[PATCH rfcv4 00/13] LIBVIRT: X86: TDX support
by Zhenzhong Duan
Hi,
This series brings libvirt the x86 TDX support.
* What's TDX?
TDX stands for Trust Domain Extensions which isolates VMs from
the virtual-machine manager (VMM)/hypervisor and any other software on
the platform.
To support TDX, multiple software components, not only KVM but also QEMU,
guest Linux and virtual bios, need to be updated. For more details, please
check link[1].
This patchset is another software component to extend libvirt to support TDX,
with which one can start a TDX guest from high level rather than running qemu
directly.
* Misc
As QEMU use a software emulated way to reset guest which isn't supported by TDX
guest for security reason. We simulate reboot for TDX guest by kill and create a
new one in FakeReboot framework.
Complete code can be found at [2], matching qemu code can be found at [3].
There is a 'debug' property for tdx-guest object which isn't in matching qemu[3]
yet. I keep them intentionally as they will be implemented in qemu as extention
series of [3].
* Test
start/stop/reboot with virsh
stop/reboot trigger in guest
stop with on_poweroff=destroy/restart
reboot with on_reboot=destroy/restart
* Patch organization
- patch 1-4: Support query of TDX capabilities.
- patch 5-8: Add TDX type to launchsecurity framework.
- patch 9-11: Add reboot support to TDX guest
- patch 12-13: Add test and docs
TODO:
- update QEMU capabilities data in tests, depending on qemu TDX merged beforehand
- add reconnect logic in virsh command
[1] https://lore.kernel.org/kvm/cover.1708933498.git.isaku.yamahata@intel.com
[2] https://github.com/intel/libvirt-tdx/commits/tdx_for_upstream_rfcv4
[3] https://github.com/intel/qemu-tdx/tree/tdx-qemu-upstream-v5
Thanks
Zhenzhong
Changelog:
rfcv4:
- add a check to tools/virt-host-validate-qemu.c (Daniel)
- remove check of q35 (Daniel)
- model 'SocktetAddress' QAPI in xml schema (Daniel)
- s/Quote-Generation-Service/quoteGenerationService/ (Daniel)
- define bits in tdx->policy and add validating logic (Daniel)
- presume QEMU choose split kernel irqchip for TDX guest by default (Daniel)
- utilize existing FakeReboot framework to do reboot for TDX guest (Daniel)
- drop patch11 'conf: Add support to keep same domid for hard reboot' (Daniel)
- add test in tests/ to validate parsing and formatting logic (Daniel)
- add doc in docs/formatdomain.rst (Daniel)
- add R-B
rfcv3:
- Change to generate qemu cmdline with -bios
- drop firmware auto match as -bios is used
- add a hard reboot method to reboot TDX guest
rfcv3: https://www.mail-archive.com/devel@lists.libvirt.org/msg00385.html
rfcv2:
- give up using qmp cmd and check TDX directly on host for TDX capabilities.
