We forgot to set cc_src, which is used for computing C.
-
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1370
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Message-Id: <20230114180601.2993644-1-richard.henderson@linaro.org>
-Cc: qemu-stable@nongnu.org
-Fixes: 1d0b926150e5 ("target/i386: move scalar 0F 38 and 0F 3A instruction to new decoder", 2022-10-18)
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit 99282098dc74c2055bde5652bde6cf0067d0c370)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- target/i386/tcg/emit.c.inc | 3 +++
- 1 file changed, 3 insertions(+)
-
-diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc
-index 99f6ba6e19..4d7702c106 100644
---- a/target/i386/tcg/emit.c.inc
-+++ b/target/i386/tcg/emit.c.inc
-@@ -1111,6 +1111,7 @@ static void gen_BLSI(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
- {
-     MemOp ot = decode->op[0].ot;
- 
-+    tcg_gen_mov_tl(cpu_cc_src, s->T0);
-     tcg_gen_neg_tl(s->T1, s->T0);
-     tcg_gen_and_tl(s->T0, s->T0, s->T1);
-     tcg_gen_mov_tl(cpu_cc_dst, s->T0);
-@@ -1121,6 +1122,7 @@ static void gen_BLSMSK(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode
- {
-     MemOp ot = decode->op[0].ot;
- 
-+    tcg_gen_mov_tl(cpu_cc_src, s->T0);
-     tcg_gen_subi_tl(s->T1, s->T0, 1);
-     tcg_gen_xor_tl(s->T0, s->T0, s->T1);
-     tcg_gen_mov_tl(cpu_cc_dst, s->T0);
-@@ -1131,6 +1133,7 @@ static void gen_BLSR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
- {
-     MemOp ot = decode->op[0].ot;
- 
-+    tcg_gen_mov_tl(cpu_cc_src, s->T0);
-     tcg_gen_subi_tl(s->T1, s->T0, 1);
-     tcg_gen_and_tl(s->T0, s->T0, s->T1);
-     tcg_gen_mov_tl(cpu_cc_dst, s->T0);

debian/patches/extra/0027-target-i386-fix-ADOX-followed-by-ADCX.patch

@@ -1,192 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Paolo Bonzini <pbonzini@redhat.com>
-Date: Tue, 31 Jan 2023 09:48:03 +0100
-Subject: [PATCH] target/i386: fix ADOX followed by ADCX
-
-When ADCX is followed by ADOX or vice versa, the second instruction's
-carry comes from EFLAGS and the condition codes use the CC_OP_ADCOX
-operation.  Retrieving the carry from EFLAGS is handled by this bit
-of gen_ADCOX:
-
-        tcg_gen_extract_tl(carry_in, cpu_cc_src,
-            ctz32(cc_op == CC_OP_ADCX ? CC_C : CC_O), 1);
-
-Unfortunately, in this case cc_op has been overwritten by the previous
-"if" statement to CC_OP_ADCOX.  This works by chance when the first
-instruction is ADCX; however, if the first instruction is ADOX,
-ADCX will incorrectly take its carry from OF instead of CF.
-
-Fix by moving the computation of the new cc_op at the end of the function.
-The included exhaustive test case fails without this patch and passes
-afterwards.
-
-Because ADCX/ADOX need not be invoked through the VEX prefix, this
-regression bisects to commit 16fc5726a6e2 ("target/i386: reimplement
-0x0f 0x38, add AVX", 2022-10-18).  However, the mistake happened a
-little earlier, when BMI instructions were rewritten using the new
-decoder framework.
-
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1471
-Reported-by: Paul Jolly <https://gitlab.com/myitcv>
-Fixes: 1d0b926150e5 ("target/i386: move scalar 0F 38 and 0F 3A instruction to new decoder", 2022-10-18)
-Cc: qemu-stable@nongnu.org
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit 60c7dd22e1383754d5f150bc9f7c2785c662a7b6)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- target/i386/tcg/emit.c.inc       | 20 +++++----
- tests/tcg/i386/Makefile.target   |  6 ++-
- tests/tcg/i386/test-i386-adcox.c | 75 ++++++++++++++++++++++++++++++++
- 3 files changed, 91 insertions(+), 10 deletions(-)
- create mode 100644 tests/tcg/i386/test-i386-adcox.c
-
-diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc
-index 4d7702c106..0d7c6e80ae 100644
---- a/target/i386/tcg/emit.c.inc
-+++ b/target/i386/tcg/emit.c.inc
-@@ -1015,6 +1015,7 @@ VSIB_AVX(VPGATHERQ, vpgatherq)
- 
- static void gen_ADCOX(DisasContext *s, CPUX86State *env, MemOp ot, int cc_op)
- {
-+    int opposite_cc_op;
-     TCGv carry_in = NULL;
-     TCGv carry_out = (cc_op == CC_OP_ADCX ? cpu_cc_dst : cpu_cc_src2);
-     TCGv zero;
-@@ -1022,14 +1023,8 @@ static void gen_ADCOX(DisasContext *s, CPUX86State *env, MemOp ot, int cc_op)
-     if (cc_op == s->cc_op || s->cc_op == CC_OP_ADCOX) {
-         /* Re-use the carry-out from a previous round.  */
-         carry_in = carry_out;
--        cc_op = s->cc_op;
--    } else if (s->cc_op == CC_OP_ADCX || s->cc_op == CC_OP_ADOX) {
--        /* Merge with the carry-out from the opposite instruction.  */
--        cc_op = CC_OP_ADCOX;
--    }
--
--    /* If we don't have a carry-in, get it out of EFLAGS.  */
--    if (!carry_in) {
-+    } else {
-+        /* We don't have a carry-in, get it out of EFLAGS.  */
-         if (s->cc_op != CC_OP_ADCX && s->cc_op != CC_OP_ADOX) {
-             gen_compute_eflags(s);
-         }
-@@ -1053,7 +1048,14 @@ static void gen_ADCOX(DisasContext *s, CPUX86State *env, MemOp ot, int cc_op)
-         tcg_gen_add2_tl(s->T0, carry_out, s->T0, carry_out, s->T1, zero);
-         break;
-     }
--    set_cc_op(s, cc_op);
-+
-+    opposite_cc_op = cc_op == CC_OP_ADCX ? CC_OP_ADOX : CC_OP_ADCX;
-+    if (s->cc_op == CC_OP_ADCOX || s->cc_op == opposite_cc_op) {
-+        /* Merge with the carry-out from the opposite instruction.  */
-+        set_cc_op(s, CC_OP_ADCOX);
-+    } else {
-+        set_cc_op(s, cc_op);
-+    }
- }
- 
- static void gen_ADCX(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
-diff --git a/tests/tcg/i386/Makefile.target b/tests/tcg/i386/Makefile.target
-index 81831cafbc..bafd8c2180 100644
---- a/tests/tcg/i386/Makefile.target
-+++ b/tests/tcg/i386/Makefile.target
-@@ -14,7 +14,7 @@ config-cc.mak: Makefile
- I386_SRCS=$(notdir $(wildcard $(I386_SRC)/*.c))
- ALL_X86_TESTS=$(I386_SRCS:.c=)
- SKIP_I386_TESTS=test-i386-ssse3 test-avx test-3dnow test-mmx
--X86_64_TESTS:=$(filter test-i386-bmi2 $(SKIP_I386_TESTS), $(ALL_X86_TESTS))
-+X86_64_TESTS:=$(filter test-i386-adcox test-i386-bmi2 $(SKIP_I386_TESTS), $(ALL_X86_TESTS))
- 
- test-i386-sse-exceptions: CFLAGS += -msse4.1 -mfpmath=sse
- run-test-i386-sse-exceptions: QEMU_OPTS += -cpu max
-@@ -28,6 +28,10 @@ test-i386-bmi2: CFLAGS=-O2
- run-test-i386-bmi2: QEMU_OPTS += -cpu max
- run-plugin-test-i386-bmi2-%: QEMU_OPTS += -cpu max
- 
-+test-i386-adcox: CFLAGS=-O2
-+run-test-i386-adcox: QEMU_OPTS += -cpu max
-+run-plugin-test-i386-adcox-%: QEMU_OPTS += -cpu max
-+
- #
- # hello-i386 is a barebones app
- #
-diff --git a/tests/tcg/i386/test-i386-adcox.c b/tests/tcg/i386/test-i386-adcox.c
-new file mode 100644
-index 0000000000..16169efff8
---- /dev/null
-+++ b/tests/tcg/i386/test-i386-adcox.c
-@@ -0,0 +1,75 @@
-+/* See if various BMI2 instructions give expected results */
-+#include <assert.h>
-+#include <stdint.h>
-+#include <stdio.h>
-+
-+#define CC_C 1
-+#define CC_O (1 << 11)
-+
-+#ifdef __x86_64__
-+#define REG uint64_t
-+#else
-+#define REG uint32_t
-+#endif
-+
-+void test_adox_adcx(uint32_t in_c, uint32_t in_o, REG adcx_operand, REG adox_operand)
-+{
-+    REG flags;
-+    REG out_adcx, out_adox;
-+
-+    asm("pushf; pop %0" : "=r"(flags));
-+    flags &= ~(CC_C | CC_O);
-+    flags |= (in_c ? CC_C : 0);
-+    flags |= (in_o ? CC_O : 0);
-+
-+    out_adcx = adcx_operand;
-+    out_adox = adox_operand;
-+    asm("push %0; popf;"
-+        "adox %3, %2;"
-+        "adcx %3, %1;"
-+        "pushf; pop %0"
-+        : "+r" (flags), "+r" (out_adcx), "+r" (out_adox)
-+        : "r" ((REG)-1), "0" (flags), "1" (out_adcx), "2" (out_adox));
-+
-+    assert(out_adcx == in_c + adcx_operand - 1);
-+    assert(out_adox == in_o + adox_operand - 1);
-+    assert(!!(flags & CC_C) == (in_c || adcx_operand));
-+    assert(!!(flags & CC_O) == (in_o || adox_operand));
-+}
-+
-+void test_adcx_adox(uint32_t in_c, uint32_t in_o, REG adcx_operand, REG adox_operand)
-+{
-+    REG flags;
-+    REG out_adcx, out_adox;
-+
-+    asm("pushf; pop %0" : "=r"(flags));
-+    flags &= ~(CC_C | CC_O);
-+    flags |= (in_c ? CC_C : 0);
-+    flags |= (in_o ? CC_O : 0);
-+
-+    out_adcx = adcx_operand;
-+    out_adox = adox_operand;
-+    asm("push %0; popf;"
-+        "adcx %3, %1;"
-+        "adox %3, %2;"
-+        "pushf; pop %0"
-+        : "+r" (flags), "+r" (out_adcx), "+r" (out_adox)
-+        : "r" ((REG)-1), "0" (flags), "1" (out_adcx), "2" (out_adox));
-+
-+    assert(out_adcx == in_c + adcx_operand - 1);
-+    assert(out_adox == in_o + adox_operand - 1);
-+    assert(!!(flags & CC_C) == (in_c || adcx_operand));
-+    assert(!!(flags & CC_O) == (in_o || adox_operand));
-+}
-+
-+int main(int argc, char *argv[]) {
-+    /* try all combinations of input CF, input OF, CF from op1+op2,  OF from op2+op1 */
-+    int i;
-+    for (i = 0; i <= 15; i++) {
-+        printf("%d\n", i);
-+        test_adcx_adox(!!(i & 1), !!(i & 2), !!(i & 4), !!(i & 8));
-+        test_adox_adcx(!!(i & 1), !!(i & 2), !!(i & 4), !!(i & 8));
-+    }
-+    return 0;
-+}
-+

debian/patches/extra/0028-target-i386-Fix-BZHI-instruction.patch

@@ -1,64 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Richard Henderson <richard.henderson@linaro.org>
-Date: Sat, 14 Jan 2023 13:32:06 -1000
-Subject: [PATCH] target/i386: Fix BZHI instruction
-
-We did not correctly handle N >= operand size.
-
-Resolves: https://gitlab.com/qemu-project/qemu/-/issues/1374
-Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
-Message-Id: <20230114233206.3118472-1-richard.henderson@linaro.org>
-Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-(cherry-picked from commit 9ad2ba6e8e7fc195d0dd0b76ab38bd2fceb1bdd4)
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- target/i386/tcg/emit.c.inc      | 14 +++++++-------
- tests/tcg/i386/test-i386-bmi2.c |  3 +++
- 2 files changed, 10 insertions(+), 7 deletions(-)
-
-diff --git a/target/i386/tcg/emit.c.inc b/target/i386/tcg/emit.c.inc
-index 0d7c6e80ae..7296f3952c 100644
---- a/target/i386/tcg/emit.c.inc
-+++ b/target/i386/tcg/emit.c.inc
-@@ -1145,20 +1145,20 @@ static void gen_BLSR(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
- static void gen_BZHI(DisasContext *s, CPUX86State *env, X86DecodedInsn *decode)
- {
-     MemOp ot = decode->op[0].ot;
--    TCGv bound;
-+    TCGv bound = tcg_constant_tl(ot == MO_64 ? 63 : 31);
-+    TCGv zero = tcg_constant_tl(0);
-+    TCGv mone = tcg_constant_tl(-1);
- 
--    tcg_gen_ext8u_tl(s->T1, cpu_regs[s->vex_v]);
--    bound = tcg_constant_tl(ot == MO_64 ? 63 : 31);
-+    tcg_gen_ext8u_tl(s->T1, s->T1);
- 
-     /*
-      * Note that since we're using BMILG (in order to get O
-      * cleared) we need to store the inverse into C.
-      */
--    tcg_gen_setcond_tl(TCG_COND_LT, cpu_cc_src, s->T1, bound);
--    tcg_gen_movcond_tl(TCG_COND_GT, s->T1, s->T1, bound, bound, s->T1);
-+    tcg_gen_setcond_tl(TCG_COND_LEU, cpu_cc_src, s->T1, bound);
- 
--    tcg_gen_movi_tl(s->A0, -1);
--    tcg_gen_shl_tl(s->A0, s->A0, s->T1);
-+    tcg_gen_shl_tl(s->A0, mone, s->T1);
-+    tcg_gen_movcond_tl(TCG_COND_LEU, s->A0, s->T1, bound, s->A0, zero);
-     tcg_gen_andc_tl(s->T0, s->T0, s->A0);
- 
-     gen_op_update1_cc(s);
-diff --git a/tests/tcg/i386/test-i386-bmi2.c b/tests/tcg/i386/test-i386-bmi2.c
-index 982d4abda4..0244df7987 100644
---- a/tests/tcg/i386/test-i386-bmi2.c
-+++ b/tests/tcg/i386/test-i386-bmi2.c
-@@ -123,6 +123,9 @@ int main(int argc, char *argv[]) {
-     result = bzhiq(mask, 0x1f);
-     assert(result == (mask & ~(-1 << 30)));
- 
-+    result = bzhiq(mask, 0x40);
-+    assert(result == mask);
-+
-     result = rorxq(0x2132435465768798, 8);
-     assert(result == 0x9821324354657687);
- 

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 b9647c5ffc..9a16d86344 100644
+index 56d1972d15..cfa0b832ba 100644
 --- a/block/file-posix.c
 +++ b/block/file-posix.c
-@@ -552,7 +552,7 @@ static QemuOptsList raw_runtime_opts = {
+@@ -565,7 +565,7 @@ static QemuOptsList raw_runtime_opts = {
          {
              .name = "locking",
              .type = QEMU_OPT_STRING,
@@ -26,7 +26,7 @@ index b9647c5ffc..9a16d86344 100644
          },
          {
              .name = "pr-manager",
-@@ -652,7 +652,7 @@ static int raw_open_common(BlockDriverState *bs, QDict *options,
+@@ -665,7 +665,7 @@ 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

@@ -9,12 +9,12 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 2 insertions(+), 2 deletions(-)
 
 diff --git a/include/net/net.h b/include/net/net.h
-index dc20b31e9f..5ae04a8693 100644
+index cdd5b109b0..653a37e9d1 100644
 --- a/include/net/net.h
 +++ b/include/net/net.h
-@@ -236,8 +236,8 @@ void netdev_add(QemuOpts *opts, Error **errp);
+@@ -305,8 +305,8 @@ 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 CONFIG_SYSCONFDIR "/qemu-ifup"
 -#define DEFAULT_NETWORK_DOWN_SCRIPT CONFIG_SYSCONFDIR "/qemu-ifdown"

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 d4bc19577a..be7da64f38 100644
+index 76f24446a5..2a47d79b49 100644
 --- a/target/i386/cpu.h
 +++ b/target/i386/cpu.h
-@@ -2174,9 +2174,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
+@@ -2556,9 +2556,9 @@ uint64_t cpu_get_tsc(CPUX86State *env);
  #define CPU_RESOLVING_TYPE TYPE_X86_CPU
  
  #ifdef TARGET_X86_64

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 c3ac20ad43..37774f1c0a 100644
+index 0326c63bec..d523d00200 100644
 --- a/ui/spice-core.c
 +++ b/ui/spice-core.c
-@@ -689,32 +689,35 @@ static void qemu_spice_init(void)
+@@ -690,32 +690,35 @@ 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 7c90f7ba4b..2e03102f00 100644
+index c6d25ae733..ccca125c3a 100644
 --- a/block/gluster.c
 +++ b/block/gluster.c
 @@ -42,7 +42,7 @@
@@ -21,15 +21,15 @@ index 7c90f7ba4b..2e03102f00 100644
  /*
   * Several versions of GlusterFS (3.12? -> 6.0.1) fail when the transfer size
   * is greater or equal to 1024 MiB, so we are limiting the transfer size to 512
-@@ -424,6 +424,7 @@ static struct glfs *qemu_gluster_glfs_init(BlockdevOptionsGluster *gconf,
+@@ -421,6 +421,7 @@ static struct glfs *qemu_gluster_glfs_init(BlockdevOptionsGluster *gconf,
      int old_errno;
      SocketAddressList *server;
-     unsigned long long port;
+     uint64_t port;
 +    const char *logfile;
  
      glfs = glfs_find_preopened(gconf->volume);
      if (glfs) {
-@@ -466,9 +467,15 @@ static struct glfs *qemu_gluster_glfs_init(BlockdevOptionsGluster *gconf,
+@@ -463,9 +464,15 @@ static struct glfs *qemu_gluster_glfs_init(BlockdevOptionsGluster *gconf,
          }
      }
  

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 f826410f40..64a8d7d48b 100644
+index af984fb7db..bf143fac00 100644
 --- a/block/rbd.c
 +++ b/block/rbd.c
-@@ -820,6 +820,8 @@ static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
+@@ -963,6 +963,8 @@ 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-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 2e03102f00..7886c5fe8c 100644
+index ccca125c3a..301a653ea7 100644
 --- a/block/gluster.c
 +++ b/block/gluster.c
 @@ -57,6 +57,7 @@ typedef struct GlusterAIOCB {
@@ -27,7 +27,7 @@ index 2e03102f00..7886c5fe8c 100644
  } GlusterAIOCB;
  
  typedef struct BDRVGlusterState {
-@@ -752,8 +753,10 @@ static void gluster_finish_aiocb(struct glfs_fd *fd, ssize_t ret,
+@@ -746,8 +747,10 @@ 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 */
@@ -39,7 +39,7 @@ index 2e03102f00..7886c5fe8c 100644
      }
  
      aio_co_schedule(acb->aio_context, acb->coroutine);
-@@ -1022,6 +1025,7 @@ static coroutine_fn int qemu_gluster_co_pwrite_zeroes(BlockDriverState *bs,
+@@ -1018,6 +1021,7 @@ 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);
@@ -47,7 +47,7 @@ index 2e03102f00..7886c5fe8c 100644
  
      ret = glfs_zerofill_async(s->fd, offset, bytes, gluster_finish_aiocb, &acb);
      if (ret < 0) {
-@@ -1203,9 +1207,11 @@ static coroutine_fn int qemu_gluster_co_rw(BlockDriverState *bs,
+@@ -1198,9 +1202,11 @@ 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 2e03102f00..7886c5fe8c 100644
          ret = glfs_preadv_async(s->fd, qiov->iov, qiov->niov, offset, 0,
                                  gluster_finish_aiocb, &acb);
      }
-@@ -1268,6 +1274,7 @@ static coroutine_fn int qemu_gluster_co_flush_to_disk(BlockDriverState *bs)
+@@ -1263,6 +1269,7 @@ 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);
@@ -67,7 +67,7 @@ index 2e03102f00..7886c5fe8c 100644
  
      ret = glfs_fsync_async(s->fd, gluster_finish_aiocb, &acb);
      if (ret < 0) {
-@@ -1316,6 +1323,7 @@ static coroutine_fn int qemu_gluster_co_pdiscard(BlockDriverState *bs,
+@@ -1311,6 +1318,7 @@ 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);

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

@@ -1,98 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Wolfgang Bumiller <w.bumiller@proxmox.com>
-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 |  2 ++
- 3 files changed, 44 insertions(+)
-
-diff --git a/net/net.c b/net/net.c
-index 840ad9dca5..28e97c5d85 100644
---- a/net/net.c
-+++ b/net/net.c
-@@ -1372,6 +1372,33 @@ void hmp_info_network(Monitor *mon, const QDict *qdict)
-     }
- }
- 
-+int64_t qmp_get_link_status(const char *name, Error **errp)
-+{
-+    NetClientState *ncs[MAX_QUEUE_NUM];
-+    NetClientState *nc;
-+    int queues;
-+    bool ret;
-+
-+    queues = qemu_find_net_clients_except(name, ncs,
-+                                          NET_CLIENT_DRIVER__MAX,
-+                                          MAX_QUEUE_NUM);
-+
-+    if (queues == 0) {
-+        error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
-+                  "Device '%s' not found", name);
-+        return (int64_t) -1;
-+    }
-+
-+    nc = ncs[0];
-+    ret = ncs[0]->link_down;
-+
-+    if (nc->peer->info->type == NET_CLIENT_DRIVER_NIC) {
-+      ret = ncs[0]->peer->link_down;
-+    }
-+
-+    return (int64_t) ret ? 0 : 1;
-+}
-+
- void colo_notify_filters_event(int event, Error **errp)
- {
-     NetClientState *nc;
-diff --git a/qapi/net.json b/qapi/net.json
-index 522ac582ed..327d7c5a37 100644
---- a/qapi/net.json
-+++ b/qapi/net.json
-@@ -36,6 +36,21 @@
- ##
- { 'command': 'set_link', 'data': {'name': 'str', 'up': 'bool'} }
- 
-+##
-+# @get_link_status:
-+#
-+# Get the current link state of the nics or nic.
-+#
-+# @name: name of the nic you get the state of
-+#
-+# Return: If link is up 1
-+#         If link is down 0
-+#         If an error occure an empty string.
-+#
-+# Notes: this is an Proxmox VE extension and not offical part of Qemu.
-+##
-+{ 'command': 'get_link_status', 'data': {'name': 'str'} , 'returns': 'int' }
-+
- ##
- # @netdev_add:
- #
-diff --git a/qapi/pragma.json b/qapi/pragma.json
-index 7f810b0e97..29233db825 100644
---- a/qapi/pragma.json
-+++ b/qapi/pragma.json
-@@ -15,6 +15,7 @@
-         'device_add',
-         'device_del',
-         'expire_password',
-+        'get_link_status',
-         'migrate_cancel',
-         'netdev_add',
-         'netdev_del',
-@@ -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-tpm-models',

debian/patches/pve/0008-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 a9b3a8103c..0bc9f1af59 100644
+index 2044c22a4c..4c8b5412c6 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -3013,7 +3013,8 @@ static int img_info(int argc, char **argv)
+@@ -3075,7 +3075,8 @@ 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/0009-PVE-Up-qemu-img-dd-add-osize-and-read-from-to-stdin-.patch

@@ -38,10 +38,10 @@ Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
  2 files changed, 133 insertions(+), 73 deletions(-)
 
 diff --git a/qemu-img-cmds.hx b/qemu-img-cmds.hx
-index 1b1dab5b17..d1616c045a 100644
+index c9dd70a892..048788b23d 100644
 --- a/qemu-img-cmds.hx
 +++ b/qemu-img-cmds.hx
-@@ -58,9 +58,9 @@ SRST
+@@ -60,9 +60,9 @@ SRST
  ERST
  
  DEF("dd", img_dd,
@@ -54,10 +54,10 @@ index 1b1dab5b17..d1616c045a 100644
  
  DEF("info", img_info,
 diff --git a/qemu-img.c b/qemu-img.c
-index 0bc9f1af59..221b9d6a16 100644
+index 4c8b5412c6..d5fa89a204 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -4829,10 +4829,12 @@ static int img_bitmap(int argc, char **argv)
+@@ -4997,10 +4997,12 @@ static int img_bitmap(int argc, char **argv)
  #define C_IF      04
  #define C_OF      010
  #define C_SKIP    020
@@ -70,7 +70,7 @@ index 0bc9f1af59..221b9d6a16 100644
  };
  
  struct DdIo {
-@@ -4908,6 +4910,19 @@ static int img_dd_skip(const char *arg,
+@@ -5076,6 +5078,19 @@ static int img_dd_skip(const char *arg,
      return 0;
  }
  
@@ -90,7 +90,7 @@ index 0bc9f1af59..221b9d6a16 100644
  static int img_dd(int argc, char **argv)
  {
      int ret = 0;
-@@ -4948,6 +4963,7 @@ static int img_dd(int argc, char **argv)
+@@ -5116,6 +5131,7 @@ 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 },
@@ -98,7 +98,7 @@ index 0bc9f1af59..221b9d6a16 100644
          { NULL, NULL, 0 }
      };
      const struct option long_options[] = {
-@@ -5023,91 +5039,112 @@ static int img_dd(int argc, char **argv)
+@@ -5191,91 +5207,112 @@ static int img_dd(int argc, char **argv)
          arg = NULL;
      }
  
@@ -275,7 +275,7 @@ index 0bc9f1af59..221b9d6a16 100644
      }
  
      if (dd.flags & C_SKIP && (in.offset > INT64_MAX / in.bsz ||
-@@ -5124,20 +5161,43 @@ static int img_dd(int argc, char **argv)
+@@ -5292,20 +5329,43 @@ static int img_dd(int argc, char **argv)
      in.buf = g_new(uint8_t, in.bsz);
  
      for (out_pos = 0; in_pos < size; ) {

debian/patches/pve/0010-PVE-Up-qemu-img-dd-add-isize-parameter.patch

@@ -16,10 +16,10 @@ Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
  1 file changed, 25 insertions(+), 3 deletions(-)
 
 diff --git a/qemu-img.c b/qemu-img.c
-index 221b9d6a16..c1306385a8 100644
+index d5fa89a204..d458e85af2 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -4830,11 +4830,13 @@ static int img_bitmap(int argc, char **argv)
+@@ -4998,11 +4998,13 @@ static int img_bitmap(int argc, char **argv)
  #define C_OF      010
  #define C_SKIP    020
  #define C_OSIZE   040
@@ -33,7 +33,7 @@ index 221b9d6a16..c1306385a8 100644
  };
  
  struct DdIo {
-@@ -4923,6 +4925,19 @@ static int img_dd_osize(const char *arg,
+@@ -5091,6 +5093,19 @@ static int img_dd_osize(const char *arg,
      return 0;
  }
  
@@ -53,7 +53,7 @@ index 221b9d6a16..c1306385a8 100644
  static int img_dd(int argc, char **argv)
  {
      int ret = 0;
-@@ -4937,12 +4952,14 @@ static int img_dd(int argc, char **argv)
+@@ -5105,12 +5120,14 @@ static int img_dd(int argc, char **argv)
      int c, i;
      const char *out_fmt = "raw";
      const char *fmt = NULL;
@@ -69,7 +69,7 @@ index 221b9d6a16..c1306385a8 100644
      };
      struct DdIo in = {
          .bsz = 512, /* Block size is by default 512 bytes */
-@@ -4964,6 +4981,7 @@ static int img_dd(int argc, char **argv)
+@@ -5132,6 +5149,7 @@ 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 },
@@ -77,7 +77,7 @@ index 221b9d6a16..c1306385a8 100644
          { NULL, NULL, 0 }
      };
      const struct option long_options[] = {
-@@ -5160,9 +5178,10 @@ static int img_dd(int argc, char **argv)
+@@ -5328,9 +5346,10 @@ static int img_dd(int argc, char **argv)
  
      in.buf = g_new(uint8_t, in.bsz);
  
@@ -90,7 +90,7 @@ index 221b9d6a16..c1306385a8 100644
          if (blk1) {
              in_ret = blk_pread(blk1, in_pos, bytes, in.buf, 0);
              if (in_ret == 0) {
-@@ -5171,6 +5190,9 @@ static int img_dd(int argc, char **argv)
+@@ -5339,6 +5358,9 @@ static int img_dd(int argc, char **argv)
          } else {
              in_ret = read(STDIN_FILENO, in.buf, bytes);
              if (in_ret == 0) {

debian/patches/pve/0011-PVE-Up-qemu-img-dd-add-n-skip_create.patch

@@ -13,10 +13,10 @@ Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
  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
+index 3653adb963..d83e8fb3c0 100644
 --- a/docs/tools/qemu-img.rst
 +++ b/docs/tools/qemu-img.rst
-@@ -208,6 +208,10 @@ Parameters to convert subcommand:
+@@ -212,6 +212,10 @@ Parameters to convert subcommand:
  
  Parameters to dd subcommand:
  
@@ -27,7 +27,7 @@ index 15aeddc6d8..5e713e231d 100644
  .. program:: qemu-img-dd
  
  .. option:: bs=BLOCK_SIZE
-@@ -488,7 +492,7 @@ Command description:
+@@ -492,7 +496,7 @@ Command description:
    it doesn't need to be specified separately in this case.
  
  
@@ -36,7 +36,7 @@ index 15aeddc6d8..5e713e231d 100644
  
    dd copies from *INPUT* file to *OUTPUT* file converting it from
    *FMT* format to *OUTPUT_FMT* format.
-@@ -499,6 +503,11 @@ Command description:
+@@ -503,6 +507,11 @@ Command description:
  
    The size syntax is similar to :manpage:`dd(1)`'s size syntax.
  
@@ -49,10 +49,10 @@ index 15aeddc6d8..5e713e231d 100644
  
    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
+index 048788b23d..0b29a67a06 100644
 --- a/qemu-img-cmds.hx
 +++ b/qemu-img-cmds.hx
-@@ -58,9 +58,9 @@ SRST
+@@ -60,9 +60,9 @@ SRST
  ERST
  
  DEF("dd", img_dd,
@@ -65,10 +65,10 @@ index d1616c045a..b5b0bb4467 100644
  
  DEF("info", img_info,
 diff --git a/qemu-img.c b/qemu-img.c
-index c1306385a8..59c403373b 100644
+index d458e85af2..dc13efba8b 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -4954,7 +4954,7 @@ static int img_dd(int argc, char **argv)
+@@ -5122,7 +5122,7 @@ static int img_dd(int argc, char **argv)
      const char *fmt = NULL;
      int64_t size = 0, readsize = 0;
      int64_t out_pos, in_pos;
@@ -77,7 +77,7 @@ index c1306385a8..59c403373b 100644
      struct DdInfo dd = {
          .flags = 0,
          .count = 0,
-@@ -4992,7 +4992,7 @@ static int img_dd(int argc, char **argv)
+@@ -5160,7 +5160,7 @@ static int img_dd(int argc, char **argv)
          { 0, 0, 0, 0 }
      };
  
@@ -86,7 +86,7 @@ index c1306385a8..59c403373b 100644
          if (c == EOF) {
              break;
          }
-@@ -5012,6 +5012,9 @@ static int img_dd(int argc, char **argv)
+@@ -5180,6 +5180,9 @@ static int img_dd(int argc, char **argv)
          case 'h':
              help();
              break;
@@ -96,7 +96,7 @@ index c1306385a8..59c403373b 100644
          case 'U':
              force_share = true;
              break;
-@@ -5142,13 +5145,15 @@ static int img_dd(int argc, char **argv)
+@@ -5310,13 +5313,15 @@ static int img_dd(int argc, char **argv)
                                  size - in.bsz * in.offset, &error_abort);
          }
  

debian/patches/pve/0012-qemu-img-dd-add-l-option-for-loading-a-snapshot.patch

@@ -1,9 +1,9 @@
 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
+Subject: [PATCH] qemu-img dd: add -l option for loading a snapshot
 
-Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  docs/tools/qemu-img.rst |  6 +++---
@@ -12,10 +12,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  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
+index d83e8fb3c0..61c6b21859 100644
 --- a/docs/tools/qemu-img.rst
 +++ b/docs/tools/qemu-img.rst
-@@ -492,10 +492,10 @@ Command description:
+@@ -496,10 +496,10 @@ Command description:
    it doesn't need to be specified separately in this case.
  