- use launchsecurity framework to support TDX
- use <os>.<loader> for general loader
- add auto firmware match feature for TDX
A example TDVF fimware description file 70-edk2-x86_64-tdx.json:
{
"description": "UEFI firmware for x86_64, supporting Intel TDX",
"interface-types": [
"uefi"
],
"mapping": {
"device": "generic",
"filename": "/usr/share/OVMF/OVMF_CODE-tdx.fd"
},
"targets": [
{
"architecture": "x86_64",
"machines": [
"pc-q35-*"
]
}
],
"features": [
"intel-tdx",
"verbose-dynamic"
],
"tags": [
]
}
rfcv2: https://www.mail-archive.com/libvir-list@redhat.com/msg219378.html
Zhenzhong Duan (13):
tools: Secure guest check for Intel in virt-host-validate
qemu: Check if INTEL Trust Domain Extention support is enabled
qemu: Add TDX capability
conf: expose TDX feature in domain capabilities
conf: add tdx as launch security type
qemu: Add command line and validation for TDX type
qemu: force special parameters enabled for TDX guest
Add Intel TDX Quote Generation Service(QGS) support
qemu: add FakeReboot support for TDX guest
qemu: Support reboot command in guest
qemu: Avoid duplicate FakeReboot for secure guest
Add test cases for Intel TDX
docs: domain: Add documentation for Intel TDX guest
docs/formatdomain.rst | 68 ++++
docs/formatdomaincaps.rst | 1 +
src/conf/domain_capabilities.c | 1 +
src/conf/domain_capabilities.h | 1 +
src/conf/domain_conf.c | 312 ++++++++++++++++++
src/conf/domain_conf.h | 75 +++++
src/conf/schemas/domaincaps.rng | 9 +
src/conf/schemas/domaincommon.rng | 135 ++++++++
src/conf/virconftypes.h | 2 +
src/qemu/qemu_capabilities.c | 36 +-
src/qemu/qemu_capabilities.h | 1 +
src/qemu/qemu_command.c | 139 ++++++++
src/qemu/qemu_firmware.c | 1 +
src/qemu/qemu_monitor.c | 28 +-
src/qemu/qemu_monitor.h | 2 +-
src/qemu/qemu_monitor_json.c | 6 +-
src/qemu/qemu_namespace.c | 1 +
src/qemu/qemu_process.c | 75 +++++
src/qemu/qemu_validate.c | 44 +++
...unch-security-tdx-qgs-fd.x86_64-latest.xml | 77 +++++
.../launch-security-tdx-qgs-fd.xml | 30 ++
...ch-security-tdx-qgs-inet.x86_64-latest.xml | 77 +++++
.../launch-security-tdx-qgs-inet.xml | 30 ++
...ch-security-tdx-qgs-unix.x86_64-latest.xml | 77 +++++
.../launch-security-tdx-qgs-unix.xml | 30 ++
...h-security-tdx-qgs-vsock.x86_64-latest.xml | 77 +++++
.../launch-security-tdx-qgs-vsock.xml | 30 ++
tests/qemuxmlconftest.c | 24 ++
tools/virt-host-validate-common.c | 22 +-
tools/virt-host-validate-common.h | 1 +
30 files changed, 1407 insertions(+), 5 deletions(-)
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-fd.x86_64-latest.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-fd.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-inet.x86_64-latest.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-inet.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-unix.x86_64-latest.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-unix.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-vsock.x86_64-latest.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-tdx-qgs-vsock.xml
--
2.34.1
3 weeks, 5 days
[PATCH 0/4] hw/s390x: Alias @dump-skeys -> @dump-s390-skey and deprecate
by Philippe Mathieu-Daudé
We are trying to unify all qemu-system-FOO to a single binary.
In order to do that we need to remove QAPI target specific code.
@dump-skeys is only available on qemu-system-s390x. This series
rename it as @dump-s390-skey, making it available on other
binaries. We take care of backward compatibility via deprecation.
Philippe Mathieu-Daudé (4):
hw/s390x: Introduce the @dump-s390-skeys QMP command
hw/s390x: Introduce the 'dump_s390_skeys' HMP command
hw/s390x: Deprecate the HMP 'dump_skeys' command
hw/s390x: Deprecate the QMP @dump-skeys command
docs/about/deprecated.rst | 5 +++++
qapi/misc-target.json | 5 +++++
qapi/misc.json | 18 ++++++++++++++++++
include/monitor/hmp.h | 1 +
hw/s390x/s390-skeys-stub.c | 24 ++++++++++++++++++++++++
hw/s390x/s390-skeys.c | 19 +++++++++++++++++--
hmp-commands.hx | 17 +++++++++++++++--
hw/s390x/meson.build | 5 +++++
8 files changed, 90 insertions(+), 4 deletions(-)
create mode 100644 hw/s390x/s390-skeys-stub.c
--
2.41.0
1 month, 2 weeks
[libvirt] [PATCH] Fix python error reporting for some storage operations
by Cole Robinson
In the python bindings, all vir* classes expect to be
passed a virConnect object when instantiated. Before
the storage stuff, these classes were only instantiated
in virConnect methods, so the generator is hardcoded to
pass 'self' as the connection instance to these classes.
Problem is there are some methods that return pool or vol
instances which aren't called from virConnect: you can
lookup a storage volume's associated pool, and can lookup
volumes from a pool. In these cases passing 'self' doesn't
give the vir* instance a connection, so when it comes time
to raise an exception crap hits the fan.
Rather than rework the generator to accomodate this edge
case, I just fixed the init functions for virStorage* to
pull the associated connection out of the passed value
if it's not a virConnect instance.
Thanks,
Cole
diff --git a/python/generator.py b/python/generator.py
index 01a17da..c706b19 100755
--- a/python/generator.py
+++ b/python/generator.py
@@ -962,8 +962,12 @@ def buildWrappers():
list = reference_keepers[classname]
for ref in list:
classes.write(" self.%s = None\n" % ref[1])
- if classname in [ "virDomain", "virNetwork", "virStoragePool", "virStorageVol" ]:
+ if classname in [ "virDomain", "virNetwork" ]:
classes.write(" self._conn = conn\n")
+ elif classname in [ "virStorageVol", "virStoragePool" ]:
+ classes.write(" self._conn = conn\n" + \
+ " if not isinstance(conn, virConnect):\n" + \
+ " self._conn = conn._conn\n")
classes.write(" if _obj != None:self._o = _obj;return\n")
classes.write(" self._o = None\n\n");
destruct=None
1 month, 3 weeks
[RFC 0/4] meson: Enable -Wundef
by Andrea Bolognani
A few days ago I have posted a patch[1] that addresses an issue
introduced when a meson check was dropped but some uses of the
corresponding WITH_ macro were not removed at the same time.
That got me thinking about what we can do to prevent such scenarios
from happening again in the future. I have come up with something
that I think would be effective, but since applying the approach
throughout the entire codebase would require a non-trivial amount of
work, I figured I'd ask for feedback before embarking on it.
The idea is that there are two types of macros we can use for
conditional compilation: external ones, coming from the OS or other
libraries, and internal ones, which are the result of meson tests.
The external ones (e.g. SIOCSIFFLAGS, __APPLE__) are usually only
defined if they apply, so it is correct to check for their presence
with #ifdef. Using #if will also work, as undefined macros evaluate
to zero, but it's not good practice to use them that way. If -Wundef
has been passed to the compiler, those incorrect uses will be
reported (only on platforms where they are not defined, of course).
The internal ones (e.g. WITH_QEMU, WITH_STRUCT_IFREQ) are similar,
but in this case we control their definition. This means that using
means that the feature is not available on the machine we're building
on, but it could also mean that we've removed the meson check and
forgot to update all users of the macro. In this case, -Wundef would
work 100% reliably to detect the issue: if the meson check doesn't
exist, neither will the macro, regardless of what platform we're
building on.
So the approach I'm suggesting is to use a syntax-check rule to
ensure that internal macros are only ever checked with #if instead of
Of course this requires a full sweep to fix all cases in which we're
not already doing things according to the proposal. Should be fairly
easy, if annoying. A couple of examples are included here for
demonstration purposes.
The bigger impact is going to be on the build system. Right now we
generally only define WITH_ macros if the check passed, but that will
have to change and the result is going to be quite a bit of
additional meson code I'm afraid.
Thoughts?
[1] https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/message/S...
Andrea Bolognani (4):
configmake: Check for WIN32 correctly
meson: Always define WITH_*_DECL macros
syntax-check: Ensure WITH_ macros are used correctly
meson: Enable -Wundef
build-aux/syntax-check.mk | 5 +++++
configmake.h.in | 2 +-
meson.build | 3 +++
tests/virmockstathelpers.c | 28 ++++++++++++++--------------
4 files changed, 23 insertions(+), 15 deletions(-)
--
2.43.2
1 month, 4 weeks
Re: [PATCH] Сheck snapshot disk is not NULL when searching it in the VM config
by Peter Krempa
On Mon, May 20, 2024 at 14:48:47 +0000, Efim Shevrin via Devel wrote:
> Hello,
>
> > If vmdisk is NULL, shouldn't this function (qemuSnapshotDeleteValidate()) return an error?