  
@@ -30,10 +30,10 @@ index 5e713e231d..9390d5e5cf 100644
    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
+index 0b29a67a06..758f397232 100644
 --- a/qemu-img-cmds.hx
 +++ b/qemu-img-cmds.hx
-@@ -58,9 +58,9 @@ SRST
+@@ -60,9 +60,9 @@ SRST
  ERST
  
  DEF("dd", img_dd,
@@ -46,10 +46,10 @@ index b5b0bb4467..36f97e1f19 100644
  
  DEF("info", img_info,
 diff --git a/qemu-img.c b/qemu-img.c
-index 59c403373b..065a54cc42 100644
+index dc13efba8b..02f2e0aa45 100644
 --- a/qemu-img.c
 +++ b/qemu-img.c
-@@ -4946,6 +4946,7 @@ static int img_dd(int argc, char **argv)
+@@ -5114,6 +5114,7 @@ static int img_dd(int argc, char **argv)
      BlockDriver *drv = NULL, *proto_drv = NULL;
      BlockBackend *blk1 = NULL, *blk2 = NULL;
      QemuOpts *opts = NULL;
@@ -57,7 +57,7 @@ index 59c403373b..065a54cc42 100644
      QemuOptsList *create_opts = NULL;
      Error *local_err = NULL;
      bool image_opts = false;
-@@ -4955,6 +4956,7 @@ static int img_dd(int argc, char **argv)
+@@ -5123,6 +5124,7 @@ 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;
@@ -65,7 +65,7 @@ index 59c403373b..065a54cc42 100644
      struct DdInfo dd = {
          .flags = 0,
          .count = 0,
-@@ -4992,7 +4994,7 @@ static int img_dd(int argc, char **argv)
+@@ -5160,7 +5162,7 @@ static int img_dd(int argc, char **argv)
          { 0, 0, 0, 0 }
      };
  
@@ -74,7 +74,7 @@ index 59c403373b..065a54cc42 100644
          if (c == EOF) {
              break;
          }
-@@ -5015,6 +5017,19 @@ static int img_dd(int argc, char **argv)
+@@ -5183,6 +5185,19 @@ static int img_dd(int argc, char **argv)
          case 'n':
              skip_create = true;
              break;
@@ -94,7 +94,7 @@ index 59c403373b..065a54cc42 100644
          case 'U':
              force_share = true;
              break;
-@@ -5074,11 +5089,24 @@ static int img_dd(int argc, char **argv)
+@@ -5242,11 +5257,24 @@ 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);
@@ -120,7 +120,7 @@ index 59c403373b..065a54cc42 100644
      }
  
      if (dd.flags & C_OSIZE) {
-@@ -5233,6 +5261,7 @@ static int img_dd(int argc, char **argv)
+@@ -5401,6 +5429,7 @@ static int img_dd(int argc, char **argv)
  out:
      g_free(arg);
      qemu_opts_del(opts);

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

@@ -7,20 +7,62 @@ 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]
+[FE: add BalloonInfo to member name exceptions list
+     rebase for 8.0 - moved to hw/core/machine-hmp-cmds.c]
 Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
+ hw/core/machine-hmp-cmds.c | 30 +++++++++++++++++++++++++++++-
  hw/virtio/virtio-balloon.c | 33 +++++++++++++++++++++++++++++++--
- monitor/hmp-cmds.c         | 30 +++++++++++++++++++++++++++++-
  qapi/machine.json          | 22 +++++++++++++++++++++-
  qapi/pragma.json           |  1 +
  4 files changed, 82 insertions(+), 4 deletions(-)
 
+diff --git a/hw/core/machine-hmp-cmds.c b/hw/core/machine-hmp-cmds.c
+index c6325cdcaa..7f817d622d 100644
+--- a/hw/core/machine-hmp-cmds.c
++++ b/hw/core/machine-hmp-cmds.c
+@@ -179,7 +179,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
+         return;
+     }
+ 
+-    monitor_printf(mon, "balloon: actual=%" PRId64 "\n", info->actual >> 20);
++    monitor_printf(mon, "balloon: actual=%" PRId64, info->actual >> 20);
++    monitor_printf(mon, " max_mem=%" PRId64, info->max_mem >> 20);
++    if (info->has_total_mem) {
++        monitor_printf(mon, " total_mem=%" PRId64, info->total_mem >> 20);
++    }
++    if (info->has_free_mem) {
++        monitor_printf(mon, " free_mem=%" PRId64, info->free_mem >> 20);
++    }
++
++    if (info->has_mem_swapped_in) {
++        monitor_printf(mon, " mem_swapped_in=%" PRId64, info->mem_swapped_in);
++    }
++    if (info->has_mem_swapped_out) {
++        monitor_printf(mon, " mem_swapped_out=%" PRId64, info->mem_swapped_out);
++    }
++    if (info->has_major_page_faults) {
++        monitor_printf(mon, " major_page_faults=%" PRId64,
++                       info->major_page_faults);
++    }
++    if (info->has_minor_page_faults) {
++        monitor_printf(mon, " minor_page_faults=%" PRId64,
++                       info->minor_page_faults);
++    }
++    if (info->has_last_update) {
++        monitor_printf(mon, " last_update=%" PRId64,
++                       info->last_update);
++    }
++
++    monitor_printf(mon, "\n");
+ 
+     qapi_free_BalloonInfo(info);
+ }
 diff --git a/hw/virtio/virtio-balloon.c b/hw/virtio/virtio-balloon.c
-index 73ac5eb675..bbfe7eca62 100644
+index 2eb5a14fa2..aa2fd6c32f 100644
 --- a/hw/virtio/virtio-balloon.c
 +++ b/hw/virtio/virtio-balloon.c
-@@ -806,8 +806,37 @@ static uint64_t virtio_balloon_get_features(VirtIODevice *vdev, uint64_t f,
+@@ -795,8 +795,37 @@ static uint64_t virtio_balloon_get_features(VirtIODevice *vdev, uint64_t f,
  static void virtio_balloon_stat(void *opaque, BalloonInfo *info)
  {
      VirtIOBalloon *dev = opaque;
@@ -60,54 +102,13 @@ index 73ac5eb675..bbfe7eca62 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 01b789a79e..480b798963 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -696,7 +696,35 @@ void hmp_info_balloon(Monitor *mon, const QDict *qdict)
-         return;
-     }
- 
--    monitor_printf(mon, "balloon: actual=%" PRId64 "\n", info->actual >> 20);
-+    monitor_printf(mon, "balloon: actual=%" PRId64, info->actual >> 20);
-+    monitor_printf(mon, " max_mem=%" PRId64, info->max_mem >> 20);
-+    if (info->has_total_mem) {
-+        monitor_printf(mon, " total_mem=%" PRId64, info->total_mem >> 20);
-+    }
-+    if (info->has_free_mem) {
-+        monitor_printf(mon, " free_mem=%" PRId64, info->free_mem >> 20);
-+    }
-+
-+    if (info->has_mem_swapped_in) {
-+        monitor_printf(mon, " mem_swapped_in=%" PRId64, info->mem_swapped_in);
-+    }
-+    if (info->has_mem_swapped_out) {
-+        monitor_printf(mon, " mem_swapped_out=%" PRId64, info->mem_swapped_out);
-+    }
-+    if (info->has_major_page_faults) {
-+        monitor_printf(mon, " major_page_faults=%" PRId64,
-+                       info->major_page_faults);
-+    }
-+    if (info->has_minor_page_faults) {
-+        monitor_printf(mon, " minor_page_faults=%" PRId64,
-+                       info->minor_page_faults);
-+    }
-+    if (info->has_last_update) {
-+        monitor_printf(mon, " last_update=%" PRId64,
-+                       info->last_update);
-+    }
-+
-+    monitor_printf(mon, "\n");
- 
-     qapi_free_BalloonInfo(info);
- }
 diff --git a/qapi/machine.json b/qapi/machine.json
-index b9228a5e46..10e77a9af3 100644
+index a6b8795b09..9f7ed0eaa0 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
+@@ -1163,9 +1163,29 @@
+ # @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
 +#
@@ -137,10 +138,10 @@ index b9228a5e46..10e77a9af3 100644
  ##
  # @query-balloon:
 diff --git a/qapi/pragma.json b/qapi/pragma.json
-index 29233db825..f2097b9020 100644
+index 023a2ef7bc..6aaa9cb975 100644
 --- a/qapi/pragma.json
 +++ b/qapi/pragma.json
-@@ -37,6 +37,7 @@
+@@ -81,6 +81,7 @@
      'member-name-exceptions': [     # visible in:
          'ACPISlotType',             # query-acpi-ospm-status
          'AcpiTableOptions',         # -acpitable

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

@@ -13,36 +13,36 @@ 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 4f4ab30f8c..76fff60a6b 100644
+index 1bc21b84a4..93fb4bc24a 100644
 --- a/hw/core/machine-qmp-cmds.c
 +++ b/hw/core/machine-qmp-cmds.c
-@@ -99,6 +99,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
-         info->hotpluggable_cpus = mc->has_hotpluggable_cpus;
+@@ -91,6 +91,12 @@ MachineInfoList *qmp_query_machines(bool has_compat_props, bool compat_props,
          info->numa_mem_supported = mc->numa_mem_supported;
          info->deprecated = !!mc->deprecation_reason;
+         info->acpi = !!object_class_property_find(OBJECT_CLASS(mc), "acpi");
 +
 +        if (strcmp(mc->name, MACHINE_GET_CLASS(current_machine)->name) == 0) {
 +            info->has_is_current = true;
 +            info->is_current = true;
 +        }
 +
-         if (mc->default_cpu_type) {
-             info->default_cpu_type = g_strdup(mc->default_cpu_type);
-             info->has_default_cpu_type = true;
+         if (default_cpu_type) {
+             info->default_cpu_type = g_strdup(default_cpu_type);
+         }
 diff --git a/qapi/machine.json b/qapi/machine.json
-index 10e77a9af3..9156103c8f 100644
+index 9f7ed0eaa0..16366b774a 100644
 --- a/qapi/machine.json
 +++ b/qapi/machine.json
-@@ -138,6 +138,8 @@
+@@ -167,6 +167,8 @@
  #
  # @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)
+ #     (since 1.5)
  #
-@@ -159,7 +161,7 @@
+@@ -199,7 +201,7 @@
  ##
  { 'struct': 'MachineInfo',
    'data': { 'name': 'str', '*alias': 'str',
@@ -50,4 +50,4 @@ index 10e77a9af3..9156103c8f 100644
 +            '*is-default': 'bool', '*is-current': 'bool', 'cpu-max': 'int',
              'hotpluggable-cpus': 'bool',  'numa-mem-supported': 'bool',
              'deprecated': 'bool', '*default-cpu-type': 'str',
-             '*default-ram-id': 'str' } }
+             '*default-ram-id': 'str', 'acpi': 'bool',

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

@@ -6,16 +6,18 @@ Subject: [PATCH] PVE: qapi: modify spice query
 Provide the last ticket in the SpiceInfo struct optionally.
 
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: adapt to QAPI change]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  qapi/ui.json    | 3 +++
- ui/spice-core.c | 5 +++++
- 2 files changed, 8 insertions(+)
+ ui/spice-core.c | 4 ++++
+ 2 files changed, 7 insertions(+)
 
 diff --git a/qapi/ui.json b/qapi/ui.json
-index 0abba3e930..bf8f441227 100644
+index c536d4e524..c2df48959b 100644
 --- a/qapi/ui.json
 +++ b/qapi/ui.json
-@@ -310,11 +310,14 @@
+@@ -312,11 +312,14 @@
  #
  # @channels: a list of @SpiceChannel for each active spice channel
  #
@@ -31,15 +33,14 @@ index 0abba3e930..bf8f441227 100644
    'if': 'CONFIG_SPICE' }
  
 diff --git a/ui/spice-core.c b/ui/spice-core.c
-index 37774f1c0a..367f77f2b4 100644
+index d523d00200..c76c224706 100644
 --- a/ui/spice-core.c
 +++ b/ui/spice-core.c
-@@ -534,6 +534,11 @@ static SpiceInfo *qmp_query_spice_real(Error **errp)
+@@ -548,6 +548,10 @@ 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);
  
 +    if (auth_passwd) {
-+        info->has_ticket = true;
 +        info->ticket =  g_strdup(auth_passwd);
 +    }
 +

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

@@ -14,27 +14,28 @@ 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>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
- migration/channel-savevm-async.c | 182 +++++++++++++++++++++++++++++++
+ migration/channel-savevm-async.c | 184 +++++++++++++++++++++++++++++++
  migration/channel-savevm-async.h |  51 +++++++++
  migration/meson.build            |   1 +
- 3 files changed, 234 insertions(+)
+ 3 files changed, 236 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
+index 0000000000..e57ab2ae40
 --- /dev/null
 +++ b/migration/channel-savevm-async.c
-@@ -0,0 +1,182 @@
+@@ -0,0 +1,184 @@
 +/*
 + * 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 "system/block-backend.h"
 +#include "trace.h"
 +
 +QIOChannelSavevmAsync *
@@ -71,6 +72,7 @@ index 0000000000..06d5484778
 +                               size_t niov,
 +                               int **fds,
 +                               size_t *nfds,
++                               int flags,
 +                               Error **errp)
 +{
 +    QIOChannelSavevmAsync *saioc = QIO_CHANNEL_SAVEVM_ASYNC(ioc);
@@ -173,8 +175,9 @@ index 0000000000..06d5484778
 +
 +static void
 +qio_channel_savevm_async_set_aio_fd_handler(QIOChannel *ioc,
-+                                            AioContext *ctx,
++                                            AioContext *read_ctx,
 +                                            IOHandler *io_read,
++                                            AioContext *write_ctx,
 +                                            IOHandler *io_write,
 +                                            void *opaque)
 +{
@@ -268,14 +271,14 @@ index 0000000000..17ae2cb261
 +
 +#endif /* QIO_CHANNEL_SAVEVM_ASYNC_H */
 diff --git a/migration/meson.build b/migration/meson.build
-index 690487cf1a..8cac83c06c 100644
+index 9aa48b290e..cf66c78681 100644
 --- a/migration/meson.build
 +++ b/migration/meson.build
-@@ -13,6 +13,7 @@ softmmu_ss.add(files(
-   'block-dirty-bitmap.c',
+@@ -14,6 +14,7 @@ system_ss.add(files(
+   'block-active.c',
    'channel.c',
    'channel-block.c',
 +  'channel-savevm-async.c',
-   'colo-failover.c',
-   'colo.c',
-   'exec.c',
+   'cpr.c',
+   'cpr-transfer.c',
+   'cpu-throttle.c',

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

@@ -21,32 +21,40 @@ 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]
+[SR: improve aborting
+     register yank before migration_incoming_state_destroy]
 Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
 [FE: further improve aborting
      adapt to removal of QEMUFileOps
-     improve condition for entering final stage]
+     improve condition for entering final stage
+     adapt to QAPI and other changes for 8.2
+     make sure to not call vm_start() from coroutine
+     stop CPU throttling after finishing
+     force raw format when loading state as suggested by Friedrich Weber
+     improve setting state in savevm-end handler
+     improve runstate preservation
+     use dedicated iothread for state file to avoid deadlock, bug #6262]
 Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  hmp-commands-info.hx         |  13 +
- hmp-commands.hx              |  33 +++
+ hmp-commands.hx              |  17 +
  include/migration/snapshot.h |   2 +
- include/monitor/hmp.h        |   5 +
+ include/monitor/hmp.h        |   3 +
  migration/meson.build        |   1 +
- migration/savevm-async.c     | 538 +++++++++++++++++++++++++++++++++++
- monitor/hmp-cmds.c           |  57 ++++
- qapi/migration.json          |  34 +++
- qapi/misc.json               |  32 +++
+ migration/savevm-async.c     | 581 +++++++++++++++++++++++++++++++++++
+ monitor/hmp-cmds.c           |  38 +++
+ qapi/migration.json          |  34 ++
+ qapi/misc.json               |  18 ++
  qemu-options.hx              |  12 +
- softmmu/vl.c                 |  10 +
- 11 files changed, 737 insertions(+)
+ system/vl.c                  |  10 +
+ 11 files changed, 729 insertions(+)
  create mode 100644 migration/savevm-async.c
 
 diff --git a/hmp-commands-info.hx b/hmp-commands-info.hx
-index 754b1e8408..489c524e9e 100644
+index c59cd6637b..d1a7b99add 100644
 --- a/hmp-commands-info.hx
 +++ b/hmp-commands-info.hx
-@@ -540,6 +540,19 @@ SRST
+@@ -512,6 +512,19 @@ SRST
      Show current migration parameters.
  ERST
  
@@ -67,11 +75,11 @@ index 754b1e8408..489c524e9e 100644
          .name       = "balloon",
          .args_type  = "",
 diff --git a/hmp-commands.hx b/hmp-commands.hx
-index 673e39a697..039be0033d 100644
+index 06746f0afc..0c7c6f2c16 100644
 --- a/hmp-commands.hx
 +++ b/hmp-commands.hx
-@@ -1815,3 +1815,36 @@ SRST
-   Dump the FDT in dtb format to *filename*.
+@@ -1859,3 +1859,20 @@ SRST
+   List event channels in the guest
  ERST
  #endif
 +
@@ -84,22 +92,6 @@ index 673e39a697..039be0033d 100644
 +    },
 +
 +    {
-+        .name       = "snapshot-drive",
-+        .args_type  = "device:s,name:s",
-+        .params     = "device name",
-+        .help       = "Create internal snapshot.",
-+        .cmd = hmp_snapshot_drive,
-+    },
-+
-+    {
-+        .name       = "delete-drive-snapshot",
-+        .args_type  = "device:s,name:s",
-+        .params     = "device name",
-+        .help       = "Delete internal snapshot.",
-+        .cmd = hmp_delete_drive_snapshot,
-+    },
-+
-+    {
 +        .name       = "savevm-end",
 +        .args_type  = "",
 +        .params     = "",
@@ -108,21 +100,21 @@ index 673e39a697..039be0033d 100644
 +        .coroutine  = true,
 +    },
 diff --git a/include/migration/snapshot.h b/include/migration/snapshot.h
-index e72083b117..c846d37806 100644
+index 9e4dcaaa75..2581730d74 100644
 --- a/include/migration/snapshot.h
 +++ b/include/migration/snapshot.h
-@@ -61,4 +61,6 @@ bool delete_snapshot(const char *name,
-                     bool has_devices, strList *devices,
-                     Error **errp);
+@@ -68,4 +68,6 @@ bool delete_snapshot(const char *name,
+  */
+ void load_snapshot_resume(RunState state);
  
 +int load_snapshot_from_blockdev(const char *filename, Error **errp);
 +
  #endif
 diff --git a/include/monitor/hmp.h b/include/monitor/hmp.h
-index dfbc0c9a2f..440f86aba8 100644
+index ae116d9804..2596cc2426 100644
 --- a/include/monitor/hmp.h
 +++ b/include/monitor/hmp.h
-@@ -27,6 +27,7 @@ void hmp_info_status(Monitor *mon, const QDict *qdict);
+@@ -28,6 +28,7 @@ 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);
@@ -130,57 +122,61 @@ index dfbc0c9a2f..440f86aba8 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);
-@@ -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);
+@@ -92,6 +93,8 @@ void hmp_closefd(Monitor *mon, const QDict *qdict);
+ void hmp_mouse_move(Monitor *mon, const QDict *qdict);
+ void hmp_mouse_button(Monitor *mon, const QDict *qdict);
+ void hmp_mouse_set(Monitor *mon, const QDict *qdict);
 +void hmp_savevm_start(Monitor *mon, const QDict *qdict);
-+void hmp_snapshot_drive(Monitor *mon, const QDict *qdict);
-+void hmp_delete_drive_snapshot(Monitor *mon, const QDict *qdict);
 +void hmp_savevm_end(Monitor *mon, const QDict *qdict);
  void hmp_sendkey(Monitor *mon, const QDict *qdict);
  void coroutine_fn hmp_screendump(Monitor *mon, const QDict *qdict);
  void hmp_chardev_add(Monitor *mon, const QDict *qdict);
 diff --git a/migration/meson.build b/migration/meson.build
-index 8cac83c06c..0842d00cd2 100644
+index cf66c78681..46e92249a1 100644
 --- a/migration/meson.build
 +++ b/migration/meson.build
-@@ -24,6 +24,7 @@ softmmu_ss.add(files(
-   'multifd-zlib.c',
+@@ -33,6 +33,7 @@ system_ss.add(files(
+   'options.c',
    'postcopy-ram.c',
    'savevm.c',
 +  'savevm-async.c',
    'socket.c',
    'tls.c',
- ), gnutls)
+   'threadinfo.c',
 diff --git a/migration/savevm-async.c b/migration/savevm-async.c
 new file mode 100644
-index 0000000000..dc30558713
+index 0000000000..56e0fa6c69
 --- /dev/null
 +++ b/migration/savevm-async.c
-@@ -0,0 +1,538 @@
+@@ -0,0 +1,581 @@
 +#include "qemu/osdep.h"
 +#include "migration/channel-savevm-async.h"
 +#include "migration/migration.h"
++#include "migration/migration-stats.h"
++#include "migration/options.h"
 +#include "migration/savevm.h"
 +#include "migration/snapshot.h"
 +#include "migration/global_state.h"
 +#include "migration/ram.h"
 +#include "migration/qemu-file.h"
-+#include "sysemu/sysemu.h"
-+#include "sysemu/runstate.h"
++#include "system/cpu-throttle.h"
++#include "system/system.h"
++#include "system/runstate.h"
 +#include "block/block.h"
-+#include "sysemu/block-backend.h"
++#include "system/block-backend.h"
 +#include "qapi/error.h"
 +#include "qapi/qmp/qerror.h"
-+#include "qapi/qmp/qdict.h"
++#include "qobject/qdict.h"
 +#include "qapi/qapi-commands-migration.h"
 +#include "qapi/qapi-commands-misc.h"
 +#include "qapi/qapi-commands-block.h"
 +#include "qemu/cutils.h"
++#include "qemu/error-report.h"
 +#include "qemu/timer.h"
 +#include "qemu/main-loop.h"
 +#include "qemu/rcu.h"
++#include "qemu/yank.h"
++#include "system/iothread.h"
 +
 +/* #define DEBUG_SAVEVM_STATE */
 +
@@ -207,12 +203,13 @@ index 0000000000..dc30558713
 +    int state;
 +    Error *error;
 +    Error *blocker;
-+    int saved_vm_running;
++    int vm_needs_start;
 +    QEMUFile *file;
 +    int64_t total_time;
 +    QEMUBH *finalize_bh;
 +    Coroutine *co;
 +    QemuCoSleep target_close_wait;
++    IOThread *iothread;
 +} snap_state;
 +
 +static bool savevm_aborted(void)
@@ -231,24 +228,20 @@ index 0000000000..dc30558713
 +        info->bytes = s->bs_pos;
 +        switch (s->state) {
 +        case SAVE_STATE_ERROR:
-+            info->has_status = true;
 +            info->status = g_strdup("failed");
 +            info->has_total_time = true;
 +            info->total_time = s->total_time;
 +            if (s->error) {
-+                info->has_error = true;
 +                info->error = g_strdup(error_get_pretty(s->error));
 +            }
 +            break;
 +        case SAVE_STATE_ACTIVE:
-+            info->has_status = true;
 +            info->status = g_strdup("active");
 +            info->has_total_time = true;
 +            info->total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME)
 +                - s->total_time;
 +            break;
 +        case SAVE_STATE_COMPLETED:
-+            info->has_status = true;
 +            info->status = g_strdup("completed");
 +            info->has_total_time = true;
 +            info->total_time = s->total_time;
@@ -274,6 +267,7 @@ index 0000000000..dc30558713
 +    }
 +
 +    if (snap_state.target) {
++        BlockDriverState *target_bs = blk_bs(snap_state.target);
 +        if (!savevm_aborted()) {
 +            /* try to truncate, but ignore errors (will fail on block devices).
 +            * note1: bdrv_read() need whole blocks, so we need to round up
@@ -282,7 +276,9 @@ index 0000000000..dc30558713
 +            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);
++        if (target_bs) {
++            bdrv_op_unblock_all(target_bs, snap_state.blocker);
++        }
 +        error_free(snap_state.blocker);
 +        snap_state.blocker = NULL;
 +        blk_unref(snap_state.target);
@@ -294,7 +290,7 @@ index 0000000000..dc30558713
 +    return ret;
 +}
 +
-+static void save_snapshot_error(const char *fmt, ...)
++static void G_GNUC_PRINTF(1, 2) save_snapshot_error(const char *fmt, ...)
 +{
 +    va_list ap;
 +    char *msg;
@@ -306,7 +302,7 @@ index 0000000000..dc30558713
 +    DPRINTF("save_snapshot_error: %s\n", msg);
 +
 +    if (!snap_state.error) {
-+        error_set(&snap_state.error, ERROR_CLASS_GENERIC_ERROR, "%s", msg);
++        error_setg(&snap_state.error, "%s", msg);
 +    }
 +
 +    g_free (msg);
@@ -317,7 +313,6 @@ index 0000000000..dc30558713
 +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();
@@ -334,10 +329,9 @@ index 0000000000..dc30558713
 +     * so move it back. It can stay in the main context and live out its live
 +     * there, since we're done with it after this method ends anyway.
 +     */
-+    aio_context_acquire(iohandler_ctx);
 +    blk_set_aio_context(snap_state.target, qemu_get_aio_context(), NULL);
-+    aio_context_release(iohandler_ctx);
 +
++    snap_state.vm_needs_start = runstate_is_running();
 +    ret = vm_stop_force_state(RUN_STATE_FINISH_MIGRATE);
 +    if (ret < 0) {
 +        save_snapshot_error("vm_stop_force_state error %d", ret);
@@ -345,7 +339,7 @@ index 0000000000..dc30558713
 +
 +    if (!aborted) {
 +        /* skip state saving if we aborted, snapshot will be invalid anyway */
-+        (void)qemu_savevm_state_complete_precopy(snap_state.file, false, false);
++        (void)qemu_savevm_state_complete_precopy(snap_state.file, false);
 +        ret = qemu_file_get_error(snap_state.file);
 +        if (ret < 0) {
 +            save_snapshot_error("qemu_savevm_state_complete_precopy error %d", ret);
@@ -361,6 +355,12 @@ index 0000000000..dc30558713
 +        ret || aborted ? MIGRATION_STATUS_FAILED : MIGRATION_STATUS_COMPLETED);
 +    ms->to_dst_file = NULL;
 +
++    /*
++     * Same as in migration_iteration_finish(): saving RAM might've turned on CPU throttling for
++     * auto-converge, make sure to disable it.
++     */
++    cpu_throttle_stop();
++
 +    qemu_savevm_state_cleanup();
 +
 +    ret = save_snapshot_cleanup();
@@ -378,9 +378,9 @@ index 0000000000..dc30558713
 +        save_snapshot_error("process_savevm_cleanup: invalid state: %d",
 +                            snap_state.state);
 +    }
-+    if (snap_state.saved_vm_running) {
++    if (snap_state.vm_needs_start) {
 +        vm_start();
-+        snap_state.saved_vm_running = false;
++        snap_state.vm_needs_start = false;
 +    }
 +
 +    DPRINTF("timing: process_savevm_finalize (full) took %ld ms\n",
@@ -405,14 +405,18 @@ index 0000000000..dc30558713
 +    }
 +
 +    while (snap_state.state == SAVE_STATE_ACTIVE) {
-+        uint64_t pending_size, pend_precopy, pend_compatible, pend_postcopy;
++        uint64_t pending_size, pend_precopy, pend_postcopy;
++        uint64_t threshold = 400 * 1000;
 +
-+        /* 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;
++        /*
++         * Similar to what is done in migration.c, call the exact variant only
++         * once pend_precopy in the estimate is below the threshold.
++         */
++        qemu_savevm_state_pending_estimate(&pend_precopy, &pend_postcopy);
++        if (pend_precopy <= threshold) {
++            qemu_savevm_state_pending_exact(&pend_precopy, &pend_postcopy);
++        }
++        pending_size = pend_precopy + pend_postcopy;
 +
 +        /*
 +         * A guest reaching this cutoff is dirtying lots of RAM. It should be
@@ -423,7 +427,7 @@ index 0000000000..dc30558713
 +        maxlen = blk_getlength(snap_state.target) - 100*1024*1024;
 +
 +        /* Note that there is no progress for pend_postcopy when iterating */
-+        if (pending_size - pend_postcopy > 400000 && snap_state.bs_pos + pending_size < maxlen) {
++        if (pend_precopy > threshold && 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);
@@ -432,11 +436,7 @@ index 0000000000..dc30558713
 +            DPRINTF("savevm iterate pending size %lu ret %d\n", pending_size, ret);
 +        } else {
 +            qemu_system_wakeup_request(QEMU_WAKEUP_REASON_OTHER, NULL);
-+            ret = global_state_store();
-+            if (ret) {
-+                save_snapshot_error("global_state_store error %d", ret);
-+                break;
-+            }
++            global_state_store();
 +
 +            DPRINTF("savevm iterate complete\n");
 +            break;
@@ -455,19 +455,25 @@ index 0000000000..dc30558713
 +     * so move there now and after every flush.
 +     */
 +    aio_co_reschedule_self(qemu_get_aio_context());
-+    for (bs = bdrv_first(&it); bs; bs = bdrv_next(&it)) {
++    bdrv_graph_co_rdlock();
++    bs = bdrv_first(&it);
++    bdrv_graph_co_rdunlock();
++    while (bs) {
 +        /* target has BDRV_O_NO_FLUSH, no sense calling bdrv_flush on it */
-+        if (bs == blk_bs(snap_state.target)) {
-+            continue;
-+        }
-+
-+        AioContext *bs_ctx = bdrv_get_aio_context(bs);
-+        if (bs_ctx != qemu_get_aio_context()) {
-+            DPRINTF("savevm: async flushing drive %s\n", bs->filename);
-+            aio_co_reschedule_self(bs_ctx);
-+            bdrv_flush(bs);
-+            aio_co_reschedule_self(qemu_get_aio_context());
++        if (bs != blk_bs(snap_state.target)) {
++            AioContext *bs_ctx = bdrv_get_aio_context(bs);
++            if (bs_ctx != qemu_get_aio_context()) {
++                DPRINTF("savevm: async flushing drive %s\n", bs->filename);
++                aio_co_reschedule_self(bs_ctx);
++                bdrv_graph_co_rdlock();
++                bdrv_flush(bs);
++                bdrv_graph_co_rdunlock();
++                aio_co_reschedule_self(qemu_get_aio_context());
++            }
 +        }
++        bdrv_graph_co_rdlock();
++        bs = bdrv_next(&it);
++        bdrv_graph_co_rdunlock();
 +    }
 +
 +    DPRINTF("timing: async flushing took %ld ms\n",
@@ -476,33 +482,33 @@ index 0000000000..dc30558713
 +    qemu_bh_schedule(snap_state.finalize_bh);
 +}
 +
-+void qmp_savevm_start(bool has_statefile, const char *statefile, Error **errp)
++static void savevm_cleanup_iothread(void) {
++    if (snap_state.iothread) {
++        iothread_destroy(snap_state.iothread);
++        snap_state.iothread = NULL;
++    }
++}
++
++void qmp_savevm_start(const char *statefile, Error **errp)
 +{
 +    Error *local_err = NULL;
 +    MigrationState *ms = migrate_get_current();
-+    AioContext *iohandler_ctx = iohandler_get_aio_context();
++    BlockDriverState *target_bs = NULL;
++    int ret = 0;
 +
 +    int bdrv_oflags = BDRV_O_RDWR | BDRV_O_RESIZE | BDRV_O_NO_FLUSH;
 +
 +    if (snap_state.state != SAVE_STATE_DONE) {
-+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+                  "VM snapshot already started\n");
++        error_setg(errp, "VM snapshot already started\n");
 +        return;
 +    }
 +
-+    if (migration_is_running(ms->state)) {
-+        error_set(errp, ERROR_CLASS_GENERIC_ERROR, QERR_MIGRATION_ACTIVE);
-+        return;
-+    }
-+
-+    if (migrate_use_block()) {
-+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+                  "Block migration and snapshots are incompatible");
++    if (migration_is_running()) {
++        error_setg(errp, "There's a migration process in progress");
 +        return;
 +    }
 +
 +    /* initialize snapshot info */
-+    snap_state.saved_vm_running = runstate_is_running();
 +    snap_state.bs_pos = 0;
 +    snap_state.total_time = qemu_clock_get_ms(QEMU_CLOCK_REALTIME);
 +    snap_state.blocker = NULL;
@@ -513,14 +519,28 @@ index 0000000000..dc30558713
 +        snap_state.error = NULL;
 +    }
 +
-+    if (!has_statefile) {
++    if (!statefile) {
++        snap_state.vm_needs_start = runstate_is_running();
 +        vm_stop(RUN_STATE_SAVE_VM);
 +        snap_state.state = SAVE_STATE_COMPLETED;
 +        return;
 +    }
 +
 +    if (qemu_savevm_state_blocked(errp)) {
-+        return;
++        goto fail;
++    }
++
++    if (snap_state.iothread) {
++        /* This is not expected, so warn about it, but no point in re-creating a new iothread. */
++        warn_report("iothread for snapshot already exists - re-using");
++    } else {
++        snap_state.iothread =
++            iothread_create("__proxmox_savevm_async_iothread__", &local_err);
++        if (!snap_state.iothread) {
++            error_setg(errp, "creating iothread failed: %s",
++                       local_err ? error_get_pretty(local_err) : "unknown error");
++            goto fail;
++        }
 +    }
 +
 +    /* Open the image */
@@ -529,8 +549,13 @@ index 0000000000..dc30558713
 +    qdict_put_str(options, "driver", "raw");
 +    snap_state.target = blk_new_open(statefile, NULL, options, bdrv_oflags, &local_err);
 +    if (!snap_state.target) {
-+        error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
-+        goto restart;
++        error_setg(errp, "failed to open '%s'", statefile);
++        goto fail;
++    }
++    target_bs = blk_bs(snap_state.target);
++    if (!target_bs) {
++        error_setg(errp, "failed to open '%s' - no block driver state", statefile);
++        goto fail;
 +    }
 +
 +    QIOChannel *ioc = QIO_CHANNEL(qio_channel_savevm_async_new(snap_state.target,
@@ -538,8 +563,8 @@ index 0000000000..dc30558713
 +    snap_state.file = qemu_file_new_output(ioc);
 +
 +    if (!snap_state.file) {
-+        error_set(errp, ERROR_CLASS_GENERIC_ERROR, "failed to open '%s'", statefile);
-+        goto restart;
++        error_setg(errp, "failed to open '%s'", statefile);
++        goto fail;
 +    }
 +
 +    /*
@@ -547,82 +572,66 @@ index 0000000000..dc30558713
 +     * State is cleared in process_savevm_co, but has to be initialized
 +     * here (blocking main thread, from QMP) to avoid race conditions.
 +     */
-+    migrate_init(ms);
-+    memset(&ram_counters, 0, sizeof(ram_counters));
++    if (migrate_init(ms, errp) != 0) {
++        goto fail;
++    }
++    memset(&mig_stats, 0, sizeof(mig_stats));
 +    ms->to_dst_file = snap_state.file;
 +
 +    error_setg(&snap_state.blocker, "block device is in use by savevm");
-+    blk_op_block_all(snap_state.target, snap_state.blocker);
++    bdrv_op_block_all(target_bs, snap_state.blocker);
 +
 +    snap_state.state = SAVE_STATE_ACTIVE;
 +    snap_state.finalize_bh = qemu_bh_new(process_savevm_finalize, &snap_state);
-+    snap_state.co = qemu_coroutine_create(&process_savevm_co, NULL);
-+    qemu_mutex_unlock_iothread();
 +    qemu_savevm_state_header(snap_state.file);
-+    qemu_savevm_state_setup(snap_state.file);
-+    qemu_mutex_lock_iothread();
++    ret = qemu_savevm_state_setup(snap_state.file, &local_err);
++    if (ret != 0) {
++        error_setg_errno(errp, -ret, "savevm state setup failed: %s",
++                         local_err ? error_get_pretty(local_err) : "unknown error");
++        goto fail;
++    }
 +
-+    /* Async processing from here on out happens in iohandler context, so let
-+     * the target bdrv have its home there.
-+     */
-+    blk_set_aio_context(snap_state.target, iohandler_ctx, &local_err);
++    ret = blk_set_aio_context(snap_state.target, snap_state.iothread->ctx, &local_err);
++    if (ret != 0) {
++        error_setg_errno(errp, -ret, "failed to set iothread context for VM state target: %s",
++                         local_err ? error_get_pretty(local_err) : "unknown error");
++        goto fail;
++    }
 +
-+    aio_co_schedule(iohandler_ctx, snap_state.co);
++    snap_state.co = qemu_coroutine_create(&process_savevm_co, NULL);
++    aio_co_schedule(snap_state.iothread->ctx, snap_state.co);
 +
 +    return;
 +
-+restart:
-+
++fail:
++    savevm_cleanup_iothread();
 +    save_snapshot_error("setup failed");
-+
-+    if (snap_state.saved_vm_running) {
-+        vm_start();
-+        snap_state.saved_vm_running = false;
-+    }
 +}
 +
-+void coroutine_fn qmp_savevm_end(Error **errp)
++static void coroutine_fn wait_for_close_co(void *opaque)
 +{
 +    int64_t timeout;
 +
-+    if (snap_state.state == SAVE_STATE_DONE) {
-+        error_set(errp, ERROR_CLASS_GENERIC_ERROR,
-+                  "VM snapshot not started\n");
-+        return;
-+    }
-+
-+    if (snap_state.state == SAVE_STATE_ACTIVE) {
-+        snap_state.state = SAVE_STATE_CANCELLED;
-+        goto wait_for_close;
-+    }
-+
-+    if (snap_state.saved_vm_running) {
-+        vm_start();
-+        snap_state.saved_vm_running = false;
-+    }
-+
-+    snap_state.state = SAVE_STATE_DONE;
-+
-+wait_for_close:
-+    if (!snap_state.target) {
-+        DPRINTF("savevm-end: no target file open\n");
-+        return;
-+    }
-+
-+    /* wait until cleanup is done before returning, this ensures that after this
-+     * call exits the statefile will be closed and can be removed immediately */
-+    DPRINTF("savevm-end: waiting for cleanup\n");
-+    timeout = 30L * 1000 * 1000 * 1000;
-+    qemu_co_sleep_ns_wakeable(&snap_state.target_close_wait,
-+                              QEMU_CLOCK_REALTIME, timeout);
 +    if (snap_state.target) {
-+        save_snapshot_error("timeout waiting for target file close in "
-+                            "qmp_savevm_end");
-+        /* we cannot assume the snapshot finished in this case, so leave the
-+         * state alone - caller has to figure something out */
-+        return;
++        /* 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;
++        }
++    } else {
++        DPRINTF("savevm-end: no target file open\n");
 +    }
 +
++    savevm_cleanup_iothread();
++
 +    // 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;
@@ -631,41 +640,59 @@ index 0000000000..dc30558713
 +    DPRINTF("savevm-end: cleanup done\n");
 +}
 +
-+// FIXME: Deprecated
-+void qmp_snapshot_drive(const char *device, const char *name, Error **errp)
++void qmp_savevm_end(Error **errp)
 +{
-+    // Compatibility to older qemu-server.
-+    qmp_blockdev_snapshot_internal_sync(device, name, errp);
-+}
++    if (snap_state.state == SAVE_STATE_DONE) {
++        error_setg(errp, "VM snapshot not started\n");
++        return;
++    }
 +
-+// FIXME: Deprecated
-+void qmp_delete_drive_snapshot(const char *device, const char *name,
-+                               Error **errp)
-+{
-+    // Compatibility to older qemu-server.
-+    (void)qmp_blockdev_snapshot_delete_internal_sync(device, false, NULL,
-+                                                     true, name, errp);
++    Coroutine *wait_for_close = qemu_coroutine_create(wait_for_close_co, NULL);
++
++    if (snap_state.state == SAVE_STATE_ACTIVE) {
++        snap_state.state = SAVE_STATE_CANCELLED;
++        qemu_coroutine_enter(wait_for_close);
++        return;
++    }
++
++    if (snap_state.vm_needs_start) {
++        vm_start();
++        snap_state.vm_needs_start = false;
++    }
++
++    qemu_coroutine_enter(wait_for_close);
 +}
 +
 +int load_snapshot_from_blockdev(const char *filename, Error **errp)
 +{
 +    BlockBackend *be;
++    BlockDriverState *bs = NULL;
 +    Error *local_err = NULL;
 +    Error *blocker = NULL;
++    QDict *options;
 +
 +    QEMUFile *f;
 +    size_t bs_pos = 0;
 +    int ret = -EINVAL;
 +
-+    be = blk_new_open(filename, NULL, NULL, 0, &local_err);
++    options = qdict_new();
++    qdict_put_str(options, "driver", "raw");
++
++    be = blk_new_open(filename, NULL, options, 0, &local_err);
 +
 +    if (!be) {
 +        error_setg(errp, "Could not open VM state file");
 +        goto the_end;
 +    }
 +
++    bs = blk_bs(be);
++    if (!bs) {
++        error_setg(errp, "Could not open VM state file - missing block driver state");
++        goto the_end;
++    }
++
 +    error_setg(&blocker, "block device is in use by load state");
-+    blk_op_block_all(be, blocker);
++    bdrv_op_block_all(bs, blocker);
 +
 +    /* restore the VM state */
 +    f = qemu_file_new_input(QIO_CHANNEL(qio_channel_savevm_async_new(be, &bs_pos)));
@@ -681,6 +708,10 @@ index 0000000000..dc30558713
 +    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");
@@ -691,46 +722,37 @@ index 0000000000..dc30558713
 +
 + the_end:
 +    if (be) {
-+        blk_op_unblock_all(be, blocker);
++        if (bs) {
++            bdrv_op_unblock_all(bs, blocker);
++        }
 +        error_free(blocker);
 +        blk_unref(be);
 +    }
 +    return ret;
 +}
 diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 480b798963..cfebfd1db5 100644
+index 7ded3378cf..bade2a4b92 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);
+@@ -24,6 +24,7 @@
+ #include "qapi/error.h"
+ #include "qapi/qapi-commands-control.h"
+ #include "qapi/qapi-commands-machine.h"
++#include "qapi/qapi-commands-migration.h"
+ #include "qapi/qapi-commands-misc.h"
+ #include "qobject/qdict.h"
+ #include "qemu/cutils.h"
+@@ -434,3 +435,40 @@ void hmp_dumpdtb(Monitor *mon, const QDict *qdict)
+     monitor_printf(mon, "DTB dumped to '%s'\n", filename);
  }
- 
+ #endif
++
 +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);
++    qmp_savevm_start(statefile, &errp);
 +    hmp_handle_error(mon, errp);
 +}
 +
@@ -747,7 +769,7 @@ index 480b798963..cfebfd1db5 100644
 +    SaveVMInfo *info;
 +    info = qmp_query_savevm(NULL);
 +
-+    if (info->has_status) {
++    if (info->status) {
 +        monitor_printf(mon, "savevm status: %s\n", info->status);
 +        monitor_printf(mon, "total time: %" PRIu64 " milliseconds\n",
 +                       info->total_time);
@@ -757,21 +779,17 @@ index 480b798963..cfebfd1db5 100644
 +    if (info->has_bytes) {
 +        monitor_printf(mon, "Bytes saved: %"PRIu64"\n", info->bytes);
 +    }
-+    if (info->has_error) {
++    if (info->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
+index 8b9c53595c..ff3479da65 100644
 --- a/qapi/migration.json
 +++ b/qapi/migration.json
-@@ -261,6 +261,40 @@
-            '*compression': 'CompressionStats',
-            '*socket-address': ['SocketAddress'] } }
+@@ -279,6 +279,40 @@
+            '*dirty-limit-throttle-time-per-round': 'uint64',
+            '*dirty-limit-ring-full-time': 'uint64'} }
  