>
> I think this qemuSnapshotDeleteValidate should not return an error.
>
> It seems to me that when vmdisk is NULL, this does not invalidate
> the snapshot itself, but indicates that the config has changed since
> the snapshot was done. And if the VM config has changed, this adds evidence that the snapshot should be deleted,
> because the snapshot does not reflect the real vm config.
>
> Since we do not have an analogue of the --force option for deleting a snapshot, in the case when qemuSnapshotDeleteValidate returns
> an error when vmdisk is NULL, we will never delete a snapshot which has invalid disk.
Snapshot deletion does have something that can be considered force and
that is the '--metadata' option that removes just the snapshot
definition (metadata) and doesn't touch the disk images.
> > Similarly, disk can be NULL too
> Thank you for the comment regarding the disk variable. I`ve reworked patch.
>
> When creating a snapshot of a VM with multiple hard disks,
> the snapshot takes into account the presence of all disks
> in the system. If, over time, one of the disks is deleted,
> the snapshot will continue to store knowledge of the deleted disk.
> This results in the fact that at the moment of deleting the snapshot,
> at the validation stage, a disk from the snapshot will be searched which
> is not in the VM configuration. As a result, vmdisk variable will
> be equal to NULL. Dereferencing a null pointer at the time of calling
> virStorageSourceIsSameLocation(vmdisk->src, disk->src)
> will result in SIGSEGV.
Crashing is obviously not okay ...
> Also, the disk variable can also be equal to NULL and this
> requires to check that disk != NULL before calling the
> virStorageSourceIsSameLocation function to avoid SIGSEGV.
.. but going ahead with the snapshot deletion isn't always okay either.
The disk isn't referenced by the VM so the disk state can't be merged,
while the state would be merged for any other disk.
When reverting back to a previous snapshot, which is still referencing
the older state of the disk which was removed from the VM, the VM would
see that the image state of disks that were present at deletion would
contain the merged state, but only a partial state for the disk which
was later removed.
2 months, 4 weeks
[libvirt PATCH] qemu_snapshot: allow reverting to external disk only snapshot
by Pavel Hrdina
When snapshot is created with disk-only flag it is always external
snapshot without memory state. Historically when there was not support
to revert external snapshots this produced error message.
error: Failed to revert snapshot s1
error: internal error: Invalid target domain state 'disk-snapshot'. Refusing snapshot reversion
Now we can simply consider this as reverting to offline snapshot as the
possible damage to file system is already done at the point of snapshot
creation.
Resolves: https://issues.redhat.com/browse/RHEL-21549
Signed-off-by: Pavel Hrdina <phrdina(a)redhat.com>
---
src/qemu/qemu_snapshot.c | 3 +--
1 file changed, 1 insertion(+), 2 deletions(-)
diff --git a/src/qemu/qemu_snapshot.c b/src/qemu/qemu_snapshot.c
index 0cac0c4146..7964f70553 100644
--- a/src/qemu/qemu_snapshot.c
+++ b/src/qemu/qemu_snapshot.c
@@ -2606,6 +2606,7 @@ qemuSnapshotRevert(virDomainObj *vm,
case VIR_DOMAIN_SNAPSHOT_SHUTDOWN:
case VIR_DOMAIN_SNAPSHOT_SHUTOFF:
case VIR_DOMAIN_SNAPSHOT_CRASHED:
+ case VIR_DOMAIN_SNAPSHOT_DISK_SNAPSHOT:
ret = qemuSnapshotRevertInactive(vm, snapshot, snap,
driver, cfg,
&inactiveConfig,
@@ -2617,8 +2618,6 @@ qemuSnapshotRevert(virDomainObj *vm,
_("qemu doesn't support reversion of snapshot taken in PMSUSPENDED state"));
goto endjob;
- case VIR_DOMAIN_SNAPSHOT_DISK_SNAPSHOT:
- /* Rejected earlier as an external snapshot */
case VIR_DOMAIN_SNAPSHOT_NOSTATE:
case VIR_DOMAIN_SNAPSHOT_BLOCKED:
case VIR_DOMAIN_SNAPSHOT_LAST:
--
2.43.0
2 months, 4 weeks
[PATCH 0/3] conf,qemu: add AIA support for RISC-V 'virt'
by Daniel Henrique Barboza
Hi,
This series adds official support for RISC-V AIA (Advanced Interrupt
Architecture). AIA and has been supported by the 'virt' RISC-V board, as
a machine property, since QEMU 7.0.