 +##
 +# @SaveVMInfo:
@@ -811,10 +829,10 @@ index 88ecf86ac8..4435866379 100644
  # @query-migrate:
  #
 diff --git a/qapi/misc.json b/qapi/misc.json
-index 27ef5a2b20..b3ce75dcae 100644
+index 559b66f201..7959e89c1e 100644
 --- a/qapi/misc.json
 +++ b/qapi/misc.json
-@@ -435,6 +435,38 @@
+@@ -454,6 +454,24 @@
  ##
  { 'command': 'query-fdsets', 'returns': ['FdsetInfo'] }
  
@@ -823,41 +841,27 @@ index 27ef5a2b20..b3ce75dcae 100644
 +#
 +# Prepare for snapshot and halt VM. Save VM state to statefile.
 +#
++# @statefile: target file that state should be written to.
++#
 +##
 +{ '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 }
++{ 'command': 'savevm-end' }
 +
  ##
  # @CommandLineParameterType:
  #
 diff --git a/qemu-options.hx b/qemu-options.hx
-index 7f99d15b23..54efb127c4 100644
+index dc694a99a3..defee0c06a 100644
 --- a/qemu-options.hx
 +++ b/qemu-options.hx
-@@ -4391,6 +4391,18 @@ SRST
+@@ -4862,6 +4862,18 @@ SRST
      Start right away with a saved state (``loadvm`` in monitor)
  ERST
  
@@ -876,11 +880,11 @@ index 7f99d15b23..54efb127c4 100644
  #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 5f7f6ca981..21f067d115 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -164,6 +164,7 @@ static const char *accelerators;
+diff --git a/system/vl.c b/system/vl.c
+index ec93988a03..9b36ace6b4 100644
+--- a/system/vl.c
++++ b/system/vl.c
+@@ -171,6 +171,7 @@ static const char *accelerators;
  static bool have_custom_ram_size;
  static const char *ram_memdev_id;
  static QDict *machine_opts_dict;
@@ -888,10 +892,10 @@ index 5f7f6ca981..21f067d115 100644
  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;
-@@ -2607,6 +2608,12 @@ void qmp_x_exit_preconfig(Error **errp)
- 
-     if (loadvm) {
+@@ -2814,6 +2815,12 @@ void qmp_x_exit_preconfig(Error **errp)
+         RunState state = autostart ? RUN_STATE_RUNNING : runstate_get();
          load_snapshot(loadvm, NULL, false, NULL, &error_fatal);
+         load_snapshot_resume(state);
 +    } else if (loadstate) {
 +        Error *local_err = NULL;
 +        if (load_snapshot_from_blockdev(loadstate, &local_err) < 0) {
@@ -901,7 +905,7 @@ index 5f7f6ca981..21f067d115 100644
      }
      if (replay_mode != REPLAY_MODE_NONE) {
          replay_vmstate_init();
-@@ -3151,6 +3158,9 @@ void qemu_init(int argc, char **argv)
+@@ -3360,6 +3367,9 @@ void qemu_init(int argc, char **argv)
              case QEMU_OPTION_loadvm:
                  loadvm = optarg;
                  break;

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

@@ -13,27 +13,27 @@ 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.c    | 48 +++++++++++++++++++++++++++-------------
  migration/qemu-file.h    |  2 ++
- migration/savevm-async.c |  5 ++--
- 3 files changed, 38 insertions(+), 18 deletions(-)
+ migration/savevm-async.c |  5 +++--
+ 3 files changed, 38 insertions(+), 17 deletions(-)
 
 diff --git a/migration/qemu-file.c b/migration/qemu-file.c
-index 2d5f74ffc2..9fd97e6fe1 100644
+index 1303a5bf58..6e2d58d5c0 100644
 --- a/migration/qemu-file.c
 +++ b/migration/qemu-file.c
-@@ -31,8 +31,8 @@
- #include "trace.h"
- #include "qapi/error.h"
+@@ -34,8 +34,8 @@
+ #include "rdma.h"
+ #include "io/channel-file.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 {
+ typedef struct FdEntry {
+     QTAILQ_ENTRY(FdEntry) entry;
+@@ -48,7 +48,8 @@ struct QEMUFile {
  
      int buf_index;
      int buf_size; /* 0 when writing */
@@ -43,8 +43,8 @@ index 2d5f74ffc2..9fd97e6fe1 100644
  
      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;
+@@ -108,7 +109,9 @@ int qemu_file_shutdown(QEMUFile *f)
+     return 0;
  }
  
 -static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
@@ -54,16 +54,16 @@ index 2d5f74ffc2..9fd97e6fe1 100644
  {
      QEMUFile *f;
  
-@@ -136,6 +139,8 @@ static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
-     object_ref(ioc);
-     f->ioc = ioc;
+@@ -119,6 +122,8 @@ static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
      f->is_writable = is_writable;
+     f->can_pass_fd = qio_channel_has_feature(ioc, QIO_CHANNEL_FEATURE_FD_PASS);
+     QTAILQ_INIT(&f->fds);
 +    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)
+@@ -129,17 +134,27 @@ static QEMUFile *qemu_file_new_impl(QIOChannel *ioc, bool is_writable)
   */
  QEMUFile *qemu_file_get_return_path(QEMUFile *f)
  {
@@ -93,26 +93,27 @@ index 2d5f74ffc2..9fd97e6fe1 100644
 +    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)
+ /*
+@@ -339,7 +354,7 @@ static ssize_t coroutine_mixed_fn 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);
+-        struct iovec iov = { f->buf + pending, IO_BUF_SIZE - pending };
++        struct iovec iov = { f->buf + pending, f->buf_allocated_size - pending };
+         len = qio_channel_readv_full(f->ioc, &iov, 1, pfds, pnfd, 0,
+                                      &local_error);
          if (len == QIO_CHANNEL_ERR_BLOCK) {
-             if (qemu_in_coroutine()) {
-@@ -464,6 +479,8 @@ int qemu_fclose(QEMUFile *f)
+@@ -443,6 +458,9 @@ int qemu_fclose(QEMUFile *f)
+         g_free(fde);
      }
      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)
+     error_free(f->last_error_obj);
+     g_free(f);
+     trace_qemu_file_fclose();
+@@ -491,7 +509,7 @@ 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;
@@ -121,7 +122,7 @@ index 2d5f74ffc2..9fd97e6fe1 100644
              qemu_fflush(f);
          }
      }
-@@ -544,7 +561,7 @@ void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, size_t size)
+@@ -516,7 +534,7 @@ void qemu_put_buffer(QEMUFile *f, const uint8_t *buf, size_t size)
      }
  
      while (size > 0) {
@@ -130,7 +131,7 @@ index 2d5f74ffc2..9fd97e6fe1 100644
          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)
+@@ -662,8 +680,8 @@ size_t coroutine_mixed_fn qemu_peek_buffer(QEMUFile *f, uint8_t **buf, size_t si
      size_t index;
  
      assert(!qemu_file_is_writable(f));
@@ -141,7 +142,7 @@ index 2d5f74ffc2..9fd97e6fe1 100644
  
      /* 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)
+@@ -713,7 +731,7 @@ size_t coroutine_mixed_fn qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size
          size_t res;
          uint8_t *src;
  
@@ -150,16 +151,16 @@ index 2d5f74ffc2..9fd97e6fe1 100644
          if (res == 0) {
              return done;
          }
-@@ -676,7 +693,7 @@ size_t qemu_get_buffer(QEMUFile *f, uint8_t *buf, size_t size)
+@@ -747,7 +765,7 @@ size_t coroutine_mixed_fn 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)
+ size_t coroutine_mixed_fn 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)
+@@ -772,7 +790,7 @@ int coroutine_mixed_fn qemu_peek_byte(QEMUFile *f, int offset)
      int index = f->buf_index + offset;
  
      assert(!qemu_file_is_writable(f));
@@ -168,34 +169,25 @@ index 2d5f74ffc2..9fd97e6fe1 100644
  
      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
+index f5b9f430e0..0179b90698 100644
 --- a/migration/qemu-file.h
 +++ b/migration/qemu-file.h
-@@ -63,7 +63,9 @@ typedef struct QEMUFileHooks {
- } QEMUFileHooks;
+@@ -30,7 +30,9 @@
+ #include "io/channel.h"
  
  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);
  
+ G_DEFINE_AUTOPTR_CLEANUP_FUNC(QEMUFile, qemu_fclose)
 diff --git a/migration/savevm-async.c b/migration/savevm-async.c
-index dc30558713..a38e7351c1 100644
+index 56e0fa6c69..730b815494 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)
+@@ -409,7 +409,7 @@ void qmp_savevm_start(const char *statefile, Error **errp)
  
      QIOChannel *ioc = QIO_CHANNEL(qio_channel_savevm_async_new(snap_state.target,
                                                                 &snap_state.bs_pos));
@@ -203,9 +195,9 @@ index dc30558713..a38e7351c1 100644
 +    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);
+         error_setg(errp, "failed to open '%s'", statefile);
+@@ -544,7 +544,8 @@ int load_snapshot_from_blockdev(const char *filename, Error **errp)
+     bdrv_op_block_all(bs, blocker);
  
      /* restore the VM state */
 -    f = qemu_file_new_input(QIO_CHANNEL(qio_channel_savevm_async_new(be, &bs_pos)));

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

@@ -4,32 +4,34 @@ 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>
+[FE: adapt to changed function signatures
+     adhere to block graph lock requirements
+     use dedicated function to open file child]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  block/meson.build |   1 +
- block/zeroinit.c  | 198 ++++++++++++++++++++++++++++++++++++++++++++++
- 2 files changed, 199 insertions(+)
+ block/zeroinit.c  | 207 ++++++++++++++++++++++++++++++++++++++++++++++
+ 2 files changed, 208 insertions(+)
  create mode 100644 block/zeroinit.c
 
 diff --git a/block/meson.build b/block/meson.build
-index b7c68b83a3..020a89ae07 100644
+index 34b1b2a306..a21d9a5411 100644
 --- a/block/meson.build
 +++ b/block/meson.build
-@@ -43,6 +43,7 @@ block_ss.add(files(
-   'vmdk.c',
-   'vpc.c',
+@@ -39,6 +39,7 @@ block_ss.add(files(
+   'throttle.c',
+   'throttle-groups.c',
    'write-threshold.c',
 +  'zeroinit.c',
- ), zstd, zlib, gnutls)
+ ), zstd, zlib)
  
- softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+ system_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..b60e1b84dc
+index 0000000000..f9d513db15
 --- /dev/null
 +++ b/block/zeroinit.c
-@@ -0,0 +1,198 @@
+@@ -0,0 +1,207 @@
 +/*
 + * Filter to fake a zero-initialized block device.
 + *
@@ -43,8 +45,10 @@ index 0000000000..b60e1b84dc
 +#include "qemu/osdep.h"
 +#include "qapi/error.h"
 +#include "block/block_int.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qmp/qstring.h"
++#include "block/block-io.h"
++#include "block/graph-lock.h"
++#include "qobject/qdict.h"
++#include "qobject/qstring.h"
 +#include "qemu/cutils.h"
 +#include "qemu/option.h"
 +#include "qemu/module.h"
@@ -108,12 +112,9 @@ index 0000000000..b60e1b84dc
 +    }
 +
 +    /* Open the raw file */
-+    bs->file = bdrv_open_child(qemu_opt_get(opts, "x-next"), options, "next",
-+                               bs, &child_of_bds,
-+                               BDRV_CHILD_FILTERED | BDRV_CHILD_PRIMARY,
-+                               false, &local_err);
-+    if (local_err) {
-+        ret = -EINVAL;
++    ret = bdrv_open_file_child(qemu_opt_get(opts, "x-next"), options, "next",
++                               bs, &local_err);
++    if (ret < 0) {
 +        error_propagate(errp, local_err);
 +        goto fail;
 +    }
@@ -124,7 +125,9 @@ index 0000000000..b60e1b84dc
 +    ret = 0;
 +fail:
 +    if (ret < 0) {
++        bdrv_graph_wrlock();
 +        bdrv_unref_child(bs, bs->file);
++        bdrv_graph_wrunlock();
 +    }
 +    qemu_opts_del(opts);
 +    return ret;
@@ -136,19 +139,22 @@ index 0000000000..b60e1b84dc
 +    (void)s;
 +}
 +
-+static int64_t zeroinit_getlength(BlockDriverState *bs)
++static coroutine_fn int64_t GRAPH_RDLOCK
++zeroinit_co_getlength(BlockDriverState *bs)
 +{
-+    return bdrv_getlength(bs->file->bs);
++    return bdrv_co_getlength(bs->file->bs);
 +}
 +
-+static int coroutine_fn zeroinit_co_preadv(BlockDriverState *bs,
-+    int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++static int coroutine_fn GRAPH_RDLOCK
++zeroinit_co_preadv(BlockDriverState *bs, int64_t offset, int64_t bytes,
++                   QEMUIOVector *qiov, BdrvRequestFlags flags)
 +{
 +    return bdrv_co_preadv(bs->file, offset, bytes, qiov, flags);
 +}
 +
-+static int coroutine_fn zeroinit_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset,
-+                                                 int64_t bytes, BdrvRequestFlags flags)
++static int coroutine_fn GRAPH_RDLOCK
++zeroinit_co_pwrite_zeroes(BlockDriverState *bs, int64_t offset, int64_t bytes,
++                          BdrvRequestFlags flags)
 +{
 +    BDRVZeroinitState *s = bs->opaque;
 +    if (offset >= s->extents)
@@ -156,8 +162,9 @@ index 0000000000..b60e1b84dc
 +    return bdrv_pwrite_zeroes(bs->file, offset, bytes, flags);
 +}
 +
-+static int coroutine_fn zeroinit_co_pwritev(BlockDriverState *bs,
-+    int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++static int coroutine_fn GRAPH_RDLOCK
++zeroinit_co_pwritev(BlockDriverState *bs, int64_t offset, int64_t bytes,
++                    QEMUIOVector *qiov, BdrvRequestFlags flags)
 +{
 +    BDRVZeroinitState *s = bs->opaque;
 +    int64_t extents = offset + bytes;
@@ -166,33 +173,37 @@ index 0000000000..b60e1b84dc
 +    return bdrv_co_pwritev(bs->file, offset, bytes, qiov, flags);
 +}
 +
-+static coroutine_fn int zeroinit_co_flush(BlockDriverState *bs)
++static coroutine_fn int GRAPH_RDLOCK
++zeroinit_co_flush(BlockDriverState *bs)
 +{
 +    return bdrv_co_flush(bs->file->bs);
 +}
 +
-+static int zeroinit_has_zero_init(BlockDriverState *bs)
++static int GRAPH_RDLOCK
++zeroinit_has_zero_init(BlockDriverState *bs)
 +{
 +    BDRVZeroinitState *s = bs->opaque;
 +    return s->has_zero_init;
 +}
 +
-+static int coroutine_fn zeroinit_co_pdiscard(BlockDriverState *bs,
-+                                             int64_t offset, int64_t bytes)
++static int coroutine_fn GRAPH_RDLOCK
++zeroinit_co_pdiscard(BlockDriverState *bs, int64_t offset, int64_t bytes)
 +{
 +    return bdrv_co_pdiscard(bs->file, offset, bytes);
 +}
 +
-+static int zeroinit_co_truncate(BlockDriverState *bs, int64_t offset,
-+                                _Bool exact, PreallocMode prealloc,
-+                                BdrvRequestFlags req_flags, Error **errp)
++static int GRAPH_RDLOCK
++zeroinit_co_truncate(BlockDriverState *bs, int64_t offset, _Bool exact,
++                     PreallocMode prealloc, BdrvRequestFlags req_flags,
++                     Error **errp)
 +{
 +    return bdrv_co_truncate(bs->file, offset, exact, prealloc, req_flags, errp);
 +}
 +
-+static int zeroinit_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
++static coroutine_fn int GRAPH_RDLOCK
++zeroinit_co_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
 +{
-+    return bdrv_get_info(bs->file->bs, bdi);
++    return bdrv_co_get_info(bs->file->bs, bdi);
 +}
 +
 +static BlockDriver bdrv_zeroinit = {
@@ -201,9 +212,9 @@ index 0000000000..b60e1b84dc
 +    .instance_size                    = sizeof(BDRVZeroinitState),
 +
 +    .bdrv_parse_filename              = zeroinit_parse_filename,
-+    .bdrv_file_open                   = zeroinit_open,
++    .bdrv_open                        = zeroinit_open,
 +    .bdrv_close                       = zeroinit_close,
-+    .bdrv_getlength                   = zeroinit_getlength,
++    .bdrv_co_getlength                = zeroinit_co_getlength,
 +    .bdrv_child_perm                  = bdrv_default_perms,
 +    .bdrv_co_flush_to_disk            = zeroinit_co_flush,
 +
@@ -219,7 +230,7 @@ index 0000000000..b60e1b84dc
 +    .bdrv_co_pdiscard                 = zeroinit_co_pdiscard,
 +
 +    .bdrv_co_truncate                 = zeroinit_co_truncate,
-+    .bdrv_get_info                    = zeroinit_get_info,
++    .bdrv_co_get_info                 = zeroinit_co_get_info,
 +};
 +
 +static void bdrv_zeroinit_init(void)

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

@@ -10,14 +10,14 @@ Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  qemu-options.hx | 3 +++
- softmmu/vl.c    | 8 ++++++++
+ system/vl.c     | 8 ++++++++
  2 files changed, 11 insertions(+)
 
 diff --git a/qemu-options.hx b/qemu-options.hx
-index 54efb127c4..ef456d03ec 100644
+index defee0c06a..fb980a05cf 100644
 --- a/qemu-options.hx
 +++ b/qemu-options.hx
-@@ -1147,6 +1147,9 @@ backend describes how QEMU handles the data.
+@@ -1280,6 +1280,9 @@ legacy PC, they are not recommended for modern configurations.
  
  ERST
  
@@ -27,11 +27,11 @@ index 54efb127c4..ef456d03ec 100644
  DEF("fda", HAS_ARG, QEMU_OPTION_fda,
      "-fda/-fdb file  use 'file' as floppy disk 0/1 image\n", QEMU_ARCH_ALL)
  DEF("fdb", HAS_ARG, QEMU_OPTION_fdb, "", QEMU_ARCH_ALL)
-diff --git a/softmmu/vl.c b/softmmu/vl.c
-index 21f067d115..9d737e7914 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -2643,6 +2643,7 @@ void qemu_init(int argc, char **argv)
+diff --git a/system/vl.c b/system/vl.c
+index 9b36ace6b4..452742ab58 100644
+--- a/system/vl.c
++++ b/system/vl.c
+@@ -2854,6 +2854,7 @@ void qemu_init(int argc, char **argv)
      MachineClass *machine_class;
      bool userconfig = true;
      FILE *vmstate_dump_file = NULL;
@@ -39,7 +39,7 @@ index 21f067d115..9d737e7914 100644
  
      qemu_add_opts(&qemu_drive_opts);
      qemu_add_drive_opts(&qemu_legacy_drive_opts);
-@@ -3263,6 +3264,13 @@ void qemu_init(int argc, char **argv)
+@@ -3472,6 +3473,13 @@ void qemu_init(int argc, char **argv)
                  machine_parse_property_opt(qemu_find_opts("smp-opts"),
                                             "smp", optarg);
                  break;
@@ -50,6 +50,6 @@ index 21f067d115..9d737e7914 100644
 +                    exit(1);
 +                }
 +                break;
+ #ifdef CONFIG_VNC
              case QEMU_OPTION_vnc:
                  vnc_parse(optarg);
-                 break;

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

@@ -11,10 +11,10 @@ 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 2a20982066..7968ad5a93 100644
+index 2a3e878c4d..efbed1aea3 100644
 --- a/hw/intc/apic_common.c
 +++ b/hw/intc/apic_common.c
-@@ -278,6 +278,15 @@ static void apic_reset_common(DeviceState *dev)
+@@ -263,6 +263,15 @@ static void apic_reset_common(DeviceState *dev)
      info->vapic_base_update(s);
  
      apic_init_reset(dev);

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

@@ -9,14 +9,14 @@ Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  block/file-posix.c   | 59 ++++++++++++++++++++++++++++++--------------
- qapi/block-core.json |  3 ++-
- 2 files changed, 42 insertions(+), 20 deletions(-)
+ qapi/block-core.json |  7 +++++-
+ 2 files changed, 46 insertions(+), 20 deletions(-)
 
 diff --git a/block/file-posix.c b/block/file-posix.c
-index 9a16d86344..bd68df57ad 100644
+index cfa0b832ba..d5c28cccc9 100644
 --- a/block/file-posix.c
 +++ b/block/file-posix.c
-@@ -2487,6 +2487,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+@@ -2897,6 +2897,7 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
      int fd;
      uint64_t perm, shared;
      int result = 0;
@@ -24,7 +24,7 @@ index 9a16d86344..bd68df57ad 100644
  
      /* Validate options and set default values */
      assert(options->driver == BLOCKDEV_DRIVER_FILE);
-@@ -2527,19 +2528,22 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+@@ -2937,19 +2938,22 @@ 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 9a16d86344..bd68df57ad 100644
      }
  
      /* Clear the file by truncating it to 0 */
-@@ -2593,13 +2597,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
+@@ -3003,13 +3007,15 @@ raw_co_create(BlockdevCreateOptions *options, Error **errp)
      }
  
  out_unlock:
@@ -82,7 +82,7 @@ index 9a16d86344..bd68df57ad 100644
      }
  
  out_close:
-@@ -2624,6 +2630,7 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+@@ -3033,6 +3039,7 @@ raw_co_create_opts(BlockDriver *drv, const char *filename,
      PreallocMode prealloc;
      char *buf = NULL;
      Error *local_err = NULL;
@@ -90,7 +90,7 @@ index 9a16d86344..bd68df57ad 100644
  
      /* Skip file: protocol prefix */
      strstart(filename, "file:", &filename);
-@@ -2646,6 +2653,18 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+@@ -3055,6 +3062,18 @@ raw_co_create_opts(BlockDriver *drv, const char *filename,
          return -EINVAL;
      }
  
@@ -109,7 +109,7 @@ index 9a16d86344..bd68df57ad 100644
      options = (BlockdevCreateOptions) {
          .driver     = BLOCKDEV_DRIVER_FILE,
          .u.file     = {
-@@ -2657,6 +2676,8 @@ static int coroutine_fn raw_co_create_opts(BlockDriver *drv,
+@@ -3066,6 +3085,8 @@ raw_co_create_opts(BlockDriver *drv, const char *filename,
              .nocow              = nocow,
              .has_extent_size_hint = has_extent_size_hint,
              .extent_size_hint   = extent_size_hint,
@@ -119,10 +119,21 @@ index 9a16d86344..bd68df57ad 100644
      };
      return raw_co_create(&options, errp);
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 7daaf545be..9e902b96bb 100644
+index 0e5f148d30..1c05413916 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -4624,7 +4624,8 @@
+@@ -5016,6 +5016,10 @@
+ # @extent-size-hint: Extent size hint to add to the image file; 0 for
+ #     not adding an extent size hint (default: 1 MB, since 5.1)
+ #
++# @locking: whether to enable file locking.  If set to 'auto', only
++#     enable when Open File Descriptor (OFD) locking API is available
++#     (default: auto).
++#
+ # Since: 2.12
+ ##
+ { 'struct': 'BlockdevCreateOptionsFile',
+@@ -5023,7 +5027,8 @@
              '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 6b8cfcf6d8..3ec67e32d3 100644
+index f093e256e9..78f1c8e3c8 100644
 --- a/monitor/qmp.c
 +++ b/monitor/qmp.c
-@@ -519,8 +519,7 @@ void monitor_init_qmp(Chardev *chr, bool pretty, Error **errp)
+@@ -534,8 +534,7 @@ 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 8d34caa31d..2df9037c4e 100644
+index 63c6ef93d2..9a34017e5a 100644
 --- a/hw/core/machine.c
 +++ b/hw/core/machine.c
-@@ -132,7 +132,8 @@ GlobalProperty hw_compat_4_0[] = {
+@@ -193,7 +193,8 @@ 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

@@ -11,90 +11,90 @@ and only if 'is-current').
 
 Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: adapt to QAPI changes]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- hw/core/machine-qmp-cmds.c |  6 ++++++
+ hw/core/machine-qmp-cmds.c |  5 +++++
  include/hw/boards.h        |  2 ++
- qapi/machine.json          |  4 +++-
- softmmu/vl.c               | 25 +++++++++++++++++++++++++
- 4 files changed, 36 insertions(+), 1 deletion(-)
+ qapi/machine.json          |  3 +++
+ system/vl.c                | 24 ++++++++++++++++++++++++
+ 4 files changed, 34 insertions(+)
 
 diff --git a/hw/core/machine-qmp-cmds.c b/hw/core/machine-qmp-cmds.c
-index 76fff60a6b..ec9201fb9a 100644
+index 93fb4bc24a..b9999423b4 100644
 --- a/hw/core/machine-qmp-cmds.c
 +++ b/hw/core/machine-qmp-cmds.c
-@@ -103,6 +103,12 @@ MachineInfoList *qmp_query_machines(Error **errp)
+@@ -95,6 +95,11 @@ MachineInfoList *qmp_query_machines(bool has_compat_props, bool compat_props,
          if (strcmp(mc->name, MACHINE_GET_CLASS(current_machine)->name) == 0) {
              info->has_is_current = true;
              info->is_current = true;
 +
 +            // PVE version string only exists for current machine
 +            if (mc->pve_version) {
-+                info->has_pve_version = true;
 +                info->pve_version = g_strdup(mc->pve_version);
 +            }
          }
  
-         if (mc->default_cpu_type) {
+         if (default_cpu_type) {
 diff --git a/include/hw/boards.h b/include/hw/boards.h
-index 90f1dd3aeb..14d60520d9 100644
+index f22b2e7fc7..8ada4d5832 100644
 --- a/include/hw/boards.h
 +++ b/include/hw/boards.h
-@@ -230,6 +230,8 @@ struct MachineClass {
+@@ -271,6 +271,8 @@ struct MachineClass {
      const char *desc;
      const char *deprecation_reason;
  
 +    const char *pve_version;
 +
      void (*init)(MachineState *state);
-     void (*reset)(MachineState *state, ShutdownCause reason);
+     void (*reset)(MachineState *state, ResetType type);
      void (*wakeup)(MachineState *state);
 diff --git a/qapi/machine.json b/qapi/machine.json
-index 9156103c8f..f4fb1b2c9c 100644
+index 16366b774a..12cfd3f260 100644
 --- a/qapi/machine.json
 +++ b/qapi/machine.json
-@@ -157,6 +157,8 @@
+@@ -189,6 +189,8 @@
  #
- # @default-ram-id: the default ID of initial RAM memory backend (since 5.2)
+ # @acpi: machine type supports ACPI (since 8.0)
  #
 +# @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',
+ # @compat-props: The machine type's compatibility properties.  Only
+ #     present when query-machines argument @compat-props is true.
+ #     (since 9.1)
+@@ -205,6 +207,7 @@
              'hotpluggable-cpus': 'bool',  'numa-mem-supported': 'bool',
              'deprecated': 'bool', '*default-cpu-type': 'str',
--            '*default-ram-id': 'str' } }
-+            '*default-ram-id': 'str', '*pve-version': 'str' } }
+             '*default-ram-id': 'str', 'acpi': 'bool',
++            '*pve-version': 'str',
+             '*compat-props': { 'type': ['CompatProperty'],
+                                'features': ['unstable'] } } }
  
- ##
- # @query-machines:
-diff --git a/softmmu/vl.c b/softmmu/vl.c
-index 9d737e7914..a64eee2fad 100644
---- a/softmmu/vl.c
-+++ b/softmmu/vl.c
-@@ -1578,6 +1578,7 @@ static const QEMUOption *lookup_opt(int argc, char **argv,
- static MachineClass *select_machine(QDict *qdict, Error **errp)
+diff --git a/system/vl.c b/system/vl.c
+index 452742ab58..c3707b2412 100644
+--- a/system/vl.c
++++ b/system/vl.c
+@@ -1674,6 +1674,7 @@ static MachineClass *select_machine(QDict *qdict, Error **errp)
  {
-     const char *optarg = qdict_get_try_str(qdict, "type");
+     ERRP_GUARD();
+     const char *machine_type = 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;
-@@ -1595,6 +1596,11 @@ static MachineClass *select_machine(QDict *qdict, Error **errp)
-         }
-     }
+     g_autoptr(GSList) machines = object_class_get_list(TYPE_MACHINE, false);
+     MachineClass *machine_class = NULL;
  
-+    if (machine_class) {
+@@ -1693,7 +1694,11 @@ static MachineClass *select_machine(QDict *qdict, Error **errp)
+     if (!machine_class) {
+         error_append_hint(errp,
+                           "Use -machine help to list supported machines\n");
++    } else {
 +        machine_class->pve_version = g_strdup(pvever);
 +        qdict_del(qdict, "pvever");
-+    }
+     }
 +
-     g_slist_free(machines);
-     if (local_err) {
-         error_append_hint(&local_err, "Use -machine help to list supported machines\n");
-@@ -3205,12 +3211,31 @@ void qemu_init(int argc, char **argv)
+     return machine_class;
+ }
+ 
+@@ -3414,12 +3419,31 @@ void qemu_init(int argc, char **argv)
              case QEMU_OPTION_machine:
                  {
                      bool help;

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

@@ -25,7 +25,7 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 4 insertions(+), 4 deletions(-)
 
 diff --git a/block/backup.c b/block/backup.c
-index 6a9ad97a53..9b0151c5be 100644
+index 79652bf57b..cc747e9163 100644
 --- a/block/backup.c
 +++ b/block/backup.c
 @@ -237,8 +237,8 @@ static void backup_init_bcs_bitmap(BackupBlockJob *job)
@@ -48,9 +48,9 @@ index 6a9ad97a53..9b0151c5be 100644
      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,
+@@ -502,6 +500,8 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
                         &error_abort);
+     bdrv_graph_wrunlock();
  
 +    backup_init_bcs_bitmap(job);
 +

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

@@ -3,40 +3,47 @@ From: Dietmar Maurer <dietmar@proxmox.com>
 Date: Mon, 6 Apr 2020 12:16:57 +0200
 Subject: [PATCH] PVE-Backup: add vma backup format code
 
+Notes about partial restoring: skipping a certain drive is done via a
+map line 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.
+
 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>
+[FE: improvements during create
+     allow partial restore
+     allow specifying disk formats for create operation]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  block/meson.build |   2 +
  meson.build       |   5 +
- vma-reader.c      | 859 ++++++++++++++++++++++++++++++++++++++++++++++
- vma-writer.c      | 791 ++++++++++++++++++++++++++++++++++++++++++
- vma.c             | 849 +++++++++++++++++++++++++++++++++++++++++++++
+ vma-reader.c      | 867 ++++++++++++++++++++++++++++++++++++++++++
+ vma-writer.c      | 816 ++++++++++++++++++++++++++++++++++++++++
+ vma.c             | 941 ++++++++++++++++++++++++++++++++++++++++++++++
  vma.h             | 150 ++++++++
- 6 files changed, 2656 insertions(+)
+ 6 files changed, 2781 insertions(+)
  create mode 100644 vma-reader.c
  create mode 100644 vma-writer.c
  create mode 100644 vma.c
  create mode 100644 vma.h
 
 diff --git a/block/meson.build b/block/meson.build
-index 020a89ae07..4feae20e37 100644
+index a21d9a5411..1373612c10 100644
 --- a/block/meson.build
 +++ b/block/meson.build
-@@ -46,6 +46,8 @@ block_ss.add(files(
+@@ -42,6 +42,8 @@ block_ss.add(files(
    'zeroinit.c',
- ), zstd, zlib, gnutls)
+ ), zstd, zlib)
  
 +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'))
+ system_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+ system_ss.add(files('block-ram-registrar.c'))
  
 diff --git a/meson.build b/meson.build
-index 5c6b5a1c75..e8cf7e3d78 100644
+index 8ec796d835..680ab48b9b 100644
 --- a/meson.build
 +++ b/meson.build
-@@ -1525,6 +1525,8 @@ keyutils = dependency('libkeyutils', required: false,
+@@ -2161,6 +2161,8 @@ endif
  
  has_gettid = cc.has_function('gettid')
  
@@ -45,22 +52,22 @@ index 5c6b5a1c75..e8cf7e3d78 100644
  # libselinux
  selinux = dependency('libselinux',
                       required: get_option('selinux'),
-@@ -3596,6 +3598,9 @@ if have_tools
-                dependencies: [blockdev, qemuutil, gnutls, selinux],
+@@ -4367,6 +4369,9 @@ if have_tools
+                dependencies: [blockdev, qemuutil, selinux],
                 install: true)
  
 +  vma = executable('vma', files('vma.c', 'vma-reader.c') + genh,
-+                   dependencies: [authz, block, crypto, io, qom], install: true)
++                   dependencies: [authz, block, crypto, io, qemuutil, qom], install: true)
 +
    subdir('storage-daemon')
-   subdir('contrib/rdmacm-mux')
-   subdir('contrib/elf2dmp')
+ 
+   foreach exe: [ 'qemu-img', 'qemu-io', 'qemu-nbd', 'qemu-storage-daemon']
 diff --git a/vma-reader.c b/vma-reader.c
 new file mode 100644
-index 0000000000..e65f1e8415
+index 0000000000..1888b21851
 --- /dev/null
 +++ b/vma-reader.c
-@@ -0,0 +1,859 @@
+@@ -0,0 +1,867 @@
 +/*
 + * VMA: Virtual Machine Archive
 + *
@@ -82,7 +89,8 @@ index 0000000000..e65f1e8415
 +#include "qemu/ratelimit.h"
 +#include "vma.h"
 +#include "block/block.h"
-+#include "sysemu/block-backend.h"
++#include "block/graph-lock.h"
++#include "system/block-backend.h"
 +
 +static unsigned char zero_vma_block[VMA_BLOCK_SIZE];
 +
@@ -91,6 +99,7 @@ index 0000000000..e65f1e8415
 +    bool write_zeroes;
 +    unsigned long *bitmap;
 +    int bitmap_size;
++    bool skip;
 +}  VmaRestoreState;
 +
 +struct VmaReader {
@@ -373,7 +382,6 @@ index 0000000000..e65f1e8415
 +    }
 +
 +
-+    int count = 0;
 +    for (i = 1; i < 256; i++) {
 +        VmaDeviceInfoHeader *dih = &h->dev_info[i];
 +        uint32_t devname_ptr = GUINT32_FROM_BE(dih->devname_ptr);
@@ -381,7 +389,6 @@ index 0000000000..e65f1e8415
 +        const char *devname =  get_header_str(vmar, devname_ptr);
 +
 +        if (size && devname) {
-+            count++;
 +            vmar->devinfo[i].size = size;
 +            vmar->devinfo[i].devname = devname;
 +
@@ -488,13 +495,14 @@ index 0000000000..e65f1e8415
 +}
 +
 +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;
 +
@@ -509,28 +517,30 @@ index 0000000000..e65f1e8415
 +}
 +
 +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);
++    assert(vmar->rstate[dev_id].target == NULL && !vmar->rstate[dev_id].skip);
 +
-+    int64_t size = blk_getlength(target);
-+    int64_t size_diff = size - vmar->devinfo[dev_id].size;
++    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;
++        /* 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;
 +}
@@ -590,8 +600,10 @@ index 0000000000..e65f1e8415
 +    } else {
 +        int res = blk_pwrite(target, sector_num * BDRV_SECTOR_SIZE, nb_sectors * BDRV_SECTOR_SIZE, buf, 0);
 +        if (res < 0) {
++            bdrv_graph_rdlock_main_loop();
 +            error_setg(errp, "blk_pwrite to %s failed (%d)",
 +                       bdrv_get_device_name(blk_bs(target)), res);
++            bdrv_graph_rdunlock_main_loop();
 +            return -1;
 +        }
 +    }
@@ -623,19 +635,23 @@ index 0000000000..e65f1e8415
 +        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 {
@@ -681,7 +697,7 @@ index 0000000000..e65f1e8415
 +                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,
@@ -717,7 +733,7 @@ index 0000000000..e65f1e8415
 +                        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,
@@ -742,7 +758,7 @@ index 0000000000..e65f1e8415
 +                            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) {
@@ -867,8 +883,7 @@ index 0000000000..e65f1e8415
 +
 +            int64_t cluster_num, end;
 +
-+            end = (vmar->devinfo[i].size + VMA_CLUSTER_SIZE - 1) /
-+                VMA_CLUSTER_SIZE;
++            end = DIV_ROUND_UP(vmar->devinfo[i].size, VMA_CLUSTER_SIZE);
 +
 +            for (cluster_num = 0; cluster_num < end; cluster_num++) {
 +                if (!vma_reader_get_bitmap(rstate, cluster_num)) {
@@ -913,7 +928,7 @@ index 0000000000..e65f1e8415
 +
 +    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);
 +        }
 +    }
 +
@@ -922,10 +937,10 @@ index 0000000000..e65f1e8415
 +
 diff --git a/vma-writer.c b/vma-writer.c
 new file mode 100644
-index 0000000000..df4b20793d
+index 0000000000..3f489092df
 --- /dev/null
 +++ b/vma-writer.c
-@@ -0,0 +1,791 @@
+@@ -0,0 +1,816 @@
 +/*
 + * VMA: Virtual Machine Archive
 + *
@@ -941,6 +956,8 @@ index 0000000000..df4b20793d
 +
 +#include "qemu/osdep.h"
 +#include <glib.h>
++#include <linux/magic.h>
++#include <sys/vfs.h>
 +#include <uuid/uuid.h>
 +
 +#include "vma.h"
@@ -949,6 +966,7 @@ index 0000000000..df4b20793d
 +#include "qemu/main-loop.h"
 +#include "qemu/coroutine.h"
 +#include "qemu/cutils.h"
++#include "qemu/error-report.h"
 +#include "qemu/memalign.h"
 +
 +#define DEBUG_VMA 0
@@ -1116,8 +1134,7 @@ index 0000000000..df4b20793d
 +    vmaw->stream_info[n].devname = g_strdup(devname);
 +    vmaw->stream_info[n].size = size;
 +
-+    vmaw->stream_info[n].cluster_count = (size + VMA_CLUSTER_SIZE - 1) /
-+        VMA_CLUSTER_SIZE;
++    vmaw->stream_info[n].cluster_count = DIV_ROUND_UP(size, VMA_CLUSTER_SIZE);
 +
 +    vmaw->stream_count = n;
 +
@@ -1132,10 +1149,10 @@ index 0000000000..df4b20793d
 +{
 +    assert(qemu_in_coroutine());
 +    AioContext *ctx = qemu_get_current_aio_context();
-+    aio_set_fd_handler(ctx, fd, false, NULL, (IOHandler *)qemu_coroutine_enter,
-+                       NULL, NULL, qemu_coroutine_self());
++    aio_set_fd_handler(ctx, fd, NULL, (IOHandler *)qemu_coroutine_enter, NULL,
++                       NULL, qemu_coroutine_self());
 +    qemu_coroutine_yield();
-+    aio_set_fd_handler(ctx, fd, false, NULL, NULL, NULL, NULL, NULL);
++    aio_set_fd_handler(ctx, fd, NULL, NULL, NULL, NULL, NULL);
 +}
 +
 +static ssize_t coroutine_fn
@@ -1184,6 +1201,23 @@ index 0000000000..df4b20793d
 +    return (done == bytes) ? bytes : -1;
 +}
 +
++static bool is_path_tmpfs(const char *path) {
++    struct statfs fs;
++    int ret;
++
++    do {
++        ret = statfs(path, &fs);
++    } while (ret != 0 && errno == EINTR);
++
++    if (ret != 0) {
++        warn_report("statfs call for %s failed, assuming not tmpfs - %s\n",
++                    path, strerror(errno));
++        return false;
++    }
++
++    return fs.f_type == TMPFS_MAGIC;
++}
++
 +VmaWriter *vma_writer_create(const char *filename, uuid_t uuid, Error **errp)
 +{
 +    const char *p;
@@ -1233,12 +1267,19 @@ index 0000000000..df4b20793d
 +            }
 +            /* try to use O_NONBLOCK */
 +            fcntl(vmaw->fd, F_SETFL, fcntl(vmaw->fd, F_GETFL)|O_NONBLOCK);
-+        } else  {
-+            oflags = O_NONBLOCK|O_DIRECT|O_WRONLY|O_EXCL;
++        } else {
++            gchar *dirname = g_path_get_dirname(filename);
++            oflags = O_NONBLOCK|O_WRONLY|O_EXCL;
++            if (!is_path_tmpfs(dirname)) {
++                oflags |= O_DIRECT;
++            }
++            g_free(dirname);
 +            vmaw->fd = qemu_create(filename, oflags, 0644, errp);
 +        }
 +
 +        if (vmaw->fd < 0) {
++            error_free(*errp);
++            *errp = NULL;
 +            error_setg(errp, "can't open file %s - %s\n", filename,
 +                       g_strerror(errno));
 +            goto err;
@@ -1473,17 +1514,16 @@ index 0000000000..df4b20793d
 +    int i;
 +
 +    g_assert(vmaw != NULL);
++    g_assert(status != NULL);
 +
-+    if (status) {
-+        status->status = vmaw->status;
-+        g_strlcpy(status->errmsg, vmaw->errmsg, sizeof(status->errmsg));
-+        for (i = 0; i <= 255; i++) {
-+            status->stream_info[i] = vmaw->stream_info[i];
-+        }
-+
-+        uuid_unparse_lower(vmaw->uuid, status->uuid_str);
++    status->status = vmaw->status;
++    g_strlcpy(status->errmsg, vmaw->errmsg, sizeof(status->errmsg));
++    for (i = 0; i <= 255; i++) {
++        status->stream_info[i] = vmaw->stream_info[i];
 +    }
 +
++    uuid_unparse_lower(vmaw->uuid, status->uuid_str);
++
 +    status->closed = vmaw->closed;
 +
 +    return vmaw->status;
@@ -1719,10 +1759,10 @@ index 0000000000..df4b20793d
 +}
 diff --git a/vma.c b/vma.c
 new file mode 100644
-index 0000000000..e8dffb43e0
+index 0000000000..0e990b5e30
 --- /dev/null
 +++ b/vma.c
-@@ -0,0 +1,849 @@
+@@ -0,0 +1,941 @@
 +/*
 + * VMA: Virtual Machine Archive
 + *
@@ -1745,8 +1785,8 @@ index 0000000000..e8dffb43e0
 +#include "qemu/main-loop.h"
 +#include "qemu/cutils.h"
 +#include "qemu/memalign.h"
-+#include "qapi/qmp/qdict.h"
-+#include "sysemu/block-backend.h"
++#include "qobject/qdict.h"
++#include "system/block-backend.h"
 +
 +static void help(void)
 +{
@@ -1754,9 +1794,9 @@ index 0000000000..e8dffb43e0
 +        "usage: vma command [command options]\n"
 +        "\n"
 +        "vma list <filename>\n"
-+        "vma config <filename> [-c config]\n"
-+        "vma create <filename> [-c config] pathname ...\n"
-+        "vma extract <filename> [-r <fifo>] <targetdir>\n"
++        "vma config <filename> [-c <config>]\n"
++        "vma create <filename> [-c <config>] [-d format=<format>:<device name>=<path> [-d ...]]\n"
++        "vma extract <filename> [-d <drive-list>] [-r <fifo>] <targetdir>\n"
 +        "vma verify <filename> [-v]\n"
 +        ;
 +
@@ -1863,6 +1903,7 @@ index 0000000000..e8dffb43e0
 +    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) {
@@ -1900,9 +1941,10 @@ index 0000000000..e8dffb43e0
 +    const char *filename;
 +    const char *dirname;
 +    const char *readmap = NULL;
++    gchar **drive_list = NULL;
 +
 +    for (;;) {
-+        c = getopt(argc, argv, "hvr:");
++        c = getopt(argc, argv, "hvd:r:");
 +        if (c == -1) {
 +            break;
 +        }
@@ -1911,6 +1953,9 @@ index 0000000000..e8dffb43e0
 +        case 'h':
 +            help();
 +            break;
++        case 'd':
++            drive_list = g_strsplit(optarg, ",", 254);
++            break;
 +        case 'r':
 +            readmap = optarg;
 +            break;
@@ -1970,74 +2015,89 @@ index 0000000000..e8dffb43e0
 +            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;
 +                }
-+            }
-+
-+            uint64_t bps_value = 0;
-+            if (bps) {
-+                bps_value = verify_u64(bps);
-+                g_free(bps);
-+            }
-+
-+            const char *path;
-+            bool write_zero;
-+            if (line[0] == '0' && line[1] == ':') {
-+                path = line + 2;
-+                write_zero = false;
-+            } else if (line[0] == '1' && line[1] == ':') {
-+                path = line + 2;
-+                write_zero = true;
 +            } else {
-+                g_error("read map failed - parse error ('%s')", inbuf);
++                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 (bps) {
++                    bps_value = verify_u64(bps);
++                    g_free(bps);
++                }
++
++                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);
 +            }
 +
-+            RestoreMap *map = g_new0(RestoreMap, 1);
-+            map->devname = g_strdup(devname);
-+            map->path = g_strdup(path);
-+            map->format = format;
-+            map->throttling_bps = bps_value;
-+            map->throttling_group = group;
-+            map->cache = cache;
-+            map->write_zero = write_zero;
++            RestoreMap *restore_map = g_new0(RestoreMap, 1);
++            restore_map->devname = g_strdup(devname);
++            restore_map->path = g_strdup(path);
++            restore_map->format = format;
++            restore_map->throttling_bps = bps_value;
++            restore_map->throttling_group = group;
++            restore_map->cache = cache;
++            restore_map->write_zero = write_zero;
++            restore_map->skip = skip;
 +
-+            g_hash_table_insert(devmap, map->devname, map);
++            g_hash_table_insert(devmap, restore_map->devname, restore_map);
 +
 +        };
 +    }
 +
 +    int i;
 +    int vmstate_fd = -1;
-+    guint8 vmstate_stream = 0;
++    bool drive_rename_bitmap[255];
++    memset(drive_rename_bitmap, 0, sizeof(drive_rename_bitmap));
 +
 +    for (i = 1; i < 255; i++) {
 +        VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
 +        if (di && (strcmp(di->devname, "vmstate") == 0)) {
-+            vmstate_stream = i;
 +            char *statefn = g_strdup_printf("%s/vmstate.bin", dirname);
 +            vmstate_fd = open(statefn, O_WRONLY|O_CREAT|O_EXCL, 0644);
 +            if (vmstate_fd < 0) {
@@ -2053,10 +2113,25 @@ index 0000000000..e8dffb43e0
 +            const char *cache = NULL;
 +            int flags = BDRV_O_RDWR;
 +            bool write_zero = true;
++            bool skip = false;
 +
 +            BlockBackend *blk = NULL;
 +
-+            if (readmap) {
++            if (drive_list) {
++                skip = true;
++                int j;
++                for (j = 0; drive_list[j]; j++) {
++                    if (strcmp(drive_list[j], di->devname) == 0) {
++                        skip = false;
++                        drive_rename_bitmap[i] = true;
++                        break;
++                    }
++                }
++            } else {
++                drive_rename_bitmap[i] = true;
++            }
++
++            if (!skip && readmap) {
 +                RestoreMap *map;
 +                map = (RestoreMap *)g_hash_table_lookup(devmap, di->devname);
 +                if (map == NULL) {
@@ -2068,7 +2143,8 @@ index 0000000000..e8dffb43e0
 +                throttling_group = map->throttling_group;
 +                cache = map->cache;
 +                write_zero = map->write_zero;
-+            } else {
++                skip = map->skip;
++            } else if (!skip) {
 +                devfn = g_strdup_printf("%s/tmp-disk-%s.raw",
 +                                        dirname, di->devname);
 +                printf("DEVINFO %s %zd\n", devfn, di->size);
@@ -2086,57 +2162,60 @@ index 0000000000..e8dffb43e0
 +                write_zero = false;
 +            }
 +
-+	    size_t devlen = strlen(devfn);
-+	    QDict *options = NULL;
-+            bool writethrough;
-+            if (format) {
-+                /* explicit format from commandline */
-+                options = qdict_new();
-+                qdict_put_str(options, "driver", format);
-+            } else if ((devlen > 4 && strcmp(devfn+devlen-4, ".raw") == 0) ||
-+	               strncmp(devfn, "/dev/", 5) == 0)
-+	    {
-+                /* This part is now deprecated for PVE as well (just as qemu
-+                 * deprecated not specifying an explicit raw format, too.
-+                 */
-+		/* explicit raw format */
-+		options = qdict_new();
-+		qdict_put_str(options, "driver", "raw");
-+	    }
-+            if (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
-+                g_error("invalid cache option: %s\n", cache);
-+            }
-+
-+	    if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
-+                g_error("can't open file %s - %s", devfn,
-+                        error_get_pretty(errp));
-+            }
-+
-+            if (cache) {
-+                blk_set_enable_write_cache(blk, !writethrough);
-+            }
-+
-+            if (throttling_group) {
-+                blk_io_limits_enable(blk, throttling_group);
-+            }
-+
-+            if (throttling_bps) {
-+                if (!throttling_group) {
-+                    blk_io_limits_enable(blk, devfn);
++            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");
 +                }
 +
-+                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 (cache && bdrv_parse_cache_mode(cache, &flags, &writethrough)) {
++                    g_error("invalid cache option: %s\n", cache);
++                }
++
++                if (errp || !(blk = blk_new_open(devfn, NULL, options, flags, &errp))) {
++                    g_error("can't open file %s - %s", devfn,
++                            error_get_pretty(errp));
++                }
++
++                if (cache) {
++                    blk_set_enable_write_cache(blk, !writethrough);
++                }
++
++                if (throttling_group) {
++                    blk_io_limits_enable(blk, throttling_group);
++                }
++
++                if (throttling_bps) {
++                    if (!throttling_group) {
++                        blk_io_limits_enable(blk, devfn);
++                    }
++
++                    ThrottleConfig cfg;
++                    throttle_config_init(&cfg);
++                    cfg.buckets[THROTTLE_BPS_WRITE].avg = throttling_bps;
++                    Error *err = NULL;
++                    if (!throttle_is_valid(&cfg, &err)) {
++                        error_report_err(err);
++                        g_error("failed to apply throttling");
++                    }
++                    blk_set_io_limits(blk, &cfg);
 +                }
-+                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));
 +            }
 +
@@ -2146,6 +2225,10 @@ index 0000000000..e8dffb43e0
 +        }
 +    }
 +
++    if (drive_list) {
++        g_strfreev(drive_list);
++    }
++
 +    if (vma_reader_restore(vmar, vmstate_fd, verbose, &errp) < 0) {
 +        g_error("restore failed - %s", error_get_pretty(errp));
 +    }
@@ -2153,7 +2236,7 @@ index 0000000000..e8dffb43e0
 +    if (!readmap) {
 +        for (i = 1; i < 255; i++) {
 +            VmaDeviceInfo *di = vma_reader_get_device_info(vmar, i);
-+            if (di && (i != vmstate_stream)) {
++            if (di && drive_rename_bitmap[i]) {
 +                char *tmpfn = g_strdup_printf("%s/tmp-disk-%s.raw",
 +                                              dirname, di->devname);
 +                char *fn = g_strdup_printf("%s/disk-%s.raw",
@@ -2252,7 +2335,7 @@ index 0000000000..e8dffb43e0
 +    struct iovec iov;
 +    QEMUIOVector qiov;
 +
-+    int64_t start, end;
++    int64_t start, end, readlen;
 +    int ret = 0;
 +
 +    unsigned char *buf = blk_blockalign(job->target, VMA_CLUSTER_SIZE);
@@ -2266,16 +2349,24 @@ index 0000000000..e8dffb43e0
 +        iov.iov_len = VMA_CLUSTER_SIZE;
 +        qemu_iovec_init_external(&qiov, &iov, 1);
 +
++        if (start + 1 == end) {
++            memset(buf, 0, VMA_CLUSTER_SIZE);
++            readlen = job->len - start * VMA_CLUSTER_SIZE;
++            assert(readlen > 0 && readlen <= VMA_CLUSTER_SIZE);
++        } else {
++            readlen = VMA_CLUSTER_SIZE;
++        }
++
 +        ret = blk_co_preadv(job->target, start * VMA_CLUSTER_SIZE,
-+                            VMA_CLUSTER_SIZE, &qiov, 0);
++                            readlen, &qiov, 0);
 +        if (ret < 0) {
-+            vma_writer_set_error(job->vmaw, "read error", -1);
++            vma_writer_set_error(job->vmaw, "read error");
 +            goto out;
 +        }
 +
 +        size_t zb = 0;
 +        if (vma_writer_write(job->vmaw, job->dev_id, start, buf, &zb) < 0) {
-+            vma_writer_set_error(job->vmaw, "backup_dump_cb vma_writer_write failed", -1);
++            vma_writer_set_error(job->vmaw, "backup_dump_cb vma_writer_write failed");
 +            goto out;
 +        }
 +    }
@@ -2293,14 +2384,16 @@ index 0000000000..e8dffb43e0
 +
 +static int create_archive(int argc, char **argv)
 +{
-+    int i, c;
++    int c;
 +    int verbose = 0;
++    bool expect_format = true;
 +    const char *archivename;
 +    GList *backup_coroutines = NULL;
 +    GList *config_files = NULL;
++    GList *disk_infos = NULL;
 +
 +    for (;;) {
-+        c = getopt(argc, argv, "hvc:");
++        c = getopt(argc, argv, "hvc:d:");
 +        if (c == -1) {
 +            break;
 +        }
@@ -2312,6 +2405,9 @@ index 0000000000..e8dffb43e0
 +        case 'c':
 +            config_files = g_list_append(config_files, optarg);
 +            break;
++        case 'd':
++            disk_infos = g_list_append(disk_infos, optarg);
++            break;
 +        case 'v':
 +            verbose = 1;
 +            break;
@@ -2357,16 +2453,48 @@ index 0000000000..e8dffb43e0
 +        l = g_list_next(l);
 +    }
 +
-+    int devcount = 0;
++    /*
++     * Don't allow mixing new and old way to specifiy disks.
++     * TODO PVE 9 drop old way and always require format.
++     */
++    if (optind < argc && g_list_first(disk_infos)) {
++        unlink(archivename);
++        g_error("Unexpected extra argument - specify all devices via '-d'");
++    }
++
 +    while (optind < argc) {
-+        const char *path = argv[optind++];
++        expect_format = false;
++        disk_infos = g_list_append(disk_infos, argv[optind++]);
++    }
++
++    int devcount = 0;
++    GList *disk_l = disk_infos;
++    while (disk_l && disk_l->data) {
++        char *disk_info = disk_l->data;
++        const char *path = NULL;
 +        char *devname = NULL;
-+        path = extract_devname(path, &devname, devcount++);
++        char *format = NULL;
++        QDict *options = qdict_new();
++
++        if (try_parse_option(&disk_info, "format", &format, disk_info)) {
++            qdict_put_str(options, "driver", format);
++        } else {
++            if (expect_format) {
++                unlink(archivename);
++                g_error("No format specified for device: '%s'", disk_info);
++            } else {
++                g_warning("Specifying a device without a format is deprecated"
++                          " - use '-d format=<format>:%s'",
++                          disk_info);
++            }
++        }
++
++        path = extract_devname(disk_info, &devname, devcount++);
 +
 +        Error *errp = NULL;
 +        BlockBackend *target;
 +
-+        target = blk_new_open(path, NULL, NULL, 0, &errp);
++        target = blk_new_open(path, NULL, options, 0, &errp);
 +        if (!target) {
 +            unlink(archivename);
 +            g_error("bdrv_open '%s' failed - %s", path, error_get_pretty(errp));
@@ -2388,6 +2516,8 @@ index 0000000000..e8dffb43e0
 +        // 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);
++
++        disk_l = g_list_next(disk_l);
 +    }
 +
 +    VmaStatus vmastat;
@@ -2451,6 +2581,7 @@ index 0000000000..e8dffb43e0
 +    vma_writer_get_status(vmaw, &vmastat);
 +
 +    if (verbose) {
++        int i;
 +        for (i = 0; i < 256; i++) {
 +            VmaStreamInfo *si = &vmastat.stream_info[i];
 +            if (si->size) {
@@ -2468,6 +2599,7 @@ index 0000000000..e8dffb43e0
 +
 +    g_list_free(backup_coroutines);
 +    g_list_free(config_files);
++    g_list_free(disk_infos);
 +    vma_writer_destroy(vmaw);
 +    return 0;
 +}
@@ -2574,7 +2706,7 @@ index 0000000000..e8dffb43e0
 +}
 diff --git a/vma.h b/vma.h
 new file mode 100644
-index 0000000000..c895c97f6d
+index 0000000000..86d2873aa5
 --- /dev/null
 +++ b/vma.h
 @@ -0,0 +1,150 @@
@@ -2712,7 +2844,7 @@ index 0000000000..c895c97f6d
 +int coroutine_fn vma_writer_flush_output(VmaWriter *vmaw);
 +
 +int vma_writer_get_status(VmaWriter *vmaw, VmaStatus *status);
-+void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...);
++void vma_writer_set_error(VmaWriter *vmaw, const char *fmt, ...) G_GNUC_PRINTF(2, 3);
 +
 +
 +VmaReader *vma_reader_create(const char *filename, Error **errp);
@@ -2722,7 +2854,7 @@ index 0000000000..c895c97f6d
 +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/0028-PVE-Backup-add-backup-dump-block-driver.patch

@@ -9,21 +9,23 @@ Subject: [PATCH] PVE-Backup: add backup-dump block driver
 - job.c: make job_should_pause non-static
 
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: adapt to coroutine changes]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- block/backup-dump.c              | 167 +++++++++++++++++++++++++++++++
+ block/backup-dump.c              | 172 +++++++++++++++++++++++++++++++
  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(-)
+ 5 files changed, 218 insertions(+), 23 deletions(-)
  create mode 100644 block/backup-dump.c
 
 diff --git a/block/backup-dump.c b/block/backup-dump.c
 new file mode 100644
-index 0000000000..04718a94e2
+index 0000000000..354593bc10
 --- /dev/null
 +++ b/block/backup-dump.c
-@@ -0,0 +1,167 @@
+@@ -0,0 +1,172 @@
 +/*
 + * BlockDriver to send backup data stream to a callback function
 + *
@@ -35,6 +37,8 @@ index 0000000000..04718a94e2
 + */
 +
 +#include "qemu/osdep.h"
++
++#include "qobject/qdict.h"
 +#include "qom/object_interfaces.h"
 +#include "block/block_int.h"
 +
@@ -45,7 +49,8 @@ index 0000000000..04718a94e2
 +    void           *dump_cb_data;
 +} BDRVBackupDumpState;
 +
-+static int qemu_backup_dump_get_info(BlockDriverState *bs, BlockDriverInfo *bdi)
++static coroutine_fn int qemu_backup_dump_co_get_info(BlockDriverState *bs,
++                                                     BlockDriverInfo *bdi)
 +{
 +    BDRVBackupDumpState *s = bs->opaque;
 +
@@ -86,7 +91,7 @@ index 0000000000..04718a94e2
 +    /* Nothing to do. */
 +}
 +
-+static int64_t qemu_backup_dump_getlength(BlockDriverState *bs)
++static coroutine_fn int64_t qemu_backup_dump_co_getlength(BlockDriverState *bs)
 +{
 +    BDRVBackupDumpState *s = bs->opaque;
 +
@@ -146,8 +151,8 @@ index 0000000000..04718a94e2
 +
 +    .bdrv_close                   = qemu_backup_dump_close,
 +    .bdrv_has_zero_init           = bdrv_has_zero_init_1,
-+    .bdrv_getlength               = qemu_backup_dump_getlength,
-+    .bdrv_get_info                = qemu_backup_dump_get_info,
++    .bdrv_co_getlength            = qemu_backup_dump_co_getlength,
++    .bdrv_co_get_info             = qemu_backup_dump_co_get_info,
 +
 +    .bdrv_co_writev               = qemu_backup_dump_co_writev,
 +
@@ -166,7 +171,7 @@ index 0000000000..04718a94e2
 +block_init(bdrv_backup_dump_init);
 +
 +
-+BlockDriverState *bdrv_backup_dump_create(
++BlockDriverState *coroutine_fn bdrv_co_backup_dump_create(
 +    int dump_cb_block_size,
 +    uint64_t byte_size,
 +    BackupDumpFunc *dump_cb,
@@ -174,9 +179,11 @@ index 0000000000..04718a94e2
 +    Error **errp)
 +{
 +    BDRVBackupDumpState *state;
-+    BlockDriverState *bs = bdrv_new_open_driver(
-+        &bdrv_backup_dump_drive, NULL, BDRV_O_RDWR, errp);
 +
++    QDict *options = qdict_new();
++    qdict_put_str(options, "driver", "backup-dump-drive");
++
++    BlockDriverState *bs = bdrv_co_open(NULL, NULL, options, BDRV_O_RDWR, errp);
 +    if (!bs) {
 +        return NULL;
 +    }
@@ -192,7 +199,7 @@ index 0000000000..04718a94e2
 +    return bs;
 +}
 diff --git a/block/backup.c b/block/backup.c
-index 9b0151c5be..6e8f6e67b3 100644
+index cc747e9163..6f7c45f922 100644
 --- a/block/backup.c
 +++ b/block/backup.c
 @@ -29,28 +29,6 @@
@@ -224,7 +231,7 @@ index 9b0151c5be..6e8f6e67b3 100644
  static const BlockJobDriver backup_job_driver;
  
  static void backup_cleanup_sync_bitmap(BackupBlockJob *job, int ret)
-@@ -454,6 +432,14 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
+@@ -462,6 +440,14 @@ BlockJob *backup_job_create(const char *job_id, BlockDriverState *bs,
      }
  
      cluster_size = block_copy_cluster_size(bcs);
@@ -240,7 +247,7 @@ index 9b0151c5be..6e8f6e67b3 100644
      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
+index 1373612c10..6278c4af0f 100644
 --- a/block/meson.build
 +++ b/block/meson.build
 @@ -4,6 +4,7 @@ block_ss.add(files(
@@ -248,28 +255,28 @@ index 4feae20e37..0d7023fc82 100644
    'amend.c',
    'backup.c',
 +  'backup-dump.c',
-   'copy-before-write.c',
    'blkdebug.c',
    'blklogwrites.c',
+   'blkverify.c',
 diff --git a/include/block/block_int-common.h b/include/block/block_int-common.h
-index 31ae91e56e..37b64bcd93 100644
+index ebb4e56a50..e717a74e5f 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/aio.h"
+ #include "block/block-common.h"
 +#include "block/block-copy.h"
- #include "block/aio-wait.h"
- #include "qemu/queue.h"
- #include "qemu/coroutine.h"
-@@ -64,6 +65,40 @@
+ #include "block/block-global-state.h"
+ #include "block/snapshot.h"
+ #include "qemu/iov.h"
+@@ -60,6 +61,40 @@
  
  #define BLOCK_PROBE_BUF_SIZE        512
  
 +typedef int BackupDumpFunc(void *opaque, uint64_t offset, uint64_t bytes, const void *buf);
 +
-+BlockDriverState *bdrv_backup_dump_create(
++BlockDriverState *coroutine_fn bdrv_co_backup_dump_create(
 +    int dump_cb_block_size,
 +    uint64_t byte_size,
 +    BackupDumpFunc *dump_cb,
@@ -305,10 +312,10 @@ index 31ae91e56e..37b64bcd93 100644
      BDRV_TRACKED_READ,
      BDRV_TRACKED_WRITE,
 diff --git a/job.c b/job.c
-index 72d57f0934..93e22d180b 100644
+index 0653bc2ba6..b981070ee8 100644
 --- a/job.c
 +++ b/job.c
-@@ -330,7 +330,8 @@ static bool job_started_locked(Job *job)
+@@ -337,7 +337,8 @@ static bool job_started_locked(Job *job)
  }
  
  /* Called with job_mutex held. */

debian/patches/pve/0029-PVE-Add-sequential-job-transaction-support.patch

@@ -11,10 +11,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  2 files changed, 46 insertions(+)
 
 diff --git a/include/qemu/job.h b/include/qemu/job.h
-index e502787dd8..963cf2bef5 100644
+index a5a04155ea..562cc7eaec 100644
 --- a/include/qemu/job.h
 +++ b/include/qemu/job.h
-@@ -381,6 +381,18 @@ void job_unlock(void);
+@@ -362,6 +362,18 @@ void job_unlock(void);
   */
  JobTxn *job_txn_new(void);
  
@@ -34,10 +34,10 @@ index e502787dd8..963cf2bef5 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 93e22d180b..2b31f1e14f 100644
+index b981070ee8..f4646866ec 100644
 --- a/job.c
 +++ b/job.c
-@@ -93,6 +93,8 @@ struct JobTxn {
+@@ -94,6 +94,8 @@ struct JobTxn {
  
      /* Reference count */
      int refcnt;
@@ -46,7 +46,7 @@ index 93e22d180b..2b31f1e14f 100644
  };
  
  void job_lock(void)
-@@ -118,6 +120,25 @@ JobTxn *job_txn_new(void)
+@@ -119,6 +121,25 @@ JobTxn *job_txn_new(void)
      return txn;
  }
  
@@ -72,7 +72,7 @@ index 93e22d180b..2b31f1e14f 100644
  /* Called with job_mutex held. */
  static void job_txn_ref_locked(JobTxn *txn)
  {
-@@ -1057,6 +1078,12 @@ static void job_completed_txn_success_locked(Job *job)
+@@ -1048,6 +1069,12 @@ static void job_completed_txn_success_locked(Job *job)
       */
      QLIST_FOREACH(other_job, &txn->jobs, txn_list) {
          if (!job_is_completed_locked(other_job)) {
@@ -85,7 +85,7 @@ index 93e22d180b..2b31f1e14f 100644
              return;
          }
          assert(other_job->ret == 0);
-@@ -1268,6 +1295,13 @@ int job_finish_sync_locked(Job *job,
+@@ -1259,6 +1286,13 @@ int job_finish_sync_locked(Job *job,
          return -EBUSY;
      }
  

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

@@ -1,452 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 29 Jun 2020 11:06:03 +0200
-Subject: [PATCH] PVE-Backup: Add dirty-bitmap tracking for incremental backups
-
-Uses QEMU's existing MIRROR_SYNC_MODE_BITMAP and a dirty-bitmap on top
-of all backed-up drives. This will only execute the data-write callback
-for any changed chunks, the PBS rust code will reuse chunks from the
-previous index for everything it doesn't receive if reuse_index is true.
-
-On error or cancellation, remove all dirty bitmaps to ensure
-consistency.
-
-Add PBS/incremental specific information to query backup info QMP and
-HMP commands.
-
-Only supported for PBS backups.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Dietmar Maurer <dietmar@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- block/monitor/block-hmp-cmds.c |   1 +
- monitor/hmp-cmds.c             |  45 ++++++++++----
- proxmox-backup-client.c        |   3 +-
- proxmox-backup-client.h        |   1 +
- pve-backup.c                   | 103 ++++++++++++++++++++++++++++++---
- qapi/block-core.json           |  12 +++-
- 6 files changed, 142 insertions(+), 23 deletions(-)
-
-diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
-index 477044c54a..556af25861 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1042,6 +1042,7 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
-         false, NULL, // PBS fingerprint
-         false, NULL, // PBS backup-id
-         false, 0, // PBS backup-time
-+        false, false, // PBS incremental
-         true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
-         false, NULL, false, NULL, !!devlist,
-         devlist, qdict_haskey(qdict, "speed"), speed, &error);
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index a40b25e906..670f783515 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -225,19 +225,42 @@ void hmp_info_backup(Monitor *mon, const QDict *qdict)
-             monitor_printf(mon, "End time: %s", ctime(&info->end_time));
-         }
- 
--        int per = (info->has_total && info->total &&
--            info->has_transferred && info->transferred) ?
--            (info->transferred * 100)/info->total : 0;
--        int zero_per = (info->has_total && info->total &&
--                        info->has_zero_bytes && info->zero_bytes) ?
--            (info->zero_bytes * 100)/info->total : 0;
-         monitor_printf(mon, "Backup file: %s\n", info->backup_file);
-         monitor_printf(mon, "Backup uuid: %s\n", info->uuid);
--        monitor_printf(mon, "Total size: %zd\n", info->total);
--        monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
--                       info->transferred, per);
--        monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
--                       info->zero_bytes, zero_per);
-+
-+        if (!(info->has_total && info->total))  {
-+            // this should not happen normally
-+            monitor_printf(mon, "Total size: %d\n", 0);
-+        } else {
-+            bool incremental = false;
-+            size_t total_or_dirty = info->total;
-+            if (info->has_transferred) {
-+                if (info->has_dirty && info->dirty) {
-+                     if (info->dirty < info->total) {
-+                        total_or_dirty = info->dirty;
-+                        incremental = true;
-+                    }
-+                }
-+            }
-+
-+            int per = (info->transferred * 100)/total_or_dirty;
-+
-+            monitor_printf(mon, "Backup mode: %s\n", incremental ? "incremental" : "full");
-+
-+            int zero_per = (info->has_zero_bytes && info->zero_bytes) ?
-+                (info->zero_bytes * 100)/info->total : 0;
-+            monitor_printf(mon, "Total size: %zd\n", info->total);
-+            monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
-+                           info->transferred, per);
-+            monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
-+                           info->zero_bytes, zero_per);
-+
-+            if (info->has_reused) {
-+                int reused_per = (info->reused * 100)/total_or_dirty;
-+                monitor_printf(mon, "Reused bytes: %zd (%d%%)\n",
-+                               info->reused, reused_per);
-+            }
-+        }
-     }
- 
-     qapi_free_BackupStatus(info);
-diff --git a/proxmox-backup-client.c b/proxmox-backup-client.c
-index a8f6653a81..4ce7bc0b5e 100644
---- a/proxmox-backup-client.c
-+++ b/proxmox-backup-client.c
-@@ -89,6 +89,7 @@ proxmox_backup_co_register_image(
-     ProxmoxBackupHandle *pbs,
-     const char *device_name,
-     uint64_t size,
-+    bool incremental,
-     Error **errp)
- {
-     Coroutine *co = qemu_coroutine_self();
-@@ -98,7 +99,7 @@ proxmox_backup_co_register_image(
-     int pbs_res = -1;
- 
-     proxmox_backup_register_image_async(
--        pbs, device_name, size ,proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-+        pbs, device_name, size, incremental, proxmox_backup_schedule_wake, &waker, &pbs_res, &pbs_err);
-     qemu_coroutine_yield();
-     if (pbs_res < 0) {
-         if (errp) error_setg(errp, "backup register image failed: %s", pbs_err ? pbs_err : "unknown error");
-diff --git a/proxmox-backup-client.h b/proxmox-backup-client.h
-index 1dda8b7d8f..8cbf645b2c 100644
---- a/proxmox-backup-client.h
-+++ b/proxmox-backup-client.h
-@@ -32,6 +32,7 @@ proxmox_backup_co_register_image(
-     ProxmoxBackupHandle *pbs,
-     const char *device_name,
-     uint64_t size,
-+    bool incremental,
-     Error **errp);
- 
- 
-diff --git a/pve-backup.c b/pve-backup.c
-index 3d28975eaa..abd7062afe 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -28,6 +28,8 @@
-  *
-  */
- 
-+const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
-+
- static struct PVEBackupState {
-     struct {
-         // Everithing accessed from qmp_backup_query command is protected using lock
-@@ -39,7 +41,9 @@ static struct PVEBackupState {
-         uuid_t uuid;
-         char uuid_str[37];
-         size_t total;
-+        size_t dirty;
-         size_t transferred;
-+        size_t reused;
-         size_t zero_bytes;
-     } stat;
-     int64_t speed;
-@@ -66,6 +70,7 @@ typedef struct PVEBackupDevInfo {
-     uint8_t dev_id;
-     bool completed;
-     char targetfile[PATH_MAX];
-+    BdrvDirtyBitmap *bitmap;
-     BlockDriverState *target;
- } PVEBackupDevInfo;
- 
-@@ -107,11 +112,12 @@ static bool pvebackup_error_or_canceled(void)
-     return error_or_canceled;
- }
- 
--static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes)
-+static void pvebackup_add_transfered_bytes(size_t transferred, size_t zero_bytes, size_t reused)
- {
-     qemu_mutex_lock(&backup_state.stat.lock);
-     backup_state.stat.zero_bytes += zero_bytes;
-     backup_state.stat.transferred += transferred;
-+    backup_state.stat.reused += reused;
-     qemu_mutex_unlock(&backup_state.stat.lock);
- }
- 
-@@ -150,7 +156,8 @@ pvebackup_co_dump_pbs_cb(
-         pvebackup_propagate_error(local_err);
-         return pbs_res;
-     } else {
--        pvebackup_add_transfered_bytes(size, !buf ? size : 0);
-+        size_t reused = (pbs_res == 0) ? size : 0;
-+        pvebackup_add_transfered_bytes(size, !buf ? size : 0, reused);
-     }
- 
-     return size;
-@@ -210,11 +217,11 @@ pvebackup_co_dump_vma_cb(
-         } else {
-             if (remaining >= VMA_CLUSTER_SIZE) {
-                 assert(ret == VMA_CLUSTER_SIZE);
--                pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes);
-+                pvebackup_add_transfered_bytes(VMA_CLUSTER_SIZE, zero_bytes, 0);
-                 remaining -= VMA_CLUSTER_SIZE;
-             } else {
-                 assert(ret == remaining);
--                pvebackup_add_transfered_bytes(remaining, zero_bytes);
-+                pvebackup_add_transfered_bytes(remaining, zero_bytes, 0);
-                 remaining = 0;
-             }
-         }
-@@ -250,6 +257,18 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
-             if (local_err != NULL) {
-                 pvebackup_propagate_error(local_err);
-             }
-+        } else {
-+            // on error or cancel we cannot ensure synchronization of dirty
-+            // bitmaps with backup server, so remove all and do full backup next
-+            GList *l = backup_state.di_list;
-+            while (l) {
-+                PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-+                l = g_list_next(l);
-+
-+                if (di->bitmap) {
-+                    bdrv_release_dirty_bitmap(di->bitmap);
-+                }
-+            }
-         }
- 
-         proxmox_backup_disconnect(backup_state.pbs);
-@@ -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);
- 
-+    if (di->bitmap && ret < 0) {
-+        // on error or cancel we cannot ensure synchronization of dirty
-+        // bitmaps with backup server, so remove all and do full backup next
-+        bdrv_release_dirty_bitmap(di->bitmap);
-+    }
-+
-     g_free(di);
- 
-     qemu_mutex_unlock(&backup_state.backup_mutex);
-@@ -469,12 +494,18 @@ static bool create_backup_jobs(void) {
- 
-         assert(di->target != NULL);
- 
-+        MirrorSyncMode sync_mode = MIRROR_SYNC_MODE_FULL;
-+        BitmapSyncMode bitmap_mode = BITMAP_SYNC_MODE_NEVER;
-+        if (di->bitmap) {
-+            sync_mode = MIRROR_SYNC_MODE_BITMAP;
-+            bitmap_mode = BITMAP_SYNC_MODE_ON_SUCCESS;
-+        }
-         AioContext *aio_context = bdrv_get_aio_context(di->bs);
-         aio_context_acquire(aio_context);
- 
-         BlockJob *job = backup_job_create(
--            NULL, di->bs, di->target, backup_state.speed, MIRROR_SYNC_MODE_FULL, NULL,
--            BITMAP_SYNC_MODE_NEVER, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-+            NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
-+            bitmap_mode, false, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT, BLOCKDEV_ON_ERROR_REPORT,
-             JOB_DEFAULT, pvebackup_complete_cb, di, NULL, &local_err);
- 
-         aio_context_release(aio_context);
-@@ -525,6 +556,8 @@ typedef struct QmpBackupTask {
-     const char *fingerprint;
-     bool has_fingerprint;
-     int64_t backup_time;
-+    bool has_use_dirty_bitmap;
-+    bool use_dirty_bitmap;
-     bool has_format;
-     BackupFormat format;
-     bool has_config_file;
-@@ -616,6 +649,7 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-     }
- 
-     size_t total = 0;
-+    size_t dirty = 0;
- 
-     l = di_list;
-     while (l) {
-@@ -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 use_dirty_bitmap = task->has_use_dirty_bitmap && task->use_dirty_bitmap;
-+
-         char *pbs_err = NULL;
-         pbs = proxmox_backup_new(
-             task->backup_file,
-@@ -672,7 +708,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-             goto err;
-         }
- 
--        if (proxmox_backup_co_connect(pbs, task->errp) < 0)
-+        int connect_result = proxmox_backup_co_connect(pbs, task->errp);
-+        if (connect_result < 0)
-             goto err;
- 
-         /* register all devices */
-@@ -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;
-+
-+            if (use_dirty_bitmap) {
-+                if (bitmap == NULL) {
-+                    bitmap = bdrv_create_dirty_bitmap(di->bs, dump_cb_block_size, PBS_BITMAP_NAME, task->errp);
-+                    if (!bitmap) {
-+                        goto err;
-+                    }
-+                } else {
-+                    expect_only_dirty = proxmox_backup_check_incremental(pbs, devname, di->size) != 0;
-+                }
-+
-+                if (expect_only_dirty) {
-+                    dirty += bdrv_get_dirty_count(bitmap);
-+                } else {
-+                    /* mark entire bitmap as dirty to make full backup */
-+                    bdrv_set_dirty_bitmap(bitmap, 0, di->size);
-+                    dirty += di->size;
-+                }
-+                di->bitmap = bitmap;
-+            } else {
-+                dirty += di->size;
-+
-+                /* after a full backup the old dirty bitmap is invalid anyway */
-+                if (bitmap != NULL) {
-+                    bdrv_release_dirty_bitmap(bitmap);
-+                }
-+            }
-+
-+            int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
-+            if (dev_id < 0) {
-                 goto err;
-+            }
- 
-             if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
-                 goto err;
-@@ -694,6 +762,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-             di->dev_id = dev_id;
-         }
-     } else if (format == BACKUP_FORMAT_VMA) {
-+        dirty = total;
-+
-         vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
-         if (!vmaw) {
-             if (local_err) {
-@@ -721,6 +791,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-             }
-         }
-     } else if (format == BACKUP_FORMAT_DIR) {
-+        dirty = total;
-+
-         if (mkdir(task->backup_file, 0640) != 0) {
-             error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
-                              task->backup_file);
-@@ -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 = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
- 
-     qemu_mutex_unlock(&backup_state.stat.lock);
- 
-@@ -818,6 +892,10 @@ err:
-         PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-         l = g_list_next(l);
- 
-+        if (di->bitmap) {
-+            bdrv_release_dirty_bitmap(di->bitmap);
-+        }
-+
-         if (di->target) {
-             bdrv_unref(di->target);
-         }
-@@ -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_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,
-@@ -877,6 +956,8 @@ UuidInfo *qmp_backup(
-         .backup_id = backup_id,
-         .has_backup_time = has_backup_time,
-         .backup_time = backup_time,
-+        .has_use_dirty_bitmap = has_use_dirty_bitmap,
-+        .use_dirty_bitmap = use_dirty_bitmap,
-         .has_format = has_format,
-         .format = format,
-         .has_config_file = has_config_file,
-@@ -945,10 +1026,14 @@ BackupStatus *qmp_query_backup(Error **errp)
- 
-     info->has_total = true;
-     info->total = backup_state.stat.total;
-+    info->has_dirty = true;
-+    info->dirty = backup_state.stat.dirty;
-     info->has_zero_bytes = true;
-     info->zero_bytes = backup_state.stat.zero_bytes;
-     info->has_transferred = true;
-     info->transferred = backup_state.stat.transferred;
-+    info->has_reused = true;
-+    info->reused = backup_state.stat.reused;
- 
-     qemu_mutex_unlock(&backup_state.stat.lock);
- 
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index c3b6b93472..992e6c1e3f 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -753,8 +753,13 @@
- #
- # @total: total amount of bytes involved in the backup process
- #
-+# @dirty: with incremental mode (PBS) this is the amount of bytes involved
-+#         in the backup process which are marked dirty.
-+#
- # @transferred: amount of bytes already backed up.
- #
-+# @reused: amount of bytes reused due to deduplication.
-+#
- # @zero-bytes: amount of 'zero' bytes detected.
- #
- # @start-time: time (epoch) when backup job started.
-@@ -767,8 +772,8 @@
- #
- ##
- { 'struct': 'BackupStatus',
--  'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int',
--           '*transferred': 'int', '*zero-bytes': 'int',
-+  'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int', '*dirty': 'int',
-+           '*transferred': 'int', '*zero-bytes': 'int', '*reused': 'int',
-            '*start-time': 'int', '*end-time': 'int',
-            '*backup-file': 'str', '*uuid': 'str' } }
- 
-@@ -811,6 +816,8 @@
- #
- # @backup-time: backup timestamp (Unix epoch, required for format 'pbs')
- #
-+# @use-dirty-bitmap: use dirty bitmap to detect incremental changes since last job (optional for format 'pbs')
-+#
- # Returns: the uuid of the backup job
- #
- ##
-@@ -821,6 +828,7 @@
-                                     '*fingerprint': 'str',
-                                     '*backup-id': 'str',
-                                     '*backup-time': 'int',
-+                                    '*use-dirty-bitmap': 'bool',
-                                     '*format': 'BackupFormat',
-                                     '*config-file': 'str',
-                                     '*firewall-file': 'str',

debian/patches/pve/0031-PVE-Backup-pbs-restore-new-command-to-restore-from-p.patch

@@ -5,33 +5,35 @@ Subject: [PATCH] PVE-Backup: pbs-restore - new command to restore from proxmox
  backup server
 
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[WB: add namespace support]
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
 ---
  meson.build   |   4 +
- pbs-restore.c | 223 ++++++++++++++++++++++++++++++++++++++++++++++++++
- 2 files changed, 227 insertions(+)
+ pbs-restore.c | 236 ++++++++++++++++++++++++++++++++++++++++++++++++++
+ 2 files changed, 240 insertions(+)
  create mode 100644 pbs-restore.c
 
 diff --git a/meson.build b/meson.build
-index 782756162c..63ea813a9a 100644
+index 1f74de1d93..8508aab9c9 100644
 --- a/meson.build
 +++ b/meson.build
-@@ -3602,6 +3602,10 @@ if have_tools
+@@ -4373,6 +4373,10 @@ if have_tools
    vma = executable('vma', files('vma.c', 'vma-reader.c') + genh,
-                    dependencies: [authz, block, crypto, io, qom], install: true)
+                    dependencies: [authz, block, crypto, io, qemuutil, qom], install: true)
  
 +  pbs_restore = executable('pbs-restore', files('pbs-restore.c') + genh,
-+                  dependencies: [authz, block, crypto, io, qom,
++                  dependencies: [authz, block, crypto, io, qemuutil, qom,
 +                    libproxmox_backup_qemu], install: true)
 +
    subdir('storage-daemon')
-   subdir('contrib/rdmacm-mux')
-   subdir('contrib/elf2dmp')
+ 
+   foreach exe: [ 'qemu-img', 'qemu-io', 'qemu-nbd', 'qemu-storage-daemon']
 diff --git a/pbs-restore.c b/pbs-restore.c
 new file mode 100644
-index 0000000000..2f834cf42e
+index 0000000000..f165f418af
 --- /dev/null
 +++ b/pbs-restore.c
-@@ -0,0 +1,223 @@
+@@ -0,0 +1,236 @@
 +/*
 + * Qemu image restore helper for Proxmox Backup
 + *
@@ -55,15 +57,15 @@ index 0000000000..2f834cf42e
 +#include "qemu/main-loop.h"
 +#include "qemu/cutils.h"
 +#include "qapi/error.h"
-+#include "qapi/qmp/qdict.h"
-+#include "sysemu/block-backend.h"
++#include "qobject/qdict.h"
++#include "system/block-backend.h"
 +
 +#include <proxmox-backup-qemu.h>
 +
 +static void help(void)
 +{
 +    const char *help_msg =
-+        "usage: pbs-restore [--repository <repo>] snapshot archive-name target [command options]\n"
++        "usage: pbs-restore [--repository <repo>] [--ns namespace] snapshot archive-name target [command options]\n"
 +        ;
 +
 +    printf("%s", help_msg);
@@ -111,6 +113,7 @@ index 0000000000..2f834cf42e
 +    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;
@@ -124,6 +127,7 @@ index 0000000000..2f834cf42e
 +            {"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}
 +        };
@@ -144,6 +148,9 @@ index 0000000000..2f834cf42e
 +            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;
@@ -194,8 +201,16 @@ index 0000000000..2f834cf42e
 +        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;

debian/patches/pve/0032-PVE-Add-PBS-block-driver-to-map-backup-archives-into.patch

@@ -7,55 +7,58 @@ 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>
+[WB: add namespace support]
+Signed-off-by: Wolfgang Bumiller <w.bumiller@proxmox.com>
 [FE: adapt to changed function signatures
-     make pbs_co_preadv return values consistent with QEMU]
+     make pbs_co_preadv return values consistent with QEMU
+     getlength is now a coroutine function]
 Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- block/meson.build    |   3 +
- block/pbs.c          | 276 +++++++++++++++++++++++++++++++++++++++++++
- configure            |   9 ++
+ block/meson.build    |   2 +
+ block/pbs.c          | 306 +++++++++++++++++++++++++++++++++++++++++++
  meson.build          |   2 +-
- qapi/block-core.json |  13 ++
+ qapi/block-core.json |  29 ++++
  qapi/pragma.json     |   1 +
- 6 files changed, 303 insertions(+), 1 deletion(-)
+ 5 files changed, 339 insertions(+), 1 deletion(-)
  create mode 100644 block/pbs.c
 
 diff --git a/block/meson.build b/block/meson.build
-index e995ae72b9..7ef2fa72d5 100644
+index d1b16e40e9..d243372c41 100644
 --- a/block/meson.build
 +++ b/block/meson.build
-@@ -53,6 +53,9 @@ block_ss.add(files(
+@@ -49,6 +49,8 @@ block_ss.add(files(
    '../pve-backup.c',
  ), libproxmox_backup_qemu)
  
-+block_ss.add(when: 'CONFIG_PBS_BDRV', if_true: files('pbs.c'))
-+block_ss.add(when: 'CONFIG_PBS_BDRV', if_true: libproxmox_backup_qemu)
++block_ss.add(files('pbs.c'), libproxmox_backup_qemu)
 +
  
- softmmu_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
- softmmu_ss.add(files('block-ram-registrar.c'))
+ system_ss.add(when: 'CONFIG_TCG', if_true: files('blkreplay.c'))
+ system_ss.add(files('block-ram-registrar.c'))
 diff --git a/block/pbs.c b/block/pbs.c
 new file mode 100644
-index 0000000000..9d1f1f39d4
+index 0000000000..3e41421716
 --- /dev/null
 +++ b/block/pbs.c
-@@ -0,0 +1,276 @@
+@@ -0,0 +1,306 @@
 +/*
 + * Proxmox Backup Server read-only block driver
 + */
 +
 +#include "qemu/osdep.h"
 +#include "qapi/error.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qmp/qstring.h"
++#include "qobject/qdict.h"
++#include "qobject/qstring.h"
 +#include "qemu/module.h"
 +#include "qemu/option.h"
 +#include "qemu/cutils.h"
 +#include "block/block_int.h"
++#include "block/block-io.h"
 +
 +#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"
@@ -65,10 +68,11 @@ index 0000000000..9d1f1f39d4
 +
 +typedef struct {
 +    ProxmoxRestoreHandle *conn;
-+    char aid;
++    uint8_t aid;
 +    int64_t length;
 +
 +    char *repository;
++    char *namespace;
 +    char *snapshot;
 +    char *archive;
 +} BDRVPBSState;
@@ -83,6 +87,11 @@ index 0000000000..9d1f1f39d4
 +            .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,
 +            .help = "The snapshot to read.",
@@ -118,7 +127,7 @@ index 0000000000..9d1f1f39d4
 +
 +
 +// 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)
 +{
@@ -154,6 +163,7 @@ index 0000000000..9d1f1f39d4
 +    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);
 +
@@ -166,9 +176,12 @@ index 0000000000..9d1f1f39d4
 +    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 */
@@ -188,12 +201,18 @@ index 0000000000..9d1f1f39d4
 +    }
 +
 +    /* acquire handle and length */
-+    s->aid = proxmox_restore_open_image(s->conn, s->archive, &pbs_error);
-+    if (s->aid < 0) {
++    ret = proxmox_restore_open_image(s->conn, s->archive, &pbs_error);
++    if (ret < 0) {
 +        if (pbs_error && errp) error_setg(errp, "PBS open_image failed: %s", pbs_error);
 +        if (pbs_error) proxmox_backup_free_error(pbs_error);
 +        return -ENODEV;
 +    }
++    if (ret > UINT8_MAX) {
++        error_setg(errp, "PBS open_image returned an ID larger than %u", UINT8_MAX);
++        return -ENODEV;
++    }
++    s->aid = ret;
++
 +    s->length = proxmox_restore_get_image_length(s->conn, s->aid, &pbs_error);
 +    if (s->length < 0) {
 +        if (pbs_error && errp) error_setg(errp, "PBS get_image_length failed: %s", pbs_error);
@@ -204,21 +223,17 @@ index 0000000000..9d1f1f39d4
 +    return 0;
 +}
 +
-+static int pbs_file_open(BlockDriverState *bs, QDict *options, int flags,
-+                         Error **errp)
-+{
-+    return pbs_open(bs, options, flags, errp);
-+}
-+
 +static void pbs_close(BlockDriverState *bs) {
 +    BDRVPBSState *s = bs->opaque;
 +    g_free(s->repository);
++    g_free(s->namespace);
 +    g_free(s->snapshot);
 +    g_free(s->archive);
 +    proxmox_restore_disconnect(s->conn);
 +}
 +
-+static int64_t pbs_getlength(BlockDriverState *bs)
++static coroutine_fn int64_t GRAPH_RDLOCK
++pbs_co_getlength(BlockDriverState *bs)
 +{
 +    BDRVPBSState *s = bs->opaque;
 +    return s->length;
@@ -235,14 +250,23 @@ index 0000000000..9d1f1f39d4
 +    aio_co_schedule(rcb->ctx, rcb->co);
 +}
 +
-+static coroutine_fn int pbs_co_preadv(BlockDriverState *bs,
-+                                      int64_t offset, int64_t bytes,
-+                                      QEMUIOVector *qiov, BdrvRequestFlags flags)
++static coroutine_fn int GRAPH_RDLOCK
++pbs_co_preadv(BlockDriverState *bs, int64_t offset, int64_t bytes,
++              QEMUIOVector *qiov, BdrvRequestFlags flags)
 +{
 +    BDRVPBSState *s = bs->opaque;
 +    int ret;
 +    char *pbs_error = NULL;
-+    uint8_t *buf = malloc(bytes);
++    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");
@@ -265,26 +289,34 @@ index 0000000000..9d1f1f39d4
 +        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;
 +}
 +
-+static coroutine_fn int pbs_co_pwritev(BlockDriverState *bs,
-+                                       int64_t offset, int64_t bytes,
-+                                       QEMUIOVector *qiov, BdrvRequestFlags flags)
++static coroutine_fn int GRAPH_RDLOCK
++pbs_co_pwritev(BlockDriverState *bs, int64_t offset, int64_t bytes,
++               QEMUIOVector *qiov, BdrvRequestFlags flags)
 +{
 +    fprintf(stderr, "pbs-bdrv: cannot write to backup file, make sure "
 +           "any attached disk devices are set to read-only!\n");
 +    return -EPERM;
 +}
 +
-+static void pbs_refresh_filename(BlockDriverState *bs)
++static void GRAPH_RDLOCK
++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[] = {
@@ -298,10 +330,9 @@ index 0000000000..9d1f1f39d4
 +
 +    .bdrv_parse_filename    = pbs_parse_filename,
 +
-+    .bdrv_file_open         = pbs_file_open,
 +    .bdrv_open              = pbs_open,
 +    .bdrv_close             = pbs_close,
-+    .bdrv_getlength         = pbs_getlength,
++    .bdrv_co_getlength      = pbs_co_getlength,
 +
 +    .bdrv_co_preadv         = pbs_co_preadv,
 +    .bdrv_co_pwritev        = pbs_co_pwritev,
@@ -316,65 +347,24 @@ index 0000000000..9d1f1f39d4
 +}
 +
 +block_init(bdrv_pbs_init);
-diff --git a/configure b/configure
-index 26c7bc5154..c587e986c7 100755
---- a/configure
-+++ b/configure
-@@ -285,6 +285,7 @@ linux_user=""
- bsd_user=""
- pie=""
- coroutine=""
-+pbs_bdrv="yes"
- plugins="$default_feature"
- meson=""
- ninja=""
-@@ -864,6 +865,10 @@ for opt do
-   --enable-uuid|--disable-uuid)
-       echo "$0: $opt is obsolete, UUID support is always built" >&2
-   ;;
-+  --disable-pbs-bdrv) pbs_bdrv="no"
-+  ;;
-+  --enable-pbs-bdrv) pbs_bdrv="yes"
-+  ;;
-   --with-git=*) git="$optarg"
-   ;;
-   --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
- 
- 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" >> $config_host_mak
-+fi
- 
- # XXX: suppress that
- if [ "$bsd" = "yes" ] ; then
 diff --git a/meson.build b/meson.build
-index 63ea813a9a..f7f5b3f253 100644
+index 8508aab9c9..9c39f54f86 100644
 --- a/meson.build
 +++ b/meson.build
-@@ -3978,7 +3978,7 @@ summary_info += {'bzip2 support':     libbzip2}
- summary_info += {'lzfse support':     liblzfse}
- summary_info += {'zstd support':      zstd}
+@@ -4838,7 +4838,7 @@ summary_info += {'Query Processing Library support': qpl}
+ summary_info += {'UADK Library support': uadk}
+ summary_info += {'qatzip support':    qatzip}
  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}
+ summary_info += {'libcbor support':   libcbor}
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 5ac6276dc1..45b63dfe26 100644
+index dd98e03bf1..0c3ebfa74e 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -3103,6 +3103,7 @@
+@@ -3470,6 +3470,7 @@
              'parallels', 'preallocate', 'qcow', 'qcow2', 'qed', 'quorum',
              'raw', 'rbd',
              { 'name': 'replication', 'if': 'CONFIG_REPLICATION' },
@@ -382,7 +372,7 @@ index 5ac6276dc1..45b63dfe26 100644
              'ssh', 'throttle', 'vdi', 'vhdx',
              { 'name': 'virtio-blk-vfio-pci', 'if': 'CONFIG_BLKIO' },
              { 'name': 'virtio-blk-vhost-user', 'if': 'CONFIG_BLKIO' },
-@@ -3179,6 +3180,17 @@
+@@ -3556,6 +3557,33 @@
  { 'struct': 'BlockdevOptionsNull',
    'data': { '*size': 'int', '*latency-ns': 'uint64', '*read-zeroes': 'bool' } }
  
@@ -391,16 +381,32 @@ index 5ac6276dc1..45b63dfe26 100644
 +#
 +# Driver specific block device options for the PBS backend.
 +#
++# @repository: Proxmox Backup Server repository.
++#
++# @snapshot: backup snapshots ID.
++#
++# @archive: archive name.
++#
++# @keyfile: keyfile to use for encryption.
++#
++# @password: password to use for connection.
++#
++# @fingerprint: backup server fingerprint.
++#
++# @key_password: password to unlock key.
++#
++# @namespace: namespace where backup snapshot lives.
++#
 +##
 +{ 'struct': 'BlockdevOptionsPbs',
 +  'data': { 'repository': 'str', 'snapshot': 'str', 'archive': 'str',
 +            '*keyfile': 'str', '*password': 'str', '*fingerprint': 'str',
-+            '*key_password': 'str' } }
++            '*key_password': 'str', '*namespace': 'str' } }
 +
  ##
  # @BlockdevOptionsNVMe:
  #
-@@ -4531,6 +4543,7 @@
+@@ -5003,6 +5031,7 @@
        'nfs':        'BlockdevOptionsNfs',
        'null-aio':   'BlockdevOptionsNull',
        'null-co':    'BlockdevOptionsNull',
@@ -409,10 +415,10 @@ index 5ac6276dc1..45b63dfe26 100644
        'nvme-io_uring': { 'type': 'BlockdevOptionsNvmeIoUring',
                           'if': 'CONFIG_BLKIO' },
 diff --git a/qapi/pragma.json b/qapi/pragma.json
-index f2097b9020..5ab1890519 100644
+index 6aaa9cb975..e9c595c4ba 100644
 --- a/qapi/pragma.json
 +++ b/qapi/pragma.json
-@@ -47,6 +47,7 @@
+@@ -91,6 +91,7 @@
          'BlockInfo',                # query-block
          'BlockdevAioOptions',       # blockdev-add, -blockdev
          'BlockdevDriver',           # blockdev-add, query-blockstats, ...

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

@@ -1,219 +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: 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 556af25861..a09f722fea 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1042,7 +1042,9 @@ void hmp_backup(Monitor *mon, const QDict *qdict)
-         false, NULL, // PBS fingerprint
-         false, NULL, // PBS backup-id
-         false, 0, // PBS backup-time
--        false, false, // PBS incremental
-+        false, false, // PBS use-dirty-bitmap
-+        false, false, // PBS compress
-+        false, false, // PBS encrypt
-         true, dir ? BACKUP_FORMAT_DIR : BACKUP_FORMAT_VMA,
-         false, NULL, false, NULL, !!devlist,
-         devlist, qdict_haskey(qdict, "speed"), speed, &error);
-diff --git a/pve-backup.c b/pve-backup.c
-index abd7062afe..e113ab61b9 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-redirect-stderr-to-journal-when-daemonized.patch

@@ -9,15 +9,15 @@ fitting.
 Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
 Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
- meson.build | 2 ++
+ meson.build | 3 ++-
  os-posix.c  | 7 +++++--
- 2 files changed, 7 insertions(+), 2 deletions(-)
+ 2 files changed, 7 insertions(+), 3 deletions(-)
 
 diff --git a/meson.build b/meson.build
-index f7f5b3f253..283b0e356e 100644
+index 9c39f54f86..60af7fa723 100644
 --- a/meson.build
 +++ b/meson.build
-@@ -1526,6 +1526,7 @@ keyutils = dependency('libkeyutils', required: false,
+@@ -2162,6 +2162,7 @@ endif
  has_gettid = cc.has_function('gettid')
  
  libuuid = cc.find_library('uuid', required: true)
@@ -25,28 +25,29 @@ index f7f5b3f253..283b0e356e 100644
  libproxmox_backup_qemu = cc.find_library('proxmox_backup_qemu', required: true)
  
  # libselinux
-@@ -3096,6 +3097,7 @@ 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')])
+@@ -3766,7 +3767,7 @@ if have_block
+   if host_os == 'windows'
+     system_ss.add(files('os-win32.c'))
+   else
+-    blockdev_ss.add(files('os-posix.c'))
++    blockdev_ss.add(files('os-posix.c'), libsystemd)
+   endif
  endif
  
 diff --git a/os-posix.c b/os-posix.c
-index 4858650c3e..c5cb12226a 100644
+index 52925c23d3..84b96d3da9 100644
 --- a/os-posix.c
 +++ b/os-posix.c
-@@ -28,6 +28,8 @@
+@@ -29,6 +29,8 @@
  #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)
+ #include "qemu/error-report.h"
+ #include "qemu/log.h"
+@@ -306,9 +308,10 @@ void os_setup_post(void)
  
          dup2(fd, 0);
          dup2(fd, 1);

debian/patches/pve/0034-PVE-Migrate-dirty-bitmap-state-via-savevm.patch

@@ -13,61 +13,68 @@ 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>
+[FE: split up state_pending for 8.0]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
  include/migration/misc.h |   3 ++
  migration/meson.build    |   2 +
  migration/migration.c    |   1 +
- migration/pbs-state.c    | 106 +++++++++++++++++++++++++++++++++++++++
+ migration/pbs-state.c    | 104 +++++++++++++++++++++++++++++++++++++++
  pve-backup.c             |   1 +
  qapi/block-core.json     |   6 +++
- 6 files changed, 119 insertions(+)
+ 6 files changed, 117 insertions(+)
  create mode 100644 migration/pbs-state.c
 
 diff --git a/include/migration/misc.h b/include/migration/misc.h
-index 465906710d..4f0aeceb6f 100644
+index 8fd36eba1d..e963e93e71 100644
 --- a/include/migration/misc.h
 +++ b/include/migration/misc.h
-@@ -75,4 +75,7 @@ bool migration_in_bg_snapshot(void);
- /* migration/block-dirty-bitmap.c */
- void dirty_bitmap_mig_init(void);
+@@ -140,4 +140,7 @@ bool multifd_device_state_save_thread_should_exit(void);
+ void multifd_abort_device_state_save_threads(void);
+ bool multifd_join_device_state_save_threads(void);
  
 +/* migration/pbs-state.c */
 +void pbs_state_mig_init(void);
 +
  #endif
 diff --git a/migration/meson.build b/migration/meson.build
-index 0842d00cd2..d012f4d8d3 100644
+index 46e92249a1..fb3fd7d7d0 100644
 --- a/migration/meson.build
 +++ b/migration/meson.build
-@@ -6,8 +6,10 @@ migration_files = files(
-   'vmstate.c',
+@@ -8,6 +8,7 @@ migration_files = files(
    'qemu-file.c',
    'yank_functions.c',
-+  'pbs-state.c',
  )
- softmmu_ss.add(migration_files)
-+softmmu_ss.add(libproxmox_backup_qemu)
++system_ss.add(libproxmox_backup_qemu)
  
- softmmu_ss.add(files(
+ system_ss.add(files(
    'block-dirty-bitmap.c',
+@@ -31,6 +32,7 @@ system_ss.add(files(
+   'multifd-zlib.c',
+   'multifd-zero-page.c',
+   'options.c',
++  'pbs-state.c',
+   'postcopy-ram.c',
+   'savevm.c',
+   'savevm-async.c',
 diff --git a/migration/migration.c b/migration/migration.c
-index f485eea5fb..89b287180f 100644
+index d46e776e24..2f3430f440 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();
+@@ -319,6 +319,7 @@ void migration_object_init(void)
+ 
+     /* Initialize cpu throttle timers */
+     cpu_throttle_init();
 +    pbs_state_mig_init();
  }
  
- void migration_cancel(const Error *error)
+ typedef struct {
 diff --git a/migration/pbs-state.c b/migration/pbs-state.c
 new file mode 100644
-index 0000000000..29f2b3860d
+index 0000000000..a97187e4d7
 --- /dev/null
 +++ b/migration/pbs-state.c
-@@ -0,0 +1,106 @@
+@@ -0,0 +1,104 @@
 +/*
 + * PBS (dirty-bitmap) state migration
 + */
@@ -86,11 +93,8 @@ index 0000000000..29f2b3860d
 +/* state is accessed via this static variable directly, 'opaque' is NULL */
 +static PBSState pbs_state;
 +
-+static void pbs_state_save_pending(QEMUFile *f, void *opaque,
-+                                      uint64_t max_size,
-+                                      uint64_t *res_precopy_only,
-+                                      uint64_t *res_compatible,
-+                                      uint64_t *res_postcopy_only)
++static void pbs_state_pending(void *opaque, uint64_t *must_precopy,
++                              uint64_t *can_postcopy)
 +{
 +    /* we send everything in save_setup, so nothing is ever pending */
 +}
@@ -116,7 +120,7 @@ index 0000000000..29f2b3860d
 +}
 +
 +/* serialize PBS state and send to target via f, called on source */
-+static int pbs_state_save_setup(QEMUFile *f, void *opaque)
++static int pbs_state_save_setup(QEMUFile *f, void *opaque, Error **errp)
 +{
 +    size_t buf_size;
 +    uint8_t *buf = proxmox_export_state(&buf_size);
@@ -160,7 +164,8 @@ index 0000000000..29f2b3860d
 +static SaveVMHandlers savevm_pbs_state_handlers = {
 +    .save_setup = pbs_state_save_setup,
 +    .has_postcopy = pbs_state_has_postcopy,
-+    .save_live_pending = pbs_state_save_pending,
++    .state_pending_exact = pbs_state_pending,
++    .state_pending_estimate = pbs_state_pending,
 +    .is_active_iterate = pbs_state_is_active_iterate,
 +    .load_state = pbs_state_load,
 +    .is_active = pbs_state_is_active,
@@ -175,22 +180,22 @@ index 0000000000..29f2b3860d
 +                         NULL);
 +}
 diff --git a/pve-backup.c b/pve-backup.c
-index 88268bb586..fa9c6c4493 100644
+index e931cb9203..366b015589 100644
 --- a/pve-backup.c
 +++ b/pve-backup.c
-@@ -1128,6 +1128,7 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
+@@ -1089,6 +1089,7 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
      ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
      ret->pbs_dirty_bitmap = true;
      ret->pbs_dirty_bitmap_savevm = true;
 +    ret->pbs_dirty_bitmap_migration = true;
      ret->query_bitmap_info = true;
-     return ret;
- }
+     ret->pbs_masterkey = true;
+     ret->backup_max_workers = true;
 diff --git a/qapi/block-core.json b/qapi/block-core.json
-index bf559c6d52..24f30260c8 100644
+index 0c3ebfa74e..6838187607 100644
 --- a/qapi/block-core.json
 +++ b/qapi/block-core.json
-@@ -879,6 +879,11 @@
+@@ -1008,6 +1008,11 @@
  # @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
  #                           safely be set for savevm-async.
  #
@@ -199,14 +204,14 @@ index bf559c6d52..24f30260c8 100644
 +#                              migration cap if this is false/unset may lead
 +#                              to crashes on migration!
 +#
- # @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
+ # @pbs-masterkey: True if the QMP backup call supports the 'master_keyfile'
+ #                 parameter.
  #
- ##
-@@ -886,6 +891,7 @@
+@@ -1021,6 +1026,7 @@
    'data': { 'pbs-dirty-bitmap': 'bool',
              'query-bitmap-info': 'bool',
              'pbs-dirty-bitmap-savevm': 'bool',
 +            'pbs-dirty-bitmap-migration': 'bool',
-             'pbs-library-version': 'str' } }
- 
- ##
+             'pbs-masterkey': 'bool',
+             'pbs-library-version': 'str',
+             'backup-max-workers': 'bool' } }

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

@@ -1,74 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 8 Jul 2020 11:57:53 +0200
-Subject: [PATCH] PVE: add query_proxmox_support QMP command
-
-Generic interface for future use, currently used for PBS dirty-bitmap
-backup support.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[PVE: query-proxmox-support: include library version]
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
----
- pve-backup.c         |  9 +++++++++
- qapi/block-core.json | 29 +++++++++++++++++++++++++++++
- 2 files changed, 38 insertions(+)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index e113ab61b9..9318ca4f0c 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -1072,3 +1072,12 @@ BackupStatus *qmp_query_backup(Error **errp)
- 
-     return info;
- }
-+
-+ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
-+{
-+    ProxmoxSupportStatus *ret = g_malloc0(sizeof(*ret));
-+    ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
-+    ret->pbs_dirty_bitmap = true;
-+    ret->pbs_dirty_bitmap_savevm = true;
-+    return ret;
-+}
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 45b63dfe26..8b0e0d92de 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -863,6 +863,35 @@
- ##
- { 'command': 'backup-cancel' }
- 
-+##
-+# @ProxmoxSupportStatus:
-+#
-+# Contains info about supported features added by Proxmox.
-+#
-+# @pbs-dirty-bitmap: True if dirty-bitmap-incremental backups to PBS are
-+#                    supported.
-+#
-+# @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
-+#                           safely be set for savevm-async.
-+#
-+# @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
-+#
-+##
-+{ 'struct': 'ProxmoxSupportStatus',
-+  'data': { 'pbs-dirty-bitmap': 'bool',
-+            'pbs-dirty-bitmap-savevm': 'bool',
-+            'pbs-library-version': 'str' } }
-+
-+##
-+# @query-proxmox-support:
-+#
-+# Returns information about supported features added by Proxmox.
-+#
-+# Returns: @ProxmoxSupportStatus
-+#
-+##
-+{ 'command': 'query-proxmox-support', 'returns': 'ProxmoxSupportStatus' }
-+
- ##
- # @BlockDeviceTimedStats:
- #

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

@@ -1,441 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Wed, 19 Aug 2020 17:02:00 +0200
-Subject: [PATCH] PVE: add query-pbs-bitmap-info QMP call
-
-Returns advanced information about dirty bitmaps used (or not used) for
-the latest PBS backup.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
----
- monitor/hmp-cmds.c   |  28 ++++++-----
- pve-backup.c         | 117 ++++++++++++++++++++++++++++++++-----------
- qapi/block-core.json |  56 +++++++++++++++++++++
- 3 files changed, 159 insertions(+), 42 deletions(-)
-
-diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
-index 670f783515..d819e5fc36 100644
---- a/monitor/hmp-cmds.c
-+++ b/monitor/hmp-cmds.c
-@@ -202,6 +202,7 @@ void hmp_info_mice(Monitor *mon, const QDict *qdict)
- void hmp_info_backup(Monitor *mon, const QDict *qdict)
- {
-     BackupStatus *info;
-+    PBSBitmapInfoList *bitmap_info;
- 
-     info = qmp_query_backup(NULL);
- 
-@@ -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 {
--            bool incremental = false;
-             size_t total_or_dirty = info->total;
--            if (info->has_transferred) {
--                if (info->has_dirty && info->dirty) {
--                     if (info->dirty < info->total) {
--                        total_or_dirty = info->dirty;
--                        incremental = true;
--                    }
--                }
-+            bitmap_info = qmp_query_pbs_bitmap_info(NULL);
-+
-+            while (bitmap_info) {
-+                monitor_printf(mon, "Drive %s:\n",
-+                        bitmap_info->value->drive);
-+                monitor_printf(mon, "  bitmap action: %s\n",
-+                        PBSBitmapAction_str(bitmap_info->value->action));
-+                monitor_printf(mon, "  size: %zd\n",
-+                        bitmap_info->value->size);
-+                monitor_printf(mon, "  dirty: %zd\n",
-+                        bitmap_info->value->dirty);
-+                bitmap_info = bitmap_info->next;
-             }
- 
--            int per = (info->transferred * 100)/total_or_dirty;
--
--            monitor_printf(mon, "Backup mode: %s\n", incremental ? "incremental" : "full");
-+            qapi_free_PBSBitmapInfoList(bitmap_info);
- 
-             int zero_per = (info->has_zero_bytes && info->zero_bytes) ?
-                 (info->zero_bytes * 100)/info->total : 0;
-             monitor_printf(mon, "Total size: %zd\n", info->total);
-+            int trans_per = (info->transferred * 100)/total_or_dirty;
-             monitor_printf(mon, "Transferred bytes: %zd (%d%%)\n",
--                           info->transferred, per);
-+                           info->transferred, trans_per);
-             monitor_printf(mon, "Zero bytes: %zd (%d%%)\n",
-                            info->zero_bytes, zero_per);
- 
-diff --git a/pve-backup.c b/pve-backup.c
-index 9318ca4f0c..c85b2ecd83 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -46,6 +46,7 @@ static struct PVEBackupState {
-         size_t transferred;
-         size_t reused;
-         size_t zero_bytes;
-+        GList *bitmap_list;
-     } stat;
-     int64_t speed;
-     VmaWriter *vmaw;
-@@ -669,7 +670,6 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-     }
- 
-     size_t total = 0;
--    size_t dirty = 0;
- 
-     l = di_list;
-     while (l) {
-@@ -690,18 +690,33 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
- 
-     uuid_generate(uuid);
- 
-+    qemu_mutex_lock(&backup_state.stat.lock);
-+    backup_state.stat.reused = 0;
-+
-+    /* clear previous backup's bitmap_list */
-+    if (backup_state.stat.bitmap_list) {
-+        GList *bl = backup_state.stat.bitmap_list;
-+        while (bl) {
-+            g_free(((PBSBitmapInfo *)bl->data)->drive);
-+            g_free(bl->data);
-+            bl = g_list_next(bl);
-+        }
-+        g_list_free(backup_state.stat.bitmap_list);
-+        backup_state.stat.bitmap_list = NULL;
-+    }
-+
-     if (format == BACKUP_FORMAT_PBS) {
-         if (!task->has_password) {
-             error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'password'");
--            goto err;
-+            goto err_mutex;
-         }
-         if (!task->has_backup_id) {
-             error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-id'");
--            goto err;
-+            goto err_mutex;
-         }
-         if (!task->has_backup_time) {
-             error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "missing parameter 'backup-time'");
--            goto err;
-+            goto err_mutex;
-         }
- 
-         int dump_cb_block_size = PROXMOX_BACKUP_DEFAULT_CHUNK_SIZE; // Hardcoded (4M)
-@@ -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);
--            goto err;
-+            goto err_mutex;
-         }
- 
-         int connect_result = proxmox_backup_co_connect(pbs, task->errp);
-         if (connect_result < 0)
--            goto err;
-+            goto err_mutex;
- 
-         /* register all devices */
-         l = di_list;
-@@ -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);
-+            PBSBitmapAction action = PBS_BITMAP_ACTION_NOT_USED;
-+            size_t dirty = di->size;
- 
-             BdrvDirtyBitmap *bitmap = bdrv_find_dirty_bitmap(di->bs, PBS_BITMAP_NAME);
-             bool expect_only_dirty = false;
-@@ -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) {
--                        goto err;
-+                        goto err_mutex;
-                     }
-+                    action = PBS_BITMAP_ACTION_NEW;
-                 } else {
-                     expect_only_dirty = proxmox_backup_check_incremental(pbs, devname, di->size) != 0;
-                 }
- 
-                 if (expect_only_dirty) {
--                    dirty += bdrv_get_dirty_count(bitmap);
-+                    /* track clean chunks as reused */
-+                    dirty = MIN(bdrv_get_dirty_count(bitmap), di->size);
-+                    backup_state.stat.reused += di->size - dirty;
-+                    action = PBS_BITMAP_ACTION_USED;
-                 } else {
-                     /* mark entire bitmap as dirty to make full backup */
-                     bdrv_set_dirty_bitmap(bitmap, 0, di->size);
--                    dirty += di->size;
-+                    if (action != PBS_BITMAP_ACTION_NEW) {
-+                        action = PBS_BITMAP_ACTION_INVALID;
-+                    }
-                 }
-                 di->bitmap = bitmap;
-             } else {
--                dirty += di->size;
--
-                 /* after a full backup the old dirty bitmap is invalid anyway */
-                 if (bitmap != NULL) {
-                     bdrv_release_dirty_bitmap(bitmap);
-+                    action = PBS_BITMAP_ACTION_NOT_USED_REMOVED;
-                 }
-             }
- 
-             int dev_id = proxmox_backup_co_register_image(pbs, devname, di->size, expect_only_dirty, task->errp);
-             if (dev_id < 0) {
--                goto err;
-+                goto err_mutex;
-             }
- 
-             if (!(di->target = bdrv_backup_dump_create(dump_cb_block_size, di->size, pvebackup_co_dump_pbs_cb, di, task->errp))) {
--                goto err;
-+                goto err_mutex;
-             }
- 
-             di->dev_id = dev_id;
-+
-+            PBSBitmapInfo *info = g_malloc(sizeof(*info));
-+            info->drive = g_strdup(devname);
-+            info->action = action;
-+            info->size = di->size;
-+            info->dirty = dirty;
-+            backup_state.stat.bitmap_list = g_list_append(backup_state.stat.bitmap_list, info);
-         }
-     } else if (format == BACKUP_FORMAT_VMA) {
--        dirty = total;
--
-         vmaw = vma_writer_create(task->backup_file, uuid, &local_err);
-         if (!vmaw) {
-             if (local_err) {
-                 error_propagate(task->errp, local_err);
-             }
--            goto err;
-+            goto err_mutex;
-         }
- 
-         /* register all devices for vma writer */
-@@ -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))) {
--                goto err;
-+                goto err_mutex;
-             }
- 
-             const char *devname = bdrv_get_device_name(di->bs);
-@@ -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");
--                goto err;
-+                goto err_mutex;
-             }
-         }
-     } else if (format == BACKUP_FORMAT_DIR) {
--        dirty = total;
--
-         if (mkdir(task->backup_file, 0640) != 0) {
-             error_setg_errno(task->errp, errno, "can't create directory '%s'\n",
-                              task->backup_file);
--            goto err;
-+            goto err_mutex;
-         }
-         backup_dir = task->backup_file;
- 
-@@ -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);
--                goto err;
-+                goto err_mutex;
-             }
- 
-             di->target = bdrv_open(di->targetfile, NULL, NULL, flags, &local_err);
-             if (!di->target) {
-                 error_propagate(task->errp, local_err);
--                goto err;
-+                goto err_mutex;
-             }
-         }
-     } else {
-         error_set(task->errp, ERROR_CLASS_GENERIC_ERROR, "unknown backup format");
--        goto err;
-+        goto err_mutex;
-     }
- 
- 
-@@ -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) {
--            goto err;
-+            goto err_mutex;
-         }
-     }
- 
-@@ -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) {
--            goto err;
-+            goto err_mutex;
-         }
-     }
-     /* initialize global backup_state now */
--
--    qemu_mutex_lock(&backup_state.stat.lock);
-+    /* note: 'reused' and 'bitmap_list' are initialized earlier */
- 
-     if (backup_state.stat.error) {
-         error_free(backup_state.stat.error);
-@@ -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;
--    backup_state.stat.dirty = dirty;
-+    backup_state.stat.dirty = total - backup_state.stat.reused;
-     backup_state.stat.transferred = 0;
-     backup_state.stat.zero_bytes = 0;
--    backup_state.stat.reused = format == BACKUP_FORMAT_PBS && dirty >= total ? 0 : total - dirty;
- 
-     qemu_mutex_unlock(&backup_state.stat.lock);
- 
-@@ -910,6 +933,9 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-     task->result = uuid_info;
-     return;
- 
-+err_mutex:
-+    qemu_mutex_unlock(&backup_state.stat.lock);
-+
- err:
- 
-     l = di_list;
-@@ -1073,11 +1099,42 @@ BackupStatus *qmp_query_backup(Error **errp)
-     return info;
- }
- 
-+PBSBitmapInfoList *qmp_query_pbs_bitmap_info(Error **errp)
-+{
-+    PBSBitmapInfoList *head = NULL, **p_next = &head;
-+
-+    qemu_mutex_lock(&backup_state.stat.lock);
-+
-+    GList *l = backup_state.stat.bitmap_list;
-+    while (l) {
-+        PBSBitmapInfo *info = (PBSBitmapInfo *)l->data;
-+        l = g_list_next(l);
-+
-+        /* clone bitmap info to avoid auto free after QMP marshalling */
-+        PBSBitmapInfo *info_ret = g_malloc0(sizeof(*info_ret));
-+        info_ret->drive = g_strdup(info->drive);
-+        info_ret->action = info->action;
-+        info_ret->size = info->size;
-+        info_ret->dirty = info->dirty;
-+
-+        PBSBitmapInfoList *info_list = g_malloc0(sizeof(*info_list));
-+        info_list->value = info_ret;
-+
-+        *p_next = info_list;
-+        p_next = &info_list->next;
-+    }
-+
-+    qemu_mutex_unlock(&backup_state.stat.lock);
-+
-+    return head;
-+}
-+
- ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
- {
-     ProxmoxSupportStatus *ret = g_malloc0(sizeof(*ret));
-     ret->pbs_library_version = g_strdup(proxmox_backup_qemu_version());
-     ret->pbs_dirty_bitmap = true;
-     ret->pbs_dirty_bitmap_savevm = true;
-+    ret->query_bitmap_info = true;
-     return ret;
- }
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 8b0e0d92de..7fde927621 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -871,6 +871,8 @@
- # @pbs-dirty-bitmap: True if dirty-bitmap-incremental backups to PBS are
- #                    supported.
- #
-+# @query-bitmap-info: True if the 'query-pbs-bitmap-info' QMP call is supported.
-+#
- # @pbs-dirty-bitmap-savevm: True if 'dirty-bitmaps' migration capability can
- #                           safely be set for savevm-async.
- #
-@@ -879,6 +881,7 @@
- ##
- { 'struct': 'ProxmoxSupportStatus',
-   'data': { 'pbs-dirty-bitmap': 'bool',
-+            'query-bitmap-info': 'bool',
-             'pbs-dirty-bitmap-savevm': 'bool',
-             'pbs-library-version': 'str' } }
- 
-@@ -892,6 +895,59 @@
- ##
- { 'command': 'query-proxmox-support', 'returns': 'ProxmoxSupportStatus' }
- 
-+##
-+# @PBSBitmapAction:
-+#
-+# An action taken on a dirty-bitmap when a backup job was started.
-+#
-+# @not-used: Bitmap mode was not enabled.
-+#
-+# @not-used-removed: Bitmap mode was not enabled, but a bitmap from a
-+#                    previous backup still existed and was removed.
-+#
-+# @new: A new bitmap was attached to the drive for this backup.
-+#
-+# @used: An existing bitmap will be used to only backup changed data.
-+#
-+# @invalid: A bitmap existed, but had to be cleared since it's associated
-+#           base snapshot did not match the base given for the current job or
-+#           the crypt mode has changed.
-+#
-+##
-+{ 'enum': 'PBSBitmapAction',
-+  'data': ['not-used', 'not-used-removed', 'new', 'used', 'invalid'] }
-+
-+##
-+# @PBSBitmapInfo:
-+#
-+# Contains information about dirty bitmaps used for each drive in a PBS backup.
-+#
-+# @drive: The underlying drive.
-+#
-+# @action: The action that was taken when the backup started.
-+#
-+# @size: The total size of the drive.
-+#
-+# @dirty: How much of the drive is considered dirty and will be backed up,
-+#         or 'size' if everything will be.
-+#
-+##
-+{ 'struct': 'PBSBitmapInfo',
-+  'data': { 'drive': 'str', 'action': 'PBSBitmapAction', 'size': 'int',
-+            'dirty': 'int' } }
-+
-+##
-+# @query-pbs-bitmap-info:
-+#
-+# Returns information about dirty bitmaps used on the most recently started
-+# backup. Returns nothing when the last backup was not using PBS or if no
-+# backup occured in this session.
-+#
-+# Returns: @PBSBitmapInfo
-+#
-+##
-+{ 'command': 'query-pbs-bitmap-info', 'returns': ['PBSBitmapInfo'] }
-+
- ##
- # @BlockDeviceTimedStats:
- #

debian/patches/pve/0035-migration-block-dirty-bitmap-migrate-other-bitmaps-e.patch

@@ -15,18 +15,22 @@ 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(-)
+ migration/block-dirty-bitmap.c | 6 +++++-
+ 1 file changed, 5 insertions(+), 1 deletion(-)
 
 diff --git a/migration/block-dirty-bitmap.c b/migration/block-dirty-bitmap.c
-index 9aba7d9c22..f4ecf9c9f9 100644
+index f2c352d4a7..931a8481e9 100644
 --- a/migration/block-dirty-bitmap.c
 +++ b/migration/block-dirty-bitmap.c
-@@ -538,7 +538,7 @@ static int add_bitmaps_to_list(DBMSaveState *s, BlockDriverState *bs,
+@@ -539,7 +539,11 @@ 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);
+         if (bdrv_dirty_bitmap_check(bitmap, BDRV_BITMAP_DEFAULT, errp)) {
 -            return -1;
++            if (errp != NULL) {
++                error_report_err(*errp);
++                *errp = NULL;
++            }
 +            continue;
          }
  

debian/patches/pve/0036-PVE-fall-back-to-open-iscsi-initiatorname.patch

@@ -21,10 +21,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 30 insertions(+)
 
 diff --git a/block/iscsi.c b/block/iscsi.c
-index a316d46d96..3ed4a50c0d 100644
+index 2f0f4dac09..b523137cff 100644
 --- a/block/iscsi.c
 +++ b/block/iscsi.c
-@@ -1387,12 +1387,42 @@ static char *get_initiator_name(QemuOpts *opts)
+@@ -1392,12 +1392,42 @@ static char *get_initiator_name(QemuOpts *opts)
      const char *name;
      char *iscsi_name;
      UuidInfo *uuid_info;

debian/patches/pve/0037-PVE-block-stream-increase-chunk-size.patch

@@ -11,10 +11,10 @@ Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
  1 file changed, 1 insertion(+), 1 deletion(-)
 
 diff --git a/block/stream.c b/block/stream.c
-index 694709bd25..e09bd5c4ef 100644
+index 999d9e56d4..e187cd1262 100644
 --- a/block/stream.c
 +++ b/block/stream.c
-@@ -28,7 +28,7 @@ enum {
+@@ -27,7 +27,7 @@ enum {
       * large enough to process multiple clusters in a single call, so
       * that populating contiguous regions of the image is efficient.
       */

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

@@ -1,293 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Thu, 20 Aug 2020 14:25:00 +0200
-Subject: [PATCH] PVE-Backup: Use a transaction to synchronize job states
-
-By using a JobTxn, we can sync dirty bitmaps only when *all* jobs were
-successful - meaning we don't need to remove them when the backup fails,
-since QEMU's BITMAP_SYNC_MODE_ON_SUCCESS will now handle that for us.
-
-To keep the rate-limiting and IO impact from before, we use a sequential
-transaction, so drives will still be backed up one after the other.
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[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 | 163 ++++++++++++++++-----------------------------------
- 1 file changed, 50 insertions(+), 113 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index c85b2ecd83..b5fb844434 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -52,6 +52,7 @@ static struct PVEBackupState {
-     VmaWriter *vmaw;
-     ProxmoxBackupHandle *pbs;
-     GList *di_list;
-+    JobTxn *txn;
-     QemuMutex backup_mutex;
-     CoMutex dump_callback_mutex;
- } backup_state;
-@@ -71,34 +72,12 @@ typedef struct PVEBackupDevInfo {
-     size_t size;
-     uint64_t block_size;
-     uint8_t dev_id;
--    bool completed;
-     char targetfile[PATH_MAX];
-     BdrvDirtyBitmap *bitmap;
-     BlockDriverState *target;
-+    BlockJob *job;
- } PVEBackupDevInfo;
- 
--static void pvebackup_run_next_job(void);
--
--static BlockJob *
--lookup_active_block_job(PVEBackupDevInfo *di)
--{
--    if (!di->completed && di->bs) {
--        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;
--                }
--            }
--        }
--    }
--    return NULL;
--}
--
- static void pvebackup_propagate_error(Error *err)
- {
-     qemu_mutex_lock(&backup_state.stat.lock);
-@@ -274,18 +253,6 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
-             if (local_err != NULL) {
-                 pvebackup_propagate_error(local_err);
-             }
--        } else {
--            // on error or cancel we cannot ensure synchronization of dirty
--            // bitmaps with backup server, so remove all and do full backup next
--            GList *l = backup_state.di_list;
--            while (l) {
--                PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
--                l = g_list_next(l);
--
--                if (di->bitmap) {
--                    bdrv_release_dirty_bitmap(di->bitmap);
--                }
--            }
-         }
- 
-         proxmox_backup_disconnect(backup_state.pbs);
-@@ -324,8 +291,6 @@ static void pvebackup_complete_cb(void *opaque, int ret)
- 
-     qemu_mutex_lock(&backup_state.backup_mutex);
- 
--    di->completed = true;
--
-     if (ret < 0) {
-         Error *local_err = NULL;
-         error_setg(&local_err, "job failed with err %d - %s", ret, strerror(-ret));
-@@ -338,20 +303,17 @@ static void pvebackup_complete_cb(void *opaque, int ret)
- 
-     block_on_coroutine_fn(pvebackup_complete_stream, di);
- 
--    // remove self from job queue
-+    // remove self from job list
-     backup_state.di_list = g_list_remove(backup_state.di_list, di);
- 
--    if (di->bitmap && ret < 0) {
--        // on error or cancel we cannot ensure synchronization of dirty
--        // bitmaps with backup server, so remove all and do full backup next
--        bdrv_release_dirty_bitmap(di->bitmap);
--    }
--
-     g_free(di);
- 
--    qemu_mutex_unlock(&backup_state.backup_mutex);
-+    /* call cleanup if we're the last job */
-+    if (!g_list_first(backup_state.di_list)) {
-+        block_on_coroutine_fn(pvebackup_co_cleanup, NULL);
-+    }
- 
--    pvebackup_run_next_job();
-+    qemu_mutex_unlock(&backup_state.backup_mutex);
- }
- 
- static void pvebackup_cancel(void)
-@@ -373,32 +335,28 @@ static void pvebackup_cancel(void)
-         proxmox_backup_abort(backup_state.pbs, "backup canceled");
-     }
- 
--    qemu_mutex_unlock(&backup_state.backup_mutex);
--
--    for(;;) {
--
--        BlockJob *next_job = NULL;
-+    /* it's enough to cancel one job in the transaction, the rest will follow
-+     * automatically */
-+    GList *bdi = g_list_first(backup_state.di_list);
-+    BlockJob *cancel_job = bdi && bdi->data ?
-+        ((PVEBackupDevInfo *)bdi->data)->job :
-+        NULL;
- 
--        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) {
-+        WITH_JOB_LOCK_GUARD() {
-+            job_ref_locked(&cancel_job->job);
-         }
-+    }
- 
--        qemu_mutex_unlock(&backup_state.backup_mutex);
-+    /* job_cancel_sync may enter the job, so we need to release the
-+     * backup_mutex to avoid deadlock */
-+    qemu_mutex_unlock(&backup_state.backup_mutex);
- 
--        if (next_job) {
--            job_cancel_sync(&next_job->job, true);
--        } else {
--            break;
-+    if (cancel_job) {
-+        WITH_JOB_LOCK_GUARD() {
-+            job_cancel_sync_locked(&cancel_job->job, true);
-+            job_unref_locked(&cancel_job->job);
-         }
-     }
- }
-@@ -458,49 +416,19 @@ static int coroutine_fn pvebackup_co_add_config(
-     goto out;
- }
- 
--bool job_should_pause_locked(Job *job);
--
--static void pvebackup_run_next_job(void)
--{
--    assert(!qemu_in_coroutine());
--
--    qemu_mutex_lock(&backup_state.backup_mutex);
--
--    GList *l = backup_state.di_list;
--    while (l) {
--        PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
--        l = g_list_next(l);
--
--        BlockJob *job = lookup_active_block_job(di);
--
--        if (job) {
--            qemu_mutex_unlock(&backup_state.backup_mutex);
--
--            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);
--                    }
--                }
--            }
--            return;
--        }
--    }
--
--    block_on_coroutine_fn(pvebackup_co_cleanup, NULL); // no more jobs, run cleanup
--
--    qemu_mutex_unlock(&backup_state.backup_mutex);
--}
--
- static bool create_backup_jobs(void) {
- 
-     assert(!qemu_in_coroutine());
- 
-     Error *local_err = NULL;
- 
-+    /* create job transaction to synchronize bitmap commit and cancel all
-+     * jobs in case one errors */
-+    if (backup_state.txn) {
-+        job_txn_unref(backup_state.txn);
-+    }
-+    backup_state.txn = job_txn_new_seq();
-+
-     BackupPerf perf = { .max_workers = 16 };
- 
-     /* create and start all jobs (paused state) */
-@@ -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, &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);
- 
-@@ -535,7 +463,8 @@ static bool create_backup_jobs(void) {
-             pvebackup_propagate_error(create_job_err);
-             break;
-         }
--        job_start(&job->job);
-+
-+        di->job = job;
- 
-         bdrv_unref(di->target);
-         di->target = NULL;
-@@ -553,6 +482,12 @@ static bool create_backup_jobs(void) {
-                 bdrv_unref(di->target);
-                 di->target = NULL;
-             }
-+
-+            if (di->job) {
-+                WITH_JOB_LOCK_GUARD() {
-+                    job_unref_locked(&di->job->job);
-+                }
-+            }
-         }
-     }
- 
-@@ -943,10 +878,6 @@ err:
-         PVEBackupDevInfo *di = (PVEBackupDevInfo *)l->data;
-         l = g_list_next(l);
- 
--        if (di->bitmap) {
--            bdrv_release_dirty_bitmap(di->bitmap);
--        }
--
-         if (di->target) {
-             bdrv_unref(di->target);
-         }
-@@ -1035,9 +966,15 @@ UuidInfo *qmp_backup(
-     block_on_coroutine_fn(pvebackup_co_prepare, &task);
- 
-     if (*errp == NULL) {
--        create_backup_jobs();
-+        bool errors = create_backup_jobs();
-         qemu_mutex_unlock(&backup_state.backup_mutex);
--        pvebackup_run_next_job();
-+
-+        if (!errors) {
-+            /* start the first job in the transaction
-+             * note: this might directly enter the job, so we need to do this
-+             * after unlocking the backup_mutex */
-+            job_txn_start_seq(backup_state.txn);
-+        }
-     } else {
-         qemu_mutex_unlock(&backup_state.backup_mutex);
-     }

debian/patches/pve/0038-block-add-alloc-track-driver.patch

@@ -19,26 +19,34 @@ 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'.
+Replacing the node cannot be done in the
+track_co_change_backing_file() callback, because replacing a node
+cannot happen in a coroutine and requires the block graph lock
+exclusively. Could either become a special option for the stream job,
+or maybe the upcoming blockdev-replace QMP command can be used in the
+future.
 
 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]
+     make error return value consistent with QEMU
+     avoid premature break during read
+     adhere to block graph lock requirements
+     avoid superfluous child permission update]
 Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
 ---
- block/alloc-track.c | 350 ++++++++++++++++++++++++++++++++++++++++++++
+ block/alloc-track.c | 343 ++++++++++++++++++++++++++++++++++++++++++++
  block/meson.build   |   1 +
- 2 files changed, 351 insertions(+)
+ block/stream.c      |  34 +++++
+ 3 files changed, 378 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
+index 0000000000..718aaabf2a
 --- /dev/null
 +++ b/block/alloc-track.c
-@@ -0,0 +1,350 @@
+@@ -0,0 +1,343 @@
 +/*
 + * 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
@@ -54,25 +62,21 @@ index 0000000000..43d40d11af
 +#include "qemu/osdep.h"
 +#include "qapi/error.h"
 +#include "block/block_int.h"
-+#include "qapi/qmp/qdict.h"
-+#include "qapi/qmp/qstring.h"
++#include "block/dirty-bitmap.h"
++#include "block/graph-lock.h"
++#include "qobject/qdict.h"
++#include "qobject/qstring.h"
 +#include "qemu/cutils.h"
++#include "qemu/error-report.h"
 +#include "qemu/option.h"
 +#include "qemu/module.h"
-+#include "sysemu/block-backend.h"
++#include "system/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;
++    uint64_t granularity;
 +} BDRVAllocTrackState;
 +
 +static QemuOptsList runtime_opts = {
@@ -89,26 +93,29 @@ index 0000000000..43d40d11af
 +    },
 +};
 +
-+static void track_refresh_limits(BlockDriverState *bs, Error **errp)
++static void GRAPH_RDLOCK
++track_refresh_limits(BlockDriverState *bs, Error **errp)
 +{
-+    BlockDriverInfo bdi;
++    BDRVAllocTrackState *s = bs->opaque;
 +
 +    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);
++    /*
++     * Always use alignment from underlying write device so RMW cycle for
++     * bdrv_pwritev reads data from our backing via track_co_preadv. Also use at
++     * least the bitmap granularity.
++     */
 +    bs->bl.request_alignment = MAX(bs->file->bs->bl.request_alignment,
-+                                   MAX(bdi.cluster_size, BDRV_SECTOR_SIZE));
++                                   s->granularity);
 +}
 +
 +static int track_open(BlockDriverState *bs, QDict *options, int flags,
 +                      Error **errp)
 +{
 +    BDRVAllocTrackState *s = bs->opaque;
++    BdrvChild *file = NULL;
 +    QemuOpts *opts;
 +    Error *local_err = NULL;
 +    int ret = 0;
@@ -121,32 +128,63 @@ index 0000000000..43d40d11af
 +        goto fail;
 +    }
 +
-+    s->auto_remove = qemu_opt_get_bool(opts, TRACK_OPT_AUTO_REMOVE, false);
++    if (!qemu_opt_get_bool(opts, TRACK_OPT_AUTO_REMOVE, false)) {
++        error_setg(errp, "alloc-track: requires auto-remove option to be set to on");
++        ret = -EINVAL;
++        goto fail;
++    }
 +
 +    /* 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);
++    file = bdrv_open_child(NULL, options, "file", bs, &child_of_bds,
++                           BDRV_CHILD_DATA | BDRV_CHILD_METADATA, false,
++                           &local_err);
++    bdrv_graph_wrlock();
++    bs->file = file;
++    bdrv_graph_wrunlock();
 +    if (local_err) {
 +        ret = -EINVAL;
 +        error_propagate(errp, local_err);
 +        goto fail;
 +    }
 +
++    bdrv_graph_rdlock_main_loop();
++    BlockDriverInfo bdi = {0};
++    ret = bdrv_get_info(bs->file->bs, &bdi);
++    if (ret < 0) {
++        /*
++         * Not a hard failure. Worst that can happen is partial cluster
++         * allocation in the write target. However, the driver here returns its
++         * allocation status based on the dirty bitmap, so any other data that
++         * maps to such a cluster will still be copied later by a stream job (or
++         * during writes to that cluster).
++         */
++        warn_report("alloc-track: unable to query cluster size for write target: %s",
++                    strerror(ret));
++    }
++    ret = 0;
++    /*
++     * Always consider alignment from underlying write device so RMW cycle for
++     * bdrv_pwritev reads data from our backing via track_co_preadv. Also try to
++     * avoid partial cluster allocation in the write target by considering the
++     * cluster size.
++     */
++    s->granularity = MAX(bs->file->bs->bl.request_alignment,
++                         MAX(bdi.cluster_size, BDRV_SECTOR_SIZE));
 +    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);
++    s->bitmap = bdrv_create_dirty_bitmap(bs->file->bs, s->granularity, NULL,
++                                         &local_err);
++    bdrv_graph_rdunlock_main_loop();
 +    if (local_err) {
 +        ret = -EIO;
 +        error_propagate(errp, local_err);
 +        goto fail;
 +    }
 +
-+    s->drop_state = DropNone;
-+
 +fail:
 +    if (ret < 0) {
++        bdrv_graph_wrlock();
 +        bdrv_unref_child(bs, bs->file);
++        bdrv_graph_wrunlock();
 +        if (s->bitmap) {
 +            bdrv_release_dirty_bitmap(s->bitmap);
 +        }
@@ -163,13 +201,15 @@ index 0000000000..43d40d11af
 +    }
 +}
 +
-+static int64_t track_getlength(BlockDriverState *bs)
++static coroutine_fn int64_t GRAPH_RDLOCK
++track_co_getlength(BlockDriverState *bs)
 +{
-+    return bdrv_getlength(bs->file->bs);
++    return bdrv_co_getlength(bs->file->bs);
 +}
 +
-+static int coroutine_fn track_co_preadv(BlockDriverState *bs,
-+    int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++static int coroutine_fn GRAPH_RDLOCK
++track_co_preadv(BlockDriverState *bs, int64_t offset, int64_t bytes,
++                QEMUIOVector *qiov, BdrvRequestFlags flags)
 +{
 +    BDRVAllocTrackState *s = bs->opaque;
 +    QEMUIOVector local_qiov;
@@ -215,7 +255,8 @@ index 0000000000..43d40d11af
 +            ret = bdrv_co_preadv(bs->backing, local_offset, local_bytes,
 +                                 &local_qiov, flags);
 +        } else {
-+            ret = qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
++            qemu_iovec_memset(&local_qiov, cur_offset, 0, local_bytes);
++            ret = 0;
 +        }
 +
 +        if (ret != 0) {
@@ -226,36 +267,39 @@ index 0000000000..43d40d11af
 +    return ret;
 +}
 +
-+static int coroutine_fn track_co_pwritev(BlockDriverState *bs,
-+    int64_t offset, int64_t bytes, QEMUIOVector *qiov, BdrvRequestFlags flags)
++static int coroutine_fn GRAPH_RDLOCK
++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)
++static int coroutine_fn GRAPH_RDLOCK
++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)
++static int coroutine_fn GRAPH_RDLOCK
++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)
++static coroutine_fn int GRAPH_RDLOCK
++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)
++static int coroutine_fn GRAPH_RDLOCK
++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;
 +
@@ -281,23 +325,13 @@ index 0000000000..43d40d11af
 +    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)
++static void GRAPH_RDLOCK
++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 {
@@ -307,55 +341,22 @@ index 0000000000..43d40d11af
 +    }
 +}
 +
-+static void track_drop(void *opaque)
++static int coroutine_fn GRAPH_RDLOCK
++track_co_change_backing_file(BlockDriverState *bs, const char *backing_file,
++                             const char *backing_fmt)
 +{
-+    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);
-+    }
-+
++    /*
++     * Note that the actual backing file graph change is already done in the
++     * stream job itself with bdrv_set_backing_hd_drained(), so no need to
++     * actually do anything here. But still needs to be implemented, to make
++     * our caller (i.e. bdrv_co_change_backing_file() do the right thing).
++     *
++     * FIXME
++     * We'd like to auto-remove ourselves from the block graph, but it cannot
++     * be done from a coroutine. Currently done in the stream job, where it
++     * kinda fits better, but in the long-term, a special parameter would be
++     * nice (or done via qemu-server via upcoming blockdev-replace QMP command).
++     */
 +    return 0;
 +}
 +
@@ -363,9 +364,9 @@ index 0000000000..43d40d11af
 +    .format_name                      = "alloc-track",
 +    .instance_size                    = sizeof(BDRVAllocTrackState),
 +
-+    .bdrv_file_open                   = track_open,
++    .bdrv_open                        = track_open,
 +    .bdrv_close                       = track_close,
-+    .bdrv_getlength                   = track_getlength,
++    .bdrv_co_getlength                = track_co_getlength,
 +    .bdrv_child_perm                  = track_child_perm,
 +    .bdrv_refresh_limits              = track_refresh_limits,
 +
@@ -380,7 +381,7 @@ index 0000000000..43d40d11af
 +    .supports_backing                 = true,
 +
 +    .bdrv_co_block_status             = track_co_block_status,
-+    .bdrv_change_backing_file         = track_change_backing_file,
++    .bdrv_co_change_backing_file      = track_co_change_backing_file,
 +};
 +
 +static void bdrv_alloc_track_init(void)
@@ -390,7 +391,7 @@ index 0000000000..43d40d11af
 +
 +block_init(bdrv_alloc_track_init);
 diff --git a/block/meson.build b/block/meson.build
-index 7ef2fa72d5..15352f579f 100644
+index d243372c41..9b45b5256d 100644
 --- a/block/meson.build
 +++ b/block/meson.build
 @@ -2,6 +2,7 @@ block_ss.add(genh)
@@ -401,3 +402,48 @@ index 7ef2fa72d5..15352f579f 100644
    'amend.c',
    'backup.c',
    'backup-dump.c',
+diff --git a/block/stream.c b/block/stream.c
+index e187cd1262..0b61029399 100644
+--- a/block/stream.c
++++ b/block/stream.c
+@@ -120,6 +120,40 @@ static int stream_prepare(Job *job)
+             ret = -EPERM;
+             goto out;
+         }
++
++        /*
++         * This cannot be done in the co_change_backing_file callback, because
++         * bdrv_replace_node() cannot be done in a coroutine. The latter also
++         * requires the graph lock exclusively. Only required for the
++         * alloc-track driver.
++         *
++         * The long-term plan is to either have an explicit parameter for the
++         * stream job or use the upcoming blockdev-replace QMP command.
++         */
++        if (base_id == NULL && strcmp(unfiltered_bs->drv->format_name, "alloc-track") == 0) {
++            BlockDriverState *file_bs;
++
++            bdrv_graph_rdlock_main_loop();
++            file_bs = unfiltered_bs->file->bs;
++            bdrv_graph_rdunlock_main_loop();
++
++            bdrv_ref(unfiltered_bs); // unrefed by bdrv_replace_node()
++            bdrv_drained_begin(file_bs);
++            bdrv_graph_wrlock();
++
++            bdrv_replace_node(unfiltered_bs, file_bs, &local_err);
++
++            bdrv_graph_wrunlock();
++            bdrv_drained_end(file_bs);
++            bdrv_unref(unfiltered_bs);
++
++            if (local_err) {
++                error_prepend(&local_err, "failed to replace alloc-track node: ");
++                error_report_err(local_err);
++                ret = -EPERM;
++                goto out;
++            }
++        }
+     }
+ 
+ out:

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

@@ -1,499 +0,0 @@
-From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
-From: Stefan Reiter <s.reiter@proxmox.com>
-Date: Mon, 28 Sep 2020 13:40:51 +0200
-Subject: [PATCH] PVE-Backup: Don't block on finishing and cleanup
- create_backup_jobs
-
-proxmox_backup_co_finish is already async, but previously we would wait
-for the coroutine using block_on_coroutine_fn(). Avoid this by
-scheduling pvebackup_co_complete_stream (and thus pvebackup_co_cleanup)
-as a real coroutine when calling from pvebackup_complete_cb. This is ok,
-since complete_stream uses the backup_mutex internally to synchronize,
-and other streams can happily continue writing in the meantime anyway.
-
-To accomodate, backup_mutex is converted to a CoMutex. This means
-converting every user to a coroutine. This is not just useful here, but
-will come in handy once this series[0] is merged, and QMP calls can be
-yield-able coroutines too. Then we can also finally get rid of
-block_on_coroutine_fn.
-
-Cases of aio_context_acquire/release from within what is now a coroutine
-are changed to aio_co_reschedule_self, which works since a running
-coroutine always holds the aio lock for the context it is running in.
-
-job_cancel_sync is called from a BH since it can't be run from a
-coroutine (uses AIO_WAIT_WHILE internally).
-
-Same thing for create_backup_jobs, which is converted to a BH too.
-
-To communicate the finishing state, a new property is introduced to
-query-backup: 'finishing'. A new state is explicitly not used, since
-that would break compatibility with older qemu-server versions.
-
-Also fix create_backup_jobs:
-
-No more weird bool returns, just the standard "errp" format used
-everywhere else too. With this, if backup_job_create fails, the error
-message is actually returned over QMP and can be shown to the user.
-
-To facilitate correct cleanup on such an error, we call
-create_backup_jobs as a bottom half directly from pvebackup_co_prepare.
-This additionally allows us to actually hold the backup_mutex during
-operation.
-
-Also add a job_cancel_sync before job_unref, since a job must be in
-STATUS_NULL to be deleted by unref, which could trigger an assert
-before.
-
-[0] https://lists.gnu.org/archive/html/qemu-devel/2020-09/msg03515.html
-
-Signed-off-by: Stefan Reiter <s.reiter@proxmox.com>
-Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
-[FE: add new force parameter to job_cancel_sync calls]
-Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
----
- pve-backup.c         | 212 +++++++++++++++++++++++++++----------------
- qapi/block-core.json |   5 +-
- 2 files changed, 138 insertions(+), 79 deletions(-)
-
-diff --git a/pve-backup.c b/pve-backup.c
-index b5fb844434..88268bb586 100644
---- a/pve-backup.c
-+++ b/pve-backup.c
-@@ -33,7 +33,9 @@ const char *PBS_BITMAP_NAME = "pbs-incremental-dirty-bitmap";
- 
- static struct PVEBackupState {
-     struct {
--        // Everithing accessed from qmp_backup_query command is protected using lock
-+        // Everything accessed from qmp_backup_query command is protected using
-+        // this lock. Do NOT hold this lock for long times, as it is sometimes
-+        // acquired from coroutines, and thus any wait time may block the guest.
-         QemuMutex lock;
-         Error *error;
-         time_t start_time;
-@@ -47,20 +49,22 @@ static struct PVEBackupState {
-         size_t reused;
-         size_t zero_bytes;
-         GList *bitmap_list;
-+        bool finishing;
-+        bool starting;
-     } stat;
-     int64_t speed;
-     VmaWriter *vmaw;
-     ProxmoxBackupHandle *pbs;
-     GList *di_list;
-     JobTxn *txn;
--    QemuMutex backup_mutex;
-+    CoMutex backup_mutex;
-     CoMutex dump_callback_mutex;
- } backup_state;
- 
- static void pvebackup_init(void)
- {
-     qemu_mutex_init(&backup_state.stat.lock);
--    qemu_mutex_init(&backup_state.backup_mutex);
-+    qemu_co_mutex_init(&backup_state.backup_mutex);
-     qemu_co_mutex_init(&backup_state.dump_callback_mutex);
- }
- 
-@@ -72,6 +76,7 @@ typedef struct PVEBackupDevInfo {
-     size_t size;
-     uint64_t block_size;
-     uint8_t dev_id;
-+    int completed_ret; // INT_MAX if not completed
-     char targetfile[PATH_MAX];
-     BdrvDirtyBitmap *bitmap;
-     BlockDriverState *target;
-@@ -227,12 +232,12 @@ pvebackup_co_dump_vma_cb(
- }
- 
- // assumes the caller holds backup_mutex
--static void coroutine_fn pvebackup_co_cleanup(void *unused)
-+static void coroutine_fn pvebackup_co_cleanup(void)
- {
-     assert(qemu_in_coroutine());
- 
-     qemu_mutex_lock(&backup_state.stat.lock);
--    backup_state.stat.end_time = time(NULL);
-+    backup_state.stat.finishing = true;
-     qemu_mutex_unlock(&backup_state.stat.lock);
- 
-     if (backup_state.vmaw) {
-@@ -261,35 +266,29 @@ static void coroutine_fn pvebackup_co_cleanup(void *unused)
- 
-     g_list_free(backup_state.di_list);
-     backup_state.di_list = NULL;
-+
-+    qemu_mutex_lock(&backup_state.stat.lock);
-+    backup_state.stat.end_time = time(NULL);
-+    backup_state.stat.finishing = false;
-+    qemu_mutex_unlock(&backup_state.stat.lock);
- }
- 
--// assumes the caller holds backup_mutex
--static void coroutine_fn pvebackup_complete_stream(void *opaque)
-+static void coroutine_fn pvebackup_co_complete_stream(void *opaque)
- {
-     PVEBackupDevInfo *di = opaque;
-+    int ret = di->completed_ret;
- 
--    bool error_or_canceled = pvebackup_error_or_canceled();
--
--    if (backup_state.vmaw) {
--        vma_writer_close_stream(backup_state.vmaw, di->dev_id);
-+    qemu_mutex_lock(&backup_state.stat.lock);
-+    bool starting = backup_state.stat.starting;
-+    qemu_mutex_unlock(&backup_state.stat.lock);
-+    if (starting) {
-+        /* in 'starting' state, no tasks have been run yet, meaning we can (and
-+         * must) skip all cleanup, as we don't know what has and hasn't been
-+         * initialized yet. */
-+        return;
-     }
- 
--    if (backup_state.pbs && !error_or_canceled) {
--        Error *local_err = NULL;
--        proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
--        if (local_err != NULL) {
--            pvebackup_propagate_error(local_err);
--        }
--    }
--}
--
--static void pvebackup_complete_cb(void *opaque, int ret)
--{
--    assert(!qemu_in_coroutine());
--
--    PVEBackupDevInfo *di = opaque;
--
--    qemu_mutex_lock(&backup_state.backup_mutex);
-+    qemu_co_mutex_lock(&backup_state.backup_mutex);
- 
-     if (ret < 0) {
-         Error *local_err = NULL;
-@@ -301,7 +300,19 @@ static void pvebackup_complete_cb(void *opaque, int ret)
- 
-     assert(di->target == NULL);
- 
--    block_on_coroutine_fn(pvebackup_complete_stream, di);
-+    bool error_or_canceled = pvebackup_error_or_canceled();
-+
-+    if (backup_state.vmaw) {
-+        vma_writer_close_stream(backup_state.vmaw, di->dev_id);
-+    }
-+
-+    if (backup_state.pbs && !error_or_canceled) {
-+        Error *local_err = NULL;
-+        proxmox_backup_co_close_image(backup_state.pbs, di->dev_id, &local_err);
-+        if (local_err != NULL) {
-+            pvebackup_propagate_error(local_err);
-+        }
-+    }
- 
-     // remove self from job list
-     backup_state.di_list = g_list_remove(backup_state.di_list, di);
-@@ -310,21 +321,46 @@ 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)) {
--        block_on_coroutine_fn(pvebackup_co_cleanup, NULL);
-+        pvebackup_co_cleanup();
-     }
- 
--    qemu_mutex_unlock(&backup_state.backup_mutex);
-+    qemu_co_mutex_unlock(&backup_state.backup_mutex);
- }
- 
--static void pvebackup_cancel(void)
-+static void pvebackup_complete_cb(void *opaque, int ret)
- {
--    assert(!qemu_in_coroutine());
-+    PVEBackupDevInfo *di = opaque;
-+    di->completed_ret = ret;
-+
-+    /*
-+     * Schedule stream cleanup in async coroutine. close_image and finish might
-+     * take a while, so we can't block on them here. This way it also doesn't
-+     * matter if we're already running in a coroutine or not.
-+     * Note: di is a pointer to an entry in the global backup_state struct, so
-+     * it stays valid.
-+     */
-+    Coroutine *co = qemu_coroutine_create(pvebackup_co_complete_stream, di);
-+    aio_co_enter(qemu_get_aio_context(), co);
-+}
-+
-+/*
-+ * job_cancel(_sync) does not like to be called from coroutines, so defer to
-+ * main loop processing via a bottom half.
-+ */
-+static void job_cancel_bh(void *opaque) {
-+    CoCtxData *data = (CoCtxData*)opaque;
-+    Job *job = (Job*)data->data;
-+    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;
-     error_setg(&cancel_err, "backup canceled");
-     pvebackup_propagate_error(cancel_err);
- 
--    qemu_mutex_lock(&backup_state.backup_mutex);
-+    qemu_co_mutex_lock(&backup_state.backup_mutex);
- 
-     if (backup_state.vmaw) {
-         /* make sure vma writer does not block anymore */
-@@ -342,28 +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) {
--        WITH_JOB_LOCK_GUARD() {
--            job_ref_locked(&cancel_job->job);
--        }
-+        CoCtxData data = {
-+            .ctx = qemu_get_current_aio_context(),
-+            .co = qemu_coroutine_self(),
-+            .data = &cancel_job->job,
-+        };
-+        aio_bh_schedule_oneshot(data.ctx, job_cancel_bh, &data);
-+        qemu_coroutine_yield();
-     }
- 
--    /* job_cancel_sync may enter the job, so we need to release the
--     * backup_mutex to avoid deadlock */
--    qemu_mutex_unlock(&backup_state.backup_mutex);
--
--    if (cancel_job) {
--        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);
- }
- 
- void qmp_backup_cancel(Error **errp)
- {
--    pvebackup_cancel();
-+    block_on_coroutine_fn(pvebackup_co_cancel, NULL);
- }
- 
- // assumes the caller holds backup_mutex
-@@ -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.
-+ * The caller is responsible that backup_mutex is held nonetheless.
-+ */
-+static void create_backup_jobs_bh(void *opaque) {
- 
-     assert(!qemu_in_coroutine());
- 
-+    CoCtxData *data = (CoCtxData*)opaque;
-+    Error **errp = (Error**)data->data;
-+
-     Error *local_err = NULL;
- 
-     /* create job transaction to synchronize bitmap commit and cancel all
-@@ -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;
- 
--// assumes the caller holds backup_mutex
- static void coroutine_fn pvebackup_co_prepare(void *opaque)
- {
-     assert(qemu_in_coroutine());
- 
-+    qemu_co_mutex_lock(&backup_state.backup_mutex);
-+
-     QmpBackupTask *task = opaque;
- 
-     task->result = NULL; // just to be sure
-@@ -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) {
--         error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-+        error_set(task->errp, ERROR_CLASS_GENERIC_ERROR,
-                   "previous backup not finished");
-+        qemu_co_mutex_unlock(&backup_state.backup_mutex);
-         return;
-     }
- 
-@@ -621,6 +658,8 @@ static void coroutine_fn pvebackup_co_prepare(void *opaque)
-         }
-         di->size = size;
-         total += size;
-+
-+        di->completed_ret = INT_MAX;
-     }
- 
-     uuid_generate(uuid);
-@@ -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);
- 
-@@ -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:
-@@ -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;
-+
-+    qemu_co_mutex_unlock(&backup_state.backup_mutex);
-     return;
- }
- 
-@@ -961,24 +1032,8 @@ UuidInfo *qmp_backup(
-         .errp = errp,
-     };
- 
--    qemu_mutex_lock(&backup_state.backup_mutex);
--
-     block_on_coroutine_fn(pvebackup_co_prepare, &task);
- 
--    if (*errp == NULL) {
--        bool errors = create_backup_jobs();
--        qemu_mutex_unlock(&backup_state.backup_mutex);
--
--        if (!errors) {
--            /* start the first job in the transaction
--             * note: this might directly enter the job, so we need to do this
--             * after unlocking the backup_mutex */
--            job_txn_start_seq(backup_state.txn);
--        }
--    } else {
--        qemu_mutex_unlock(&backup_state.backup_mutex);
--    }
--
-     return task.result;
- }
- 
-@@ -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;
-+    info->finishing = backup_state.stat.finishing;
- 
-     qemu_mutex_unlock(&backup_state.stat.lock);
- 
-diff --git a/qapi/block-core.json b/qapi/block-core.json
-index 7fde927621..bf559c6d52 100644
---- a/qapi/block-core.json
-+++ b/qapi/block-core.json
-@@ -770,12 +770,15 @@
- #
- # @uuid: uuid for this backup job
- #
-+# @finishing: if status='active' and finishing=true, then the backup process is
-+#             waiting for the target to finish.
-+#
- ##
- { 'struct': 'BackupStatus',
-   'data': {'*status': 'str', '*errmsg': 'str', '*total': 'int', '*dirty': 'int',
-            '*transferred': 'int', '*zero-bytes': 'int', '*reused': 'int',
-            '*start-time': 'int', '*end-time': 'int',
--           '*backup-file': 'str', '*uuid': 'str' } }
-+           '*backup-file': 'str', '*uuid': 'str', 'finishing': 'bool' } }
- 
- ##
- # @BackupFormat:

debian/patches/pve/0039-Revert-block-rbd-workaround-for-ceph-issue-53784.patch

@@ -7,15 +7,16 @@ This reverts commit fc176116cdea816ceb8dd969080b2b95f58edbc0 in
 preparation to revert 0347a8fd4c3faaedf119be04c197804be40a384b.
 
 Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  block/rbd.c | 42 ++----------------------------------------
  1 file changed, 2 insertions(+), 40 deletions(-)
 
 diff --git a/block/rbd.c b/block/rbd.c
-index 64a8d7d48b..9fc6dcb957 100644
+index bf143fac00..70d92966f7 100644
 --- a/block/rbd.c
 +++ b/block/rbd.c
-@@ -1348,7 +1348,6 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
+@@ -1515,7 +1515,6 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
      int status, r;
      RBDDiffIterateReq req = { .offs = offset };
      uint64_t features, flags;
@@ -23,7 +24,7 @@ index 64a8d7d48b..9fc6dcb957 100644
  
      assert(offset + bytes <= s->image_size);
  
-@@ -1376,43 +1375,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
+@@ -1543,43 +1542,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
          return status;
      }
  
@@ -68,7 +69,7 @@ index 64a8d7d48b..9fc6dcb957 100644
                            qemu_rbd_diff_iterate_cb, &req);
      if (r < 0 && r != QEMU_RBD_EXIT_DIFF_ITERATE2) {
          return status;
-@@ -1431,8 +1394,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
+@@ -1598,8 +1561,7 @@ static int coroutine_fn qemu_rbd_co_block_status(BlockDriverState *bs,
          status = BDRV_BLOCK_ZERO | BDRV_BLOCK_OFFSET_VALID;
      }
  

debian/patches/pve/0040-Revert-block-rbd-fix-handling-of-holes-in-.bdrv_co_b.patch

@@ -8,15 +8,16 @@ This reverts commit 9e302f64bb407a9bb097b626da97228c2654cfee in
 preparation to revert 0347a8fd4c3faaedf119be04c197804be40a384b.
 
 Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  block/rbd.c | 10 +++++-----
  1 file changed, 5 insertions(+), 5 deletions(-)
 
 diff --git a/block/rbd.c b/block/rbd.c
-index 9fc6dcb957..98f4ba2620 100644
+index 70d92966f7..931b513828 100644
 --- a/block/rbd.c
 +++ b/block/rbd.c
-@@ -1307,11 +1307,11 @@ static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
+@@ -1474,11 +1474,11 @@ static int qemu_rbd_diff_iterate_cb(uint64_t offs, size_t len,
      RBDDiffIterateReq *req = opaque;
  
      assert(req->offs + req->bytes <= offs);

debian/patches/pve/0041-Revert-block-rbd-implement-bdrv_co_block_status.patch

@@ -18,15 +18,16 @@ Upstream bug report:
 https://gitlab.com/qemu-project/qemu/-/issues/1026
 
 Signed-off-by: Fabian Ebner <f.ebner@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
 ---
  block/rbd.c | 112 ----------------------------------------------------
  1 file changed, 112 deletions(-)
 
 diff --git a/block/rbd.c b/block/rbd.c
-index 98f4ba2620..efcbbe5949 100644
+index 931b513828..4ab9bb5e02 100644
 --- a/block/rbd.c
 +++ b/block/rbd.c
-@@ -97,12 +97,6 @@ typedef struct RBDTask {
+@@ -108,12 +108,6 @@ typedef struct RBDTask {
      int64_t ret;
  } RBDTask;
  
@@ -39,7 +40,7 @@ index 98f4ba2620..efcbbe5949 100644
  static int qemu_rbd_connect(rados_t *cluster, rados_ioctx_t *io_ctx,
                              BlockdevOptionsRbd *opts, bool cache,
                              const char *keypairs, const char *secretid,
-@@ -1293,111 +1287,6 @@ static ImageInfoSpecific *qemu_rbd_get_specific_info(BlockDriverState *bs,
+@@ -1460,111 +1454,6 @@ static ImageInfoSpecific *qemu_rbd_get_specific_info(BlockDriverState *bs,
      return spec_info;
  }
  
@@ -148,10 +149,10 @@ index 98f4ba2620..efcbbe5949 100644
 -    return status;
 -}
 -
- static int64_t qemu_rbd_getlength(BlockDriverState *bs)
+ static int64_t coroutine_fn qemu_rbd_co_getlength(BlockDriverState *bs)
  {
      BDRVRBDState *s = bs->opaque;
-@@ -1633,7 +1522,6 @@ static BlockDriver bdrv_rbd = {
+@@ -1801,7 +1690,6 @@ static BlockDriver bdrv_rbd = {
  #ifdef LIBRBD_SUPPORTS_WRITE_ZEROES
      .bdrv_co_pwrite_zeroes  = qemu_rbd_co_pwrite_zeroes,
  #endif

debian/patches/pve/0042-PVE-backup-add-fleecing-option.patch

@@ -0,0 +1,472 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Thu, 11 Apr 2024 11:29:28 +0200
+Subject: [PATCH] PVE backup: add fleecing option
+
+When a fleecing option is given, it is expected that each device has
+a corresponding "-fleecing" block device already attached, except for
+EFI disk and TPM state, where fleecing is never used.
+
+The following graph was adapted from [0] which also contains more
+details about fleecing.
+
+[guest]
+   |
+   | root
+   v                 file
+[copy-before-write]<------[snapshot-access]
+   |           |
+   | file      | target
+   v           v
+[source] [fleecing]
+
+For fleecing, a copy-before-write filter is inserted on top of the
+source node, as well as a snapshot-access node pointing to the filter
+node which allows to read the consistent state of the image at the
+time it was inserted. New guest writes are passed through the
+copy-before-write filter which will first copy over old data to the
+fleecing image in case that old data is still needed by the
+snapshot-access node.
+
+The backup process will sequentially read from the snapshot access,
+which has a bitmap and knows whether to read from the original image
+or the fleecing image to get the "snapshot" state, i.e. data from the
+source image at the time when the copy-before-write filter was
+inserted. After reading, the copied sections are discarded from the
+fleecing image to reduce space usage.
+
+All of this can be restricted by an initial dirty bitmap to parts of
+the source image that are required for an incremental backup.
+
+For discard to work, it is necessary that the fleecing image does not
+have a larger cluster size than the backup job granularity. Since
+querying that size does not always work, e.g. for RBD with krbd, the
+cluster size will not be reported, a minimum of 4 MiB is used. A job
+with PBS target already has at least this granularity, so it's just
+relevant for other targets. I.e. edge cases where this minimum is not
+enough should be very rare in practice. If ever necessary in the
+future, can still add a passed-in value for the backup QMP command to
+override.
+
+Additionally, the cbw-timeout and on-cbw-error=break-snapshot options
+are set when installing the copy-before-write filter and
+snapshot-access. When an error or timeout occurs, the problematic (and
+each further) snapshot operation will fail and thus cancel the backup
+instead of breaking the guest write.
+
+Note that job_id cannot be inferred from the snapshot-access bs because
+it has no parent, so just pass the one from the original bs.
+
+[0]: https://www.mail-archive.com/qemu-devel@nongnu.org/msg876056.html
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+Signed-off-by: Thomas Lamprecht <t.lamprecht@proxmox.com>
+[FE: improve error when cbw fails as reported by Friedrich Weber]
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ block/copy-before-write.c      |  18 ++--
+ block/copy-before-write.h      |   1 +
+ block/monitor/block-hmp-cmds.c |   1 +
+ pve-backup.c                   | 175 ++++++++++++++++++++++++++++++++-
+ qapi/block-core.json           |  10 +-
+ 5 files changed, 195 insertions(+), 10 deletions(-)
+
+diff --git a/block/copy-before-write.c b/block/copy-before-write.c
+index fd470f5f92..5c23b578ef 100644
+--- a/block/copy-before-write.c
++++ b/block/copy-before-write.c
+@@ -27,6 +27,7 @@
+ #include "qobject/qjson.h"
+ 
+ #include "system/block-backend.h"
++#include "qemu/atomic.h"
+ #include "qemu/cutils.h"
+ #include "qapi/error.h"
+ #include "block/block_int.h"
+@@ -75,7 +76,8 @@ typedef struct BDRVCopyBeforeWriteState {
+      * @snapshot_error is normally zero. But on first copy-before-write failure
+      * when @on_cbw_error == ON_CBW_ERROR_BREAK_SNAPSHOT, @snapshot_error takes
+      * value of this error (<0). After that all in-flight and further
+-     * snapshot-API requests will fail with that error.
++     * snapshot-API requests will fail with that error. To be accessed with
++     * atomics.
+      */
+     int snapshot_error;
+ } BDRVCopyBeforeWriteState;
+@@ -115,7 +117,7 @@ static coroutine_fn int cbw_do_copy_before_write(BlockDriverState *bs,
+         return 0;
+     }
+ 
+-    if (s->snapshot_error) {
++    if (qatomic_read(&s->snapshot_error)) {
+         return 0;
+     }
+ 
+@@ -139,9 +141,7 @@ static coroutine_fn int cbw_do_copy_before_write(BlockDriverState *bs,
+     WITH_QEMU_LOCK_GUARD(&s->lock) {
+         if (ret < 0) {
+             assert(s->on_cbw_error == ON_CBW_ERROR_BREAK_SNAPSHOT);
+-            if (!s->snapshot_error) {
+-                s->snapshot_error = ret;
+-            }
++            qatomic_cmpxchg(&s->snapshot_error, 0, ret);
+         } else {
+             bdrv_set_dirty_bitmap(s->done_bitmap, off, end - off);
+         }
+@@ -215,7 +215,7 @@ cbw_snapshot_read_lock(BlockDriverState *bs, int64_t offset, int64_t bytes,
+ 
+     QEMU_LOCK_GUARD(&s->lock);
+ 
+-    if (s->snapshot_error) {
++    if (qatomic_read(&s->snapshot_error)) {
+         g_free(req);
+         return NULL;
+     }
+@@ -595,6 +595,12 @@ void bdrv_cbw_drop(BlockDriverState *bs)
+     bdrv_unref(bs);
+ }
+ 
++int bdrv_cbw_snapshot_error(BlockDriverState *bs)
++{
++    BDRVCopyBeforeWriteState *s = bs->opaque;
++    return qatomic_read(&s->snapshot_error);
++}
++
+ static void cbw_init(void)
+ {
+     bdrv_register(&bdrv_cbw_filter);
+diff --git a/block/copy-before-write.h b/block/copy-before-write.h
+index 2a5d4ba693..969da3620f 100644
+--- a/block/copy-before-write.h
++++ b/block/copy-before-write.h
+@@ -44,5 +44,6 @@ BlockDriverState *bdrv_cbw_append(BlockDriverState *source,
+                                   BlockCopyState **bcs,
+                                   Error **errp);
+ void bdrv_cbw_drop(BlockDriverState *bs);
++int bdrv_cbw_snapshot_error(BlockDriverState *bs);
+ 
+ #endif /* COPY_BEFORE_WRITE_H */
+diff --git a/block/monitor/block-hmp-cmds.c b/block/monitor/block-hmp-cmds.c
+index 4f30f99644..66d16d342f 100644
+--- a/block/monitor/block-hmp-cmds.c
++++ b/block/monitor/block-hmp-cmds.c
+@@ -1045,6 +1045,7 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
+         NULL, NULL,
+         devlist, qdict_haskey(qdict, "speed"), speed,
+         false, 0, // BackupPerf max-workers
++        false, false, // fleecing
+         &error);
+ 
+     hmp_handle_error(mon, error);
+diff --git a/pve-backup.c b/pve-backup.c
+index 366b015589..9b66788ab5 100644
+--- a/pve-backup.c
++++ b/pve-backup.c
+@@ -7,6 +7,7 @@
+ #include "system/blockdev.h"
+ #include "block/block_int-global-state.h"
+ #include "block/blockjob.h"
++#include "block/copy-before-write.h"
+ #include "block/dirty-bitmap.h"
+ #include "block/graph-lock.h"
+ #include "qapi/qapi-commands-block.h"
+@@ -81,8 +82,15 @@ static void pvebackup_init(void)
+ // initialize PVEBackupState at startup
+ opts_init(pvebackup_init);
+ 
++typedef struct PVEBackupFleecingInfo {
++    BlockDriverState *bs;
++    BlockDriverState *cbw;
++    BlockDriverState *snapshot_access;
++} PVEBackupFleecingInfo;
++
+ typedef struct PVEBackupDevInfo {
+     BlockDriverState *bs;
++    PVEBackupFleecingInfo fleecing;
+     size_t size;
+     uint64_t block_size;
+     uint8_t dev_id;
+@@ -352,11 +360,44 @@ static void coroutine_fn pvebackup_co_complete_stream(void *opaque)
+     qemu_co_mutex_unlock(&backup_state.backup_mutex);
+ }
+ 
++static void cleanup_snapshot_access(PVEBackupDevInfo *di)
++{
++    if (di->fleecing.snapshot_access) {
++        bdrv_unref(di->fleecing.snapshot_access);
++        di->fleecing.snapshot_access = NULL;
++    }
++    if (di->fleecing.cbw) {
++        bdrv_cbw_drop(di->fleecing.cbw);
++        di->fleecing.cbw = NULL;
++    }
++}
++
+ static void pvebackup_complete_cb(void *opaque, int ret)
+ {
+     PVEBackupDevInfo *di = opaque;
+     di->completed_ret = ret;
+ 
++    if (di->fleecing.cbw) {
++        /*
++         * With fleecing, failure for cbw does not fail the guest write, but only sets the snapshot
++         * error, making further requests to the snapshot fail with EACCES, which then also fail the
++         * job. But that code is not the root cause and just confusing, so update it.
++         */
++        int snapshot_error = bdrv_cbw_snapshot_error(di->fleecing.cbw);
++        if (di->completed_ret == -EACCES && snapshot_error) {
++            di->completed_ret = snapshot_error;
++        }
++    }
++
++    /*
++     * Handle block-graph specific cleanup (for fleecing) outside of the coroutine, because the work
++     * won't be done as a coroutine anyways:
++     * - For snapshot_access, allows doing bdrv_unref() directly. Doing it via bdrv_co_unref() would
++     *   just spawn a BH calling bdrv_unref().
++     * - For cbw, draining would need to spawn a BH.
++     */
++    cleanup_snapshot_access(di);
++
+     /*
+      * Needs to happen outside of coroutine, because it takes the graph write lock.
+      */
+@@ -487,6 +528,65 @@ static int coroutine_fn pvebackup_co_add_config(
+     goto out;
+ }
+ 
++/*
++ * Setup a snapshot-access block node for a device with associated fleecing image.
++ */
++static int setup_snapshot_access(PVEBackupDevInfo *di, Error **errp)
++{
++    Error *local_err = NULL;
++
++    if (!di->fleecing.bs) {
++        error_setg(errp, "no associated fleecing image");
++        return -1;
++    }
++
++    QDict *cbw_opts = qdict_new();
++    qdict_put_str(cbw_opts, "driver", "copy-before-write");
++    qdict_put_str(cbw_opts, "file", bdrv_get_node_name(di->bs));
++    qdict_put_str(cbw_opts, "target", bdrv_get_node_name(di->fleecing.bs));
++
++    if (di->bitmap) {
++        /*
++         * Only guest writes to parts relevant for the backup need to be intercepted with
++         * old data being copied to the fleecing image.
++         */
++        qdict_put_str(cbw_opts, "bitmap.node", bdrv_get_node_name(di->bs));
++        qdict_put_str(cbw_opts, "bitmap.name", bdrv_dirty_bitmap_name(di->bitmap));
++    }
++    /*
++     * Fleecing storage is supposed to be fast and it's better to break backup than guest
++     * writes. Certain guest drivers like VirtIO-win have 60 seconds timeout by default, so
++     * abort a bit before that.
++     */
++    qdict_put_str(cbw_opts, "on-cbw-error", "break-snapshot");
++    qdict_put_int(cbw_opts, "cbw-timeout", 45);
++
++    di->fleecing.cbw = bdrv_insert_node(di->bs, cbw_opts, BDRV_O_RDWR, &local_err);
++
++    if (!di->fleecing.cbw) {
++        error_setg(errp, "appending cbw node for fleecing failed: %s",
++                   local_err ? error_get_pretty(local_err) : "unknown error");
++        return -1;
++    }
++
++    QDict *snapshot_access_opts = qdict_new();
++    qdict_put_str(snapshot_access_opts, "driver", "snapshot-access");
++    qdict_put_str(snapshot_access_opts, "file", bdrv_get_node_name(di->fleecing.cbw));
++
++    di->fleecing.snapshot_access =
++        bdrv_open(NULL, NULL, snapshot_access_opts, BDRV_O_RDWR | BDRV_O_UNMAP, &local_err);
++    if (!di->fleecing.snapshot_access) {
++        bdrv_cbw_drop(di->fleecing.cbw);
++        di->fleecing.cbw = NULL;
++
++        error_setg(errp, "setting up snapshot access for fleecing failed: %s",
++                   local_err ? error_get_pretty(local_err) : "unknown error");
++        return -1;
++    }
++
++    return 0;
++}
++
+ /*
+  * backup_job_create can *not* be run from a coroutine, so this can't either.
+  * The caller is responsible that backup_mutex is held nonetheless.
+@@ -523,9 +623,42 @@ static void create_backup_jobs_bh(void *opaque) {
+         }
+         bdrv_drained_begin(di->bs);
+ 
++        BackupPerf perf = (BackupPerf){ .max_workers = backup_state.perf.max_workers };
++
++        BlockDriverState *source_bs = di->bs;
++        bool discard_source = false;
++        if (di->fleecing.bs) {
++            if (setup_snapshot_access(di, &local_err) < 0) {
++                error_setg(errp, "%s - setting up snapshot access for fleecing failed: %s",
++                           di->device_name,
++                           local_err ? error_get_pretty(local_err) : "unknown error");
++                bdrv_drained_end(di->bs);
++                break;
++            }
++
++            source_bs = di->fleecing.snapshot_access;
++            discard_source = true;
++
++            /*
++             * bdrv_get_info() just retuns 0 (= doesn't matter) for RBD when using krbd. But discard
++             * on the fleecing image won't work if the backup job's granularity is less than the RBD
++             * object size (default 4 MiB), so it does matter. Always use at least 4 MiB. With a PBS
++             * target, the backup job granularity would already be at least this much.
++             */
++            perf.min_cluster_size = 4 * 1024 * 1024;
++            /*
++             * For discard to work, cluster size for the backup job must be at least the same as for
++             * the fleecing image.
++             */
++            BlockDriverInfo bdi;
++            if (bdrv_get_info(di->fleecing.bs, &bdi) >= 0) {
++                perf.min_cluster_size = MAX(perf.min_cluster_size, bdi.cluster_size);
++            }
++        }
++
+         BlockJob *job = backup_job_create(
+-            NULL, di->bs, di->target, backup_state.speed, sync_mode, di->bitmap,
+-            bitmap_mode, false, NULL, &backup_state.perf, BLOCKDEV_ON_ERROR_REPORT,
++            di->device_name, source_bs, di->target, backup_state.speed, sync_mode, di->bitmap,
++            bitmap_mode, false, discard_source, NULL, &perf, BLOCKDEV_ON_ERROR_REPORT,
+             BLOCKDEV_ON_ERROR_REPORT, JOB_DEFAULT, pvebackup_complete_cb, di, backup_state.txn,
+             &local_err);
+ 
+@@ -539,6 +672,7 @@ static void create_backup_jobs_bh(void *opaque) {
+         }
+ 
+         if (!job || local_err) {
++            cleanup_snapshot_access(di);
+             error_setg(errp, "backup_job_create failed: %s",
+                        local_err ? error_get_pretty(local_err) : "null");
+             break;
+@@ -581,6 +715,14 @@ static void create_backup_jobs_bh(void *opaque) {
+     aio_co_enter(data->ctx, data->co);
+ }
+ 
++/*
++ * EFI disk and TPM state are small and it's just not worth setting up fleecing for them.
++ */
++static bool device_uses_fleecing(const char *device_id)
++{
++    return strncmp(device_id, "drive-efidisk", 13) && strncmp(device_id, "drive-tpmstate", 14);
++}
++
+ /*
+  * Returns a list of device infos, which needs to be freed by the caller. In
+  * case of an error, errp will be set, but the returned value might still be a
+@@ -588,6 +730,7 @@ static void create_backup_jobs_bh(void *opaque) {
+  */
+ static GList coroutine_fn GRAPH_RDLOCK *get_device_info(
+     const char *devlist,
++    bool fleecing,
+     Error **errp)
+ {
+     gchar **devs = NULL;
+@@ -613,6 +756,30 @@ static GList coroutine_fn GRAPH_RDLOCK *get_device_info(
+             di->bs = bs;
+             di->device_name = g_strdup(bdrv_get_device_name(bs));
+ 
++            if (fleecing && device_uses_fleecing(*d)) {
++                g_autofree gchar *fleecing_devid = g_strconcat(*d, "-fleecing", NULL);
++                BlockBackend *fleecing_blk = blk_by_name(fleecing_devid);
++                if (!fleecing_blk) {
++                    error_set(errp, ERROR_CLASS_DEVICE_NOT_FOUND,
++                              "Device '%s' not found", fleecing_devid);
++                    goto err;
++                }
++                BlockDriverState *fleecing_bs = blk_bs(fleecing_blk);
++                if (!bdrv_co_is_inserted(fleecing_bs)) {
++                    error_setg(errp, "Device '%s' has no medium", fleecing_devid);
++                    goto err;
++                }
++                /*
++                 * Fleecing image needs to be the same size to act as a cbw target.
++                 */
++                if (bs->total_sectors != fleecing_bs->total_sectors) {
++                    error_setg(errp, "Size mismatch for '%s' - sector count %ld != %ld",
++                               fleecing_devid, fleecing_bs->total_sectors, bs->total_sectors);
++                    goto err;
++                }
++                di->fleecing.bs = fleecing_bs;
++            }
++
+             di_list = g_list_append(di_list, di);
+             d++;
+         }
+@@ -663,6 +830,7 @@ UuidInfo coroutine_fn *qmp_backup(
+     const char *devlist,
+     bool has_speed, int64_t speed,
+     bool has_max_workers, int64_t max_workers,
++    bool has_fleecing, bool fleecing,
+     Error **errp)
+ {
+     assert(qemu_in_coroutine());
+@@ -691,7 +859,7 @@ UuidInfo coroutine_fn *qmp_backup(
+     format = has_format ? format : BACKUP_FORMAT_VMA;
+ 
+     bdrv_graph_co_rdlock();
+-    di_list = get_device_info(devlist, &local_err);
++    di_list = get_device_info(devlist, has_fleecing && fleecing, &local_err);
+     bdrv_graph_co_rdunlock();
+     if (local_err) {
+         error_propagate(errp, local_err);
+@@ -1093,5 +1261,6 @@ ProxmoxSupportStatus *qmp_query_proxmox_support(Error **errp)
+     ret->query_bitmap_info = true;
+     ret->pbs_masterkey = true;
+     ret->backup_max_workers = true;
++    ret->backup_fleecing = true;
+     return ret;
+ }
+diff --git a/qapi/block-core.json b/qapi/block-core.json
+index 6838187607..9bdcfa31ea 100644
+--- a/qapi/block-core.json
++++ b/qapi/block-core.json
+@@ -952,6 +952,10 @@
+ #
+ # @max-workers: see @BackupPerf for details. Default 16.
+ #
++# @fleecing: perform a backup with fleecing. For each device in @devlist, a
++#            corresponing '-fleecing' device with the same size already needs to
++#            be present.
++#
+ # Returns: the uuid of the backup job
+ #
+ ##
+@@ -972,7 +976,8 @@
+                                     '*firewall-file': 'str',
+                                     '*devlist': 'str',
+                                     '*speed': 'int',
+-                                    '*max-workers': 'int' },
++                                    '*max-workers': 'int',
++                                    '*fleecing': 'bool' },
+   'returns': 'UuidInfo', 'coroutine': true }
+ 
+ ##
+@@ -1018,6 +1023,8 @@
+ #
+ # @pbs-library-version: Running version of libproxmox-backup-qemu0 library.
+ #
++# @backup-fleecing: Whether backup fleecing is supported or not.
++#
+ # @backup-max-workers: Whether the 'max-workers' @BackupPerf setting is
+ #     supported or not.
+ #
+@@ -1029,6 +1036,7 @@
+             'pbs-dirty-bitmap-migration': 'bool',
+             'pbs-masterkey': 'bool',
+             'pbs-library-version': 'str',
++            'backup-fleecing': 'bool',
+             'backup-max-workers': 'bool' } }
+ 
+ ##

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

@@ -1,598 +0,0 @@
-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 a09f722fea..71ed202491 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1016,7 +1016,7 @@ void hmp_info_snapshots(Monitor *mon, const QDict *qdict)
-     g_free(global_snapshots);
- }
- 
--void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
-+void coroutine_fn hmp_backup_cancel(Monitor *mon, const QDict *qdict)
- {
-     Error *error = NULL;
- 
-@@ -1025,7 +1025,7 @@ void hmp_backup_cancel(Monitor *mon, const QDict *qdict)
-     hmp_handle_error(mon, error);
- }
- 
--void hmp_backup(Monitor *mon, const QDict *qdict)
-+void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
- {
-     Error *error = NULL;
- 
-diff --git a/hmp-commands.hx b/hmp-commands.hx
-index 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/0043-adapt-machine-version-deprecation-for-Proxmox-VE.patch

@@ -0,0 +1,137 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Fri, 3 Jan 2025 14:03:12 +0100
+Subject: [PATCH] adapt machine version deprecation for Proxmox VE
+
+In commit a35f8577a0 ("include/hw: add macros for deprecation &
+removal of versioned machines"), a new machine version deprecation and
+removal policy was introduced. After only 3 years a machine version
+will be deprecated while being removed after 6 years.
+
+The deprecation is a bit early considering major PVE releases are
+approximately every 2 years. This means that a deprecation warning can
+already happen for a machine version that was introduced during the
+previous major release. This would scare users for no good reason, so
+avoid deprecating machine versions in PVE too early and define a
+baseline of machine versions that will be supported throughout a
+single major PVE release.
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ include/hw/boards.h | 78 +++++++++++++++++++++++++++++----------------
+ 1 file changed, 51 insertions(+), 27 deletions(-)
+
+diff --git a/include/hw/boards.h b/include/hw/boards.h
+index 8ada4d5832..f9f3b75284 100644
+--- a/include/hw/boards.h
++++ b/include/hw/boards.h
+@@ -636,42 +636,66 @@ struct MachineState {
+ 
+ 
+ /*
+- * How many years/major releases for each phase
+- * of the life cycle. Assumes use of versioning
+- * scheme where major is bumped each year
++ * Baseline of machine versions that are still considered supported throughout
++ * current major Proxmox VE release. Machine versions older than this are
++ * considered to be deprecated in Proxmox VE.
++ *
++ * Machine versions older than 6 years are removed just like in upstream QEMU.
++ * (policy takes effect with QEMU 10.1). Assumes yearly major QEMU release.
++ *
++ * QEMU release cylce N.0 in ~April, N.1 in ~August, N.2 in ~December
++ * Debian/PVE release cylce ~every two years in summer
++ *
++ * PVE - last QEMU - machine versions dropped - baseline
++ *   8         9.2              2.3 and older        2.4
++ *   9        11.2              5.2 and older        6.0
++ *  10        13.2              7.2 and older        8.0
++ */
++#define MACHINE_VER_BASELINE_PVE_MAJOR 2
++#define MACHINE_VER_BASELINE_PVE_MINOR 4
++#define MACHINE_VER_DELETION_MAJOR (QEMU_VERSION_MAJOR - 6)
++#define MACHINE_VER_DELETION_MINOR QEMU_VERSION_MINOR
++
++/*
++ * Proxmox VE needs to support the baseline throughout a major PVE release. So
++ * a QEMU release where the baseline is already deleted cannot be used.
++ * Removal policy after 6 years takes effect with QEMU 10.1.
+  */
+-#define MACHINE_VER_DELETION_MAJOR 6
+-#define MACHINE_VER_DEPRECATION_MAJOR 3
++#if ((QEMU_VERSION_MAJOR > 10) || ((QEMU_VERSION_MAJOR == 10) && (QEMU_VERSION_MINOR >= 1)))
++#if ((MACHINE_VER_BASELINE_PVE_MAJOR < MACHINE_VER_DELETION_MAJOR) || \
++     ((MACHINE_VER_BASELINE_PVE_MAJOR == MACHINE_VER_DELETION_MAJOR) && \
++      (MACHINE_VER_BASELINE_PVE_MINOR < MACHINE_VER_DELETION_MINOR)))
++#error "Baseline machine version needed by Proxmox VE not supported anymore by this QEMU release"
++#endif
++#endif
+ 
+ /*
+  * Expands to a static string containing a deprecation
+  * message for a versioned machine type
+  */
+ #define MACHINE_VER_DEPRECATION_MSG \
+-    "machines more than " stringify(MACHINE_VER_DEPRECATION_MAJOR) \
+-    " years old are subject to deletion after " \
+-    stringify(MACHINE_VER_DELETION_MAJOR) " years"
+-
+-#define _MACHINE_VER_IS_EXPIRED_IMPL(cutoff, major, minor) \
+-    (((QEMU_VERSION_MAJOR - major) > cutoff) || \
+-     (((QEMU_VERSION_MAJOR - major) == cutoff) && \
+-      (QEMU_VERSION_MINOR - minor) >= 0))
+-
+-#define _MACHINE_VER_IS_EXPIRED2(cutoff, major, minor) \
+-    _MACHINE_VER_IS_EXPIRED_IMPL(cutoff, major, minor)
+-#define _MACHINE_VER_IS_EXPIRED3(cutoff, major, minor, micro) \
+-    _MACHINE_VER_IS_EXPIRED_IMPL(cutoff, major, minor)
+-#define _MACHINE_VER_IS_EXPIRED4(cutoff, major, minor, _unused, tag) \
+-    _MACHINE_VER_IS_EXPIRED_IMPL(cutoff, major, minor)
+-#define _MACHINE_VER_IS_EXPIRED5(cutoff, major, minor, micro, _unused, tag)   \
+-    _MACHINE_VER_IS_EXPIRED_IMPL(cutoff, major, minor)
+-
+-#define _MACHINE_IS_EXPIRED(cutoff, ...) \
++    "old machine version is subject to deletion during current major Proxmox VE release"
++
++#define _MACHINE_VER_IS_EXPIRED_IMPL(baseline_major, baseline_minor, major, minor) \
++    ((major < baseline_major) || \
++     ((major == baseline_major) && \
++      (minor < baseline_minor)))
++
++#define _MACHINE_VER_IS_EXPIRED2(baseline_major, baseline_minor, major, minor) \
++    _MACHINE_VER_IS_EXPIRED_IMPL(baseline_major, baseline_minor, major, minor)
++#define _MACHINE_VER_IS_EXPIRED3(baseline_major, baseline_minor, major, minor, micro) \
++    _MACHINE_VER_IS_EXPIRED_IMPL(baseline_major, baseline_minor, major, minor)
++#define _MACHINE_VER_IS_EXPIRED4(baseline_major, baseline_minor, major, minor, _unused, tag) \
++    _MACHINE_VER_IS_EXPIRED_IMPL(baseline_major, baseline_minor, major, minor)
++#define _MACHINE_VER_IS_EXPIRED5(baseline_major, baseline_minor, major, minor, micro, _unused, tag)   \
++    _MACHINE_VER_IS_EXPIRED_IMPL(baseline_major, baseline_minor, major, minor)
++
++#define _MACHINE_IS_EXPIRED(baseline_major, baseline_minor, ...) \
+     _MACHINE_VER_PICK(__VA_ARGS__, \
+                       _MACHINE_VER_IS_EXPIRED5, \
+                       _MACHINE_VER_IS_EXPIRED4, \
+                       _MACHINE_VER_IS_EXPIRED3, \
+-                      _MACHINE_VER_IS_EXPIRED2) (cutoff, __VA_ARGS__)
++                      _MACHINE_VER_IS_EXPIRED2) (baseline_major, baseline_minor, __VA_ARGS__)
+ 
+ /*
+  * Evaluates true when a machine type with (major, minor)
+@@ -680,7 +704,7 @@ struct MachineState {
+  * lifecycle rules
+  */
+ #define MACHINE_VER_IS_DEPRECATED(...) \
+-    _MACHINE_IS_EXPIRED(MACHINE_VER_DEPRECATION_MAJOR, __VA_ARGS__)
++    _MACHINE_IS_EXPIRED(MACHINE_VER_BASELINE_PVE_MAJOR, MACHINE_VER_BASELINE_PVE_MINOR, __VA_ARGS__)
+ 
+ /*
+  * Evaluates true when a machine type with (major, minor)
+@@ -689,7 +713,7 @@ struct MachineState {
+  * lifecycle rules
+  */
+ #define MACHINE_VER_SHOULD_DELETE(...) \
+-    _MACHINE_IS_EXPIRED(MACHINE_VER_DELETION_MAJOR, __VA_ARGS__)
++    _MACHINE_IS_EXPIRED(MACHINE_VER_DELETION_MAJOR, MACHINE_VER_DELETION_MINOR, __VA_ARGS__)
+ 
+ /*
+  * Sets the deprecation reason for a versioned machine based

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

@@ -1,98 +0,0 @@
-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 71ed202491..c7468e5d3b 100644
---- a/block/monitor/block-hmp-cmds.c
-+++ b/block/monitor/block-hmp-cmds.c
-@@ -1039,6 +1039,7 @@ void coroutine_fn hmp_backup(Monitor *mon, const QDict *qdict)
-         false, NULL, // PBS password
-         false, NULL, // PBS keyfile
-         false, NULL, // PBS key_password
-+        false, NULL, // PBS master_keyfile
-         false, NULL, // PBS fingerprint
-         false, NULL, // PBS backup-id
-         false, 0, // PBS backup-time
-diff --git a/pve-backup.c b/pve-backup.c
-index 109498eaf9..4b5134ed27 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/0044-Revert-hpet-avoid-timer-storms-on-periodic-timers.patch

@@ -0,0 +1,50 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Wed, 19 Mar 2025 17:31:05 +0100
+Subject: [PATCH] Revert "hpet: avoid timer storms on periodic timers"
+
+This reverts commit 7c912ffb59e8137091894d767433e65c3df8b0bf.
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hw/timer/hpet.c | 13 ++-----------
+ 1 file changed, 2 insertions(+), 11 deletions(-)
+
+diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
+index ccb97b6806..0f45af8bbe 100644
+--- a/hw/timer/hpet.c
++++ b/hw/timer/hpet.c
+@@ -61,7 +61,6 @@ typedef struct HPETTimer {  /* timers */
+     uint8_t wrap_flag;      /* timer pop will indicate wrap for one-shot 32-bit
+                              * mode. Next pop will be actual timer expiration.
+                              */
+-    uint64_t last;          /* last value armed, to avoid timer storms */
+ } HPETTimer;
+ 
+ struct HPETState {
+@@ -262,7 +261,6 @@ static int hpet_post_load(void *opaque, int version_id)
+     for (i = 0; i < s->num_timers; i++) {
+         HPETTimer *t = &s->timer[i];
+         t->cmp64 = hpet_calculate_cmp64(t, s->hpet_counter, t->cmp);
+-        t->last = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) - NANOSECONDS_PER_SECOND;
+     }
+     /* Recalculate the offset between the main counter and guest time */
+     if (!s->hpet_offset_saved) {
+@@ -350,15 +348,8 @@ static const VMStateDescription vmstate_hpet = {
+ 
+ static void hpet_arm(HPETTimer *t, uint64_t tick)
+ {
+-    uint64_t ns = hpet_get_ns(t->state, tick);
+-
+-    /* Clamp period to reasonable min value (1 us) */
+-    if (timer_is_periodic(t) && ns - t->last < 1000) {
+-        ns = t->last + 1000;
+-    }
+-
+-    t->last = ns;
+-    timer_mod(t->qemu_timer, ns);
++    /* FIXME: Clamp period to reasonable min value? */
++    timer_mod(t->qemu_timer, hpet_get_ns(t->state, tick));
+ }
+ 
+ /*

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

@@ -1,53 +0,0 @@
-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-Revert-hpet-store-full-64-bit-target-value-of-the-co.patch

@@ -0,0 +1,202 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Wed, 19 Mar 2025 17:31:08 +0100
+Subject: [PATCH] Revert "hpet: store full 64-bit target value of the counter"
+
+This reverts commit 242d665396407f83a6acbffc804882eeb21cfdad.
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hw/timer/hpet.c | 109 +++++++++++++++++++++++++++---------------------
+ 1 file changed, 61 insertions(+), 48 deletions(-)
+
+diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
+index 0f45af8bbe..635a060d38 100644
+--- a/hw/timer/hpet.c
++++ b/hw/timer/hpet.c
+@@ -56,7 +56,6 @@ typedef struct HPETTimer {  /* timers */
+     uint64_t cmp;           /* comparator */
+     uint64_t fsb;           /* FSB route */
+     /* Hidden register state */
+-    uint64_t cmp64;         /* comparator (extended to counter width) */
+     uint64_t period;        /* Last value written to comparator */
+     uint8_t wrap_flag;      /* timer pop will indicate wrap for one-shot 32-bit
+                              * mode. Next pop will be actual timer expiration.
+@@ -119,6 +118,11 @@ static uint32_t timer_enabled(HPETTimer *t)
+ }
+ 
+ static uint32_t hpet_time_after(uint64_t a, uint64_t b)
++{
++    return ((int32_t)(b - a) < 0);
++}
++
++static uint32_t hpet_time_after64(uint64_t a, uint64_t b)
+ {
+     return ((int64_t)(b - a) < 0);
+ }
+@@ -155,32 +159,27 @@ static uint64_t hpet_get_ticks(HPETState *s)
+     return ns_to_ticks(qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + s->hpet_offset);
+ }
+ 
+-static uint64_t hpet_get_ns(HPETState *s, uint64_t tick)
+-{
+-    return ticks_to_ns(tick) - s->hpet_offset;
+-}
+-
+ /*
+- * calculate next value of the general counter that matches the
+- * target (either entirely, or the low 32-bit only depending on
+- * the timer mode).
++ * calculate diff between comparator value and current ticks
+  */
+-static uint64_t hpet_calculate_cmp64(HPETTimer *t, uint64_t cur_tick, uint64_t target)
++static inline uint64_t hpet_calculate_diff(HPETTimer *t, uint64_t current)
+ {
++
+     if (t->config & HPET_TN_32BIT) {
+-        uint64_t result = deposit64(cur_tick, 0, 32, target);
+-        if (result < cur_tick) {
+-            result += 0x100000000ULL;
+-        }
+-        return result;
++        uint32_t diff, cmp;
++
++        cmp = (uint32_t)t->cmp;
++        diff = cmp - (uint32_t)current;
++        diff = (int32_t)diff > 0 ? diff : (uint32_t)1;
++        return (uint64_t)diff;
+     } else {
+-        return target;
+-    }
+-}
++        uint64_t diff, cmp;
+ 
+-static uint64_t hpet_next_wrap(uint64_t cur_tick)
+-{
+-    return (cur_tick | 0xffffffffU) + 1;
++        cmp = t->cmp;
++        diff = cmp - current;
++        diff = (int64_t)diff > 0 ? diff : (uint64_t)1;
++        return diff;
++    }
+ }
+ 
+ static void update_irq(struct HPETTimer *timer, int set)
+@@ -256,12 +255,7 @@ static bool hpet_validate_num_timers(void *opaque, int version_id)
+ static int hpet_post_load(void *opaque, int version_id)
+ {
+     HPETState *s = opaque;
+-    int i;
+ 
+-    for (i = 0; i < s->num_timers; i++) {
+-        HPETTimer *t = &s->timer[i];
+-        t->cmp64 = hpet_calculate_cmp64(t, s->hpet_counter, t->cmp);
+-    }
+     /* Recalculate the offset between the main counter and guest time */
+     if (!s->hpet_offset_saved) {
+         s->hpet_offset = ticks_to_ns(s->hpet_counter)
+@@ -346,10 +340,14 @@ static const VMStateDescription vmstate_hpet = {
+     }
+ };
+ 
+-static void hpet_arm(HPETTimer *t, uint64_t tick)
++static void hpet_arm(HPETTimer *t, uint64_t ticks)
+ {
+-    /* FIXME: Clamp period to reasonable min value? */
+-    timer_mod(t->qemu_timer, hpet_get_ns(t->state, tick));
++    if (ticks < ns_to_ticks(INT64_MAX / 2)) {
++        timer_mod(t->qemu_timer,
++                  qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) + ticks_to_ns(ticks));
++    } else {
++        timer_del(t->qemu_timer);
++    }
+ }
+ 
+ /*
+@@ -358,44 +356,54 @@ static void hpet_arm(HPETTimer *t, uint64_t tick)
+ static void hpet_timer(void *opaque)
+ {
+     HPETTimer *t = opaque;
++    uint64_t diff;
++
+     uint64_t period = t->period;
+     uint64_t cur_tick = hpet_get_ticks(t->state);
+ 
+     if (timer_is_periodic(t) && period != 0) {
+-        while (hpet_time_after(cur_tick, t->cmp64)) {
+-            t->cmp64 += period;
+-        }
+         if (t->config & HPET_TN_32BIT) {
+-            t->cmp = (uint32_t)t->cmp64;
++            while (hpet_time_after(cur_tick, t->cmp)) {
++                t->cmp = (uint32_t)(t->cmp + t->period);
++            }
+         } else {
+-            t->cmp = t->cmp64;
++            while (hpet_time_after64(cur_tick, t->cmp)) {
++                t->cmp += period;
++            }
++        }
++        diff = hpet_calculate_diff(t, cur_tick);
++        hpet_arm(t, diff);
++    } else if (t->config & HPET_TN_32BIT && !timer_is_periodic(t)) {
++        if (t->wrap_flag) {
++            diff = hpet_calculate_diff(t, cur_tick);
++            hpet_arm(t, diff);
++            t->wrap_flag = 0;
+         }
+-        hpet_arm(t, t->cmp64);
+-    } else if (t->wrap_flag) {
+-        t->wrap_flag = 0;
+-        hpet_arm(t, t->cmp64);
+     }
+     update_irq(t, 1);
+ }
+ 
+ static void hpet_set_timer(HPETTimer *t)
+ {
++    uint64_t diff;
++    uint32_t wrap_diff;  /* how many ticks until we wrap? */
+     uint64_t cur_tick = hpet_get_ticks(t->state);
+ 
++    /* whenever new timer is being set up, make sure wrap_flag is 0 */
+     t->wrap_flag = 0;
+-    t->cmp64 = hpet_calculate_cmp64(t, cur_tick, t->cmp);
+-    if (t->config & HPET_TN_32BIT) {
++    diff = hpet_calculate_diff(t, cur_tick);
+ 
+-        /* hpet spec says in one-shot 32-bit mode, generate an interrupt when
+-         * counter wraps in addition to an interrupt with comparator match.
+-         */
+-        if (!timer_is_periodic(t) && t->cmp64 > hpet_next_wrap(cur_tick)) {
++    /* hpet spec says in one-shot 32-bit mode, generate an interrupt when
++     * counter wraps in addition to an interrupt with comparator match.
++     */
++    if (t->config & HPET_TN_32BIT && !timer_is_periodic(t)) {
++        wrap_diff = 0xffffffff - (uint32_t)cur_tick;
++        if (wrap_diff < (uint32_t)diff) {
++            diff = wrap_diff;
+             t->wrap_flag = 1;
+-            hpet_arm(t, hpet_next_wrap(cur_tick));
+-            return;
+         }
+     }
+-    hpet_arm(t, t->cmp64);
++    hpet_arm(t, diff);
+ }
+ 
+ static void hpet_del_timer(HPETTimer *t)
+@@ -526,7 +534,12 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+                 timer->cmp = deposit64(timer->cmp, shift, len, value);
+             }
+             if (timer_is_periodic(timer)) {
+-                timer->period = deposit64(timer->period, shift, len, value);
++                /*
++                 * FIXME: Clamp period to reasonable min value?
++                 * Clamp period to reasonable max value
++                 */
++                new_val = deposit64(timer->period, shift, len, value);
++                timer->period = MIN(new_val, (timer->config & HPET_TN_32BIT ? ~0u : ~0ull) >> 1);
+             }
+             timer->config &= ~HPET_TN_SETVAL;
+             if (hpet_enabled(s)) {

debian/patches/pve/0046-Revert-hpet-accept-64-bit-reads-and-writes.patch

@@ -0,0 +1,281 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Wed, 19 Mar 2025 17:31:09 +0100
+Subject: [PATCH] Revert "hpet: accept 64-bit reads and writes"
+
+This reverts commit c2366567378dd8fb89329816003801f54e30e6f3.
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hw/timer/hpet.c       | 137 +++++++++++++++++++++++++++++-------------
+ hw/timer/trace-events |   3 +-
+ 2 files changed, 96 insertions(+), 44 deletions(-)
+
+diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
+index 635a060d38..5f4bb5667d 100644
+--- a/hw/timer/hpet.c
++++ b/hw/timer/hpet.c
+@@ -421,7 +421,6 @@ static uint64_t hpet_ram_read(void *opaque, hwaddr addr,
+                               unsigned size)
+ {
+     HPETState *s = opaque;
+-    int shift = (addr & 4) * 8;
+     uint64_t cur_tick;
+ 
+     trace_hpet_ram_read(addr);
+@@ -436,33 +435,52 @@ static uint64_t hpet_ram_read(void *opaque, hwaddr addr,
+             return 0;
+         }
+ 
+-        switch (addr & 0x18) {
+-        case HPET_TN_CFG: // including interrupt capabilities
+-            return timer->config >> shift;
++        switch ((addr - 0x100) % 0x20) {
++        case HPET_TN_CFG:
++            return timer->config;
++        case HPET_TN_CFG + 4: // Interrupt capabilities
++            return timer->config >> 32;
+         case HPET_TN_CMP: // comparator register
+-            return timer->cmp >> shift;
++            return timer->cmp;
++        case HPET_TN_CMP + 4:
++            return timer->cmp >> 32;
+         case HPET_TN_ROUTE:
+-            return timer->fsb >> shift;
++            return timer->fsb;
++        case HPET_TN_ROUTE + 4:
++            return timer->fsb >> 32;
+         default:
+             trace_hpet_ram_read_invalid();
+             break;
+         }
+     } else {
+-        switch (addr & ~4) {
+-        case HPET_ID: // including HPET_PERIOD
+-            return s->capability >> shift;
++        switch (addr) {
++        case HPET_ID:
++            return s->capability;
++        case HPET_PERIOD:
++            return s->capability >> 32;
+         case HPET_CFG:
+-            return s->config >> shift;
++            return s->config;
++        case HPET_CFG + 4:
++            trace_hpet_invalid_hpet_cfg(4);
++            return 0;
+         case HPET_COUNTER:
+             if (hpet_enabled(s)) {
+                 cur_tick = hpet_get_ticks(s);
+             } else {
+                 cur_tick = s->hpet_counter;
+             }
+-            trace_hpet_ram_read_reading_counter(addr & 4, cur_tick);
+-            return cur_tick >> shift;
++            trace_hpet_ram_read_reading_counter(0, cur_tick);
++            return cur_tick;
++        case HPET_COUNTER + 4:
++            if (hpet_enabled(s)) {
++                cur_tick = hpet_get_ticks(s);
++            } else {
++                cur_tick = s->hpet_counter;
++            }
++            trace_hpet_ram_read_reading_counter(4, cur_tick);
++            return cur_tick >> 32;
+         case HPET_STATUS:
+-            return s->isr >> shift;
++            return s->isr;
+         default:
+             trace_hpet_ram_read_invalid();
+             break;
+@@ -476,11 +494,11 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+ {
+     int i;
+     HPETState *s = opaque;
+-    int shift = (addr & 4) * 8;
+-    int len = MIN(size * 8, 64 - shift);
+     uint64_t old_val, new_val, cleared;
+ 
+     trace_hpet_ram_write(addr, value);
++    old_val = hpet_ram_read(opaque, addr, 4);
++    new_val = value;
+ 
+     /*address range of all TN regs*/
+     if (addr >= 0x100 && addr <= 0x3ff) {
+@@ -492,12 +510,9 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+             trace_hpet_timer_id_out_of_range(timer_id);
+             return;
+         }
+-        switch (addr & 0x18) {
++        switch ((addr - 0x100) % 0x20) {
+         case HPET_TN_CFG:
+-            trace_hpet_ram_write_tn_cfg(addr & 4);
+-            old_val = timer->config;
+-            new_val = deposit64(old_val, shift, len, value);
+-            new_val = hpet_fixup_reg(new_val, old_val, HPET_TN_CFG_WRITE_MASK);
++            trace_hpet_ram_write_tn_cfg();
+             if (deactivating_bit(old_val, new_val, HPET_TN_TYPE_LEVEL)) {
+                 /*
+                  * Do this before changing timer->config; otherwise, if
+@@ -505,7 +520,8 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+                  */
+                 update_irq(timer, 0);
+             }
+-            timer->config = new_val;
++            new_val = hpet_fixup_reg(new_val, old_val, HPET_TN_CFG_WRITE_MASK);
++            timer->config = (timer->config & 0xffffffff00000000ULL) | new_val;
+             if (activating_bit(old_val, new_val, HPET_TN_ENABLE)
+                 && (s->isr & (1 << timer_id))) {
+                 update_irq(timer, 1);
+@@ -518,28 +534,56 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+                 hpet_set_timer(timer);
+             }
+             break;
++        case HPET_TN_CFG + 4: // Interrupt capabilities
++            trace_hpet_ram_write_invalid_tn_cfg(4);
++            break;
+         case HPET_TN_CMP: // comparator register
++            trace_hpet_ram_write_tn_cmp(0);
+             if (timer->config & HPET_TN_32BIT) {
+-                /* High 32-bits are zero, leave them untouched.  */
+-                if (shift) {
+-                    trace_hpet_ram_write_invalid_tn_cmp();
+-                    break;
++                new_val = (uint32_t)new_val;
++            }
++            if (!timer_is_periodic(timer)
++                || (timer->config & HPET_TN_SETVAL)) {
++                timer->cmp = (timer->cmp & 0xffffffff00000000ULL) | new_val;
++            }
++            if (timer_is_periodic(timer)) {
++                /*
++                 * FIXME: Clamp period to reasonable min value?
++                 * Clamp period to reasonable max value
++                 */
++                if (timer->config & HPET_TN_32BIT) {
++                    new_val = MIN(new_val, ~0u >> 1);
+                 }
+-                len = 64;
+-                value = (uint32_t) value;
++                timer->period =
++                    (timer->period & 0xffffffff00000000ULL) | new_val;
++            }
++            /*
++             * FIXME: on a 64-bit write, HPET_TN_SETVAL should apply to the
++             * high bits part as well.
++             */
++            timer->config &= ~HPET_TN_SETVAL;
++            if (hpet_enabled(s)) {
++                hpet_set_timer(timer);
+             }
+-            trace_hpet_ram_write_tn_cmp(addr & 4);
++            break;
++        case HPET_TN_CMP + 4: // comparator register high order
++            if (timer->config & HPET_TN_32BIT) {
++                trace_hpet_ram_write_invalid_tn_cmp();
++                break;
++            }
++            trace_hpet_ram_write_tn_cmp(4);
+             if (!timer_is_periodic(timer)
+                 || (timer->config & HPET_TN_SETVAL)) {
+-                timer->cmp = deposit64(timer->cmp, shift, len, value);
++                timer->cmp = (timer->cmp & 0xffffffffULL) | new_val << 32;
+             }
+             if (timer_is_periodic(timer)) {
+                 /*
+                  * FIXME: Clamp period to reasonable min value?
+                  * Clamp period to reasonable max value
+                  */
+-                new_val = deposit64(timer->period, shift, len, value);
+-                timer->period = MIN(new_val, (timer->config & HPET_TN_32BIT ? ~0u : ~0ull) >> 1);
++                new_val = MIN(new_val, ~0u >> 1);
++                timer->period =
++                    (timer->period & 0xffffffffULL) | new_val << 32;
+             }
+             timer->config &= ~HPET_TN_SETVAL;
+             if (hpet_enabled(s)) {
+@@ -547,7 +591,10 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+             }
+             break;
+         case HPET_TN_ROUTE:
+-            timer->fsb = deposit64(timer->fsb, shift, len, value);
++            timer->fsb = (timer->fsb & 0xffffffff00000000ULL) | new_val;
++            break;
++        case HPET_TN_ROUTE + 4:
++            timer->fsb = (new_val << 32) | (timer->fsb & 0xffffffff);
+             break;
+         default:
+             trace_hpet_ram_write_invalid();
+@@ -555,14 +602,12 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+         }
+         return;
+     } else {
+-        switch (addr & ~4) {
++        switch (addr) {
+         case HPET_ID:
+             return;
+         case HPET_CFG:
+-            old_val = s->config;
+-            new_val = deposit64(old_val, shift, len, value);
+             new_val = hpet_fixup_reg(new_val, old_val, HPET_CFG_WRITE_MASK);
+-            s->config = new_val;
++            s->config = (s->config & 0xffffffff00000000ULL) | new_val;
+             if (activating_bit(old_val, new_val, HPET_CFG_ENABLE)) {
+                 /* Enable main counter and interrupt generation. */
+                 s->hpet_offset =
+@@ -592,8 +637,10 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+                 qemu_set_irq(s->irqs[RTC_ISA_IRQ], s->rtc_irq_level);
+             }
+             break;
++        case HPET_CFG + 4:
++            trace_hpet_invalid_hpet_cfg(4);
++            break;
+         case HPET_STATUS:
+-            new_val = value << shift;
+             cleared = new_val & s->isr;
+             for (i = 0; i < s->num_timers; i++) {
+                 if (cleared & (1 << i)) {
+@@ -605,7 +652,15 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+             if (hpet_enabled(s)) {
+                 trace_hpet_ram_write_counter_write_while_enabled();
+             }
+-            s->hpet_counter = deposit64(s->hpet_counter, shift, len, value);
++            s->hpet_counter =
++                (s->hpet_counter & 0xffffffff00000000ULL) | value;
++            trace_hpet_ram_write_counter_written(0, value, s->hpet_counter);
++            break;
++        case HPET_COUNTER + 4:
++            trace_hpet_ram_write_counter_write_while_enabled();
++            s->hpet_counter =
++                (s->hpet_counter & 0xffffffffULL) | (((uint64_t)value) << 32);
++            trace_hpet_ram_write_counter_written(4, value, s->hpet_counter);
+             break;
+         default:
+             trace_hpet_ram_write_invalid();
+@@ -619,11 +674,7 @@ static const MemoryRegionOps hpet_ram_ops = {
+     .write = hpet_ram_write,
+     .valid = {
+         .min_access_size = 4,
+-        .max_access_size = 8,
+-    },
+-    .impl = {
+-        .min_access_size = 4,
+-        .max_access_size = 8,
++        .max_access_size = 4,
+     },
+     .endianness = DEVICE_NATIVE_ENDIAN,
+ };
+diff --git a/hw/timer/trace-events b/hw/timer/trace-events
+index c5b6db49f5..dd8a53c690 100644
+--- a/hw/timer/trace-events
++++ b/hw/timer/trace-events
+@@ -114,7 +114,8 @@ hpet_ram_read_reading_counter(uint8_t reg_off, uint64_t cur_tick) "reading count
+ hpet_ram_read_invalid(void) "invalid hpet_ram_readl"
+ hpet_ram_write(uint64_t addr, uint64_t value) "enter hpet_ram_writel at 0x%" PRIx64 " = 0x%" PRIx64
+ hpet_ram_write_timer_id(uint64_t timer_id) "hpet_ram_writel timer_id = 0x%" PRIx64
+-hpet_ram_write_tn_cfg(uint8_t reg_off) "hpet_ram_writel HPET_TN_CFG + %" PRIu8
++hpet_ram_write_tn_cfg(void) "hpet_ram_writel HPET_TN_CFG"
++hpet_ram_write_invalid_tn_cfg(uint8_t reg_off) "invalid HPET_TN_CFG + %" PRIu8 " write"
+ hpet_ram_write_tn_cmp(uint8_t reg_off) "hpet_ram_writel HPET_TN_CMP + %" PRIu8
+ hpet_ram_write_invalid_tn_cmp(void) "invalid HPET_TN_CMP + 4 write"
+ hpet_ram_write_invalid(void) "invalid hpet_ram_writel"

debian/patches/pve/0047-Revert-hpet-place-read-only-bits-directly-in-new_val.patch

@@ -0,0 +1,64 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Wed, 19 Mar 2025 17:31:10 +0100
+Subject: [PATCH] Revert "hpet: place read-only bits directly in "new_val""
+
+This reverts commit ba88935b0fac2588b0a739f810b58dfabf7f92c8.
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hw/timer/hpet.c | 15 ++++++++-------
+ 1 file changed, 8 insertions(+), 7 deletions(-)
+
+diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
+index 5f4bb5667d..5e3bf1f153 100644
+--- a/hw/timer/hpet.c
++++ b/hw/timer/hpet.c
+@@ -494,7 +494,7 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+ {
+     int i;
+     HPETState *s = opaque;
+-    uint64_t old_val, new_val, cleared;
++    uint64_t old_val, new_val, val;
+ 
+     trace_hpet_ram_write(addr, value);
+     old_val = hpet_ram_read(opaque, addr, 4);
+@@ -520,12 +520,13 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+                  */
+                 update_irq(timer, 0);
+             }
+-            new_val = hpet_fixup_reg(new_val, old_val, HPET_TN_CFG_WRITE_MASK);
+-            timer->config = (timer->config & 0xffffffff00000000ULL) | new_val;
++            val = hpet_fixup_reg(new_val, old_val, HPET_TN_CFG_WRITE_MASK);
++            timer->config = (timer->config & 0xffffffff00000000ULL) | val;
+             if (activating_bit(old_val, new_val, HPET_TN_ENABLE)
+                 && (s->isr & (1 << timer_id))) {
+                 update_irq(timer, 1);
+             }
++
+             if (new_val & HPET_TN_32BIT) {
+                 timer->cmp = (uint32_t)timer->cmp;
+                 timer->period = (uint32_t)timer->period;
+@@ -606,8 +607,8 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+         case HPET_ID:
+             return;
+         case HPET_CFG:
+-            new_val = hpet_fixup_reg(new_val, old_val, HPET_CFG_WRITE_MASK);
+-            s->config = (s->config & 0xffffffff00000000ULL) | new_val;
++            val = hpet_fixup_reg(new_val, old_val, HPET_CFG_WRITE_MASK);
++            s->config = (s->config & 0xffffffff00000000ULL) | val;
+             if (activating_bit(old_val, new_val, HPET_CFG_ENABLE)) {
+                 /* Enable main counter and interrupt generation. */
+                 s->hpet_offset =
+@@ -641,9 +642,9 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+             trace_hpet_invalid_hpet_cfg(4);
+             break;
+         case HPET_STATUS:
+-            cleared = new_val & s->isr;
++            val = new_val & s->isr;
+             for (i = 0; i < s->num_timers; i++) {
+-                if (cleared & (1 << i)) {
++                if (val & (1 << i)) {
+                     update_irq(&s->timer[i], 0);
+                 }
+             }

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

@@ -1,33 +0,0 @@
-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 b9424024f9..01f50d28c8 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-Revert-hpet-remove-unnecessary-variable-index.patch

@@ -0,0 +1,68 @@
+From 0000000000000000000000000000000000000000 Mon Sep 17 00:00:00 2001
+From: Fiona Ebner <f.ebner@proxmox.com>
+Date: Wed, 19 Mar 2025 17:31:11 +0100
+Subject: [PATCH] Revert "hpet: remove unnecessary variable "index""
+
+This reverts commit 5895879aca252f4ebb2d1078eaf836c61ec54e9b.
+
+Signed-off-by: Fiona Ebner <f.ebner@proxmox.com>
+---
+ hw/timer/hpet.c | 15 ++++++++-------
+ 1 file changed, 8 insertions(+), 7 deletions(-)
+
+diff --git a/hw/timer/hpet.c b/hw/timer/hpet.c
+index 5e3bf1f153..daef12c8cf 100644
+--- a/hw/timer/hpet.c
++++ b/hw/timer/hpet.c
+@@ -421,12 +421,12 @@ static uint64_t hpet_ram_read(void *opaque, hwaddr addr,
+                               unsigned size)
+ {
+     HPETState *s = opaque;
+-    uint64_t cur_tick;
++    uint64_t cur_tick, index;
+ 
+     trace_hpet_ram_read(addr);
+-
++    index = addr;
+     /*address range of all TN regs*/
+-    if (addr >= 0x100 && addr <= 0x3ff) {
++    if (index >= 0x100 && index <= 0x3ff) {
+         uint8_t timer_id = (addr - 0x100) / 0x20;
+         HPETTimer *timer = &s->timer[timer_id];
+ 
+@@ -453,7 +453,7 @@ static uint64_t hpet_ram_read(void *opaque, hwaddr addr,
+             break;
+         }
+     } else {
+-        switch (addr) {
++        switch (index) {
+         case HPET_ID:
+             return s->capability;
+         case HPET_PERIOD:
+@@ -494,14 +494,15 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+ {
+     int i;
+     HPETState *s = opaque;
+-    uint64_t old_val, new_val, val;
++    uint64_t old_val, new_val, val, index;
+ 
+     trace_hpet_ram_write(addr, value);
++    index = addr;
+     old_val = hpet_ram_read(opaque, addr, 4);
+     new_val = value;
+ 
+     /*address range of all TN regs*/
+-    if (addr >= 0x100 && addr <= 0x3ff) {
++    if (index >= 0x100 && index <= 0x3ff) {
+         uint8_t timer_id = (addr - 0x100) / 0x20;
+         HPETTimer *timer = &s->timer[timer_id];
+ 
+@@ -603,7 +604,7 @@ static void hpet_ram_write(void *opaque, hwaddr addr,
+         }
+         return;
+     } else {
+-        switch (addr) {
++        switch (index) {
+         case HPET_ID:
+             return;
+         case HPET_CFG:

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

@@ -1,35 +0,0 @@
-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");