Daniel Henrique Barboza (3):
qemu: add capability for RISC-V AIA feature
conf,qemu: implement RISC-V 'aia' virt domain feature
qemu: add RISC-V 'aia' command line
docs/formatdomain.rst | 8 ++++
src/conf/domain_conf.c | 39 +++++++++++++++++++
src/conf/domain_conf.h | 11 ++++++
src/conf/schemas/domaincommon.rng | 15 +++++++
src/libvirt_private.syms | 2 +
src/qemu/qemu_capabilities.c | 2 +
src/qemu/qemu_capabilities.h | 1 +
src/qemu/qemu_command.c | 5 +++
src/qemu/qemu_validate.c | 15 +++++++
.../caps_8.0.0_riscv64.xml | 1 +
.../caps_9.1.0_riscv64.xml | 1 +
...cv64-virt-features-aia.riscv64-latest.args | 31 +++++++++++++++
...scv64-virt-features-aia.riscv64-latest.xml | 1 +
.../riscv64-virt-features-aia.xml | 27 +++++++++++++
tests/qemuxmlconftest.c | 2 +
15 files changed, 161 insertions(+)
create mode 100644 tests/qemuxmlconfdata/riscv64-virt-features-aia.riscv64-latest.args
create mode 120000 tests/qemuxmlconfdata/riscv64-virt-features-aia.riscv64-latest.xml
create mode 100644 tests/qemuxmlconfdata/riscv64-virt-features-aia.xml
--
2.45.2
3 months
[PATCH 00/12] Introduce SEV-SNP support
by Michal Privoznik
SEV-SNP support just landed in QEMU. Here is the first round of patches
to incorporate support into libvirt.
TODOs (aka problems of future me):
- Teach tools/virt-qemu-sev-validate how to deal with SEV-SNP
- Try to find a SEV-SNP machine a test these patches in real worl
- Write a kbase article on attestation with SEV-SNP
Michal Prívozník (12):
qemu_monitor_json: Report error in error paths in SEV related code
conf: Move some members of virDomainSEVDef into virDomainSEVCommonDef
conf: Separate SEV formatting into a function
Drop needless typecast to virDomainLaunchSecurity
src: Convert some _virDomainSecDef::sectype checks to switch()
qemu_monitor: Allow querying SEV-SNP state in 'query-sev'
qemu: Report snp-policy in virDomainGetLaunchSecurityInfo()
qemu_capabilities: Introduce QEMU_CAPS_SEV_SNP_GUEST
conf: Introduce SEV-SNP support
qemu: Build cmd line for SEV-SNP
qemu: Allow setting launch security for SEV-SNP
qemu_firmware: Pick the right firmware for SEV-SNP guests
docs/formatdomain.rst | 108 ++++++++++++
include/libvirt/libvirt-domain.h | 10 ++
src/conf/domain_conf.c | 156 ++++++++++++++----
src/conf/domain_conf.h | 28 +++-
src/conf/domain_validate.c | 44 +++++
src/conf/schemas/domaincommon.rng | 73 ++++++--
src/conf/virconftypes.h | 4 +
src/qemu/qemu_capabilities.c | 4 +
src/qemu/qemu_capabilities.h | 3 +
src/qemu/qemu_cgroup.c | 19 ++-
src/qemu/qemu_command.c | 56 ++++++-
src/qemu/qemu_driver.c | 60 +++++--
src/qemu/qemu_firmware.c | 20 ++-
src/qemu/qemu_monitor.c | 7 +-
src/qemu/qemu_monitor.h | 41 ++++-
src/qemu/qemu_monitor_json.c | 67 ++++++--
src/qemu/qemu_monitor_json.h | 8 +-
src/qemu/qemu_namespace.c | 3 +-
src/qemu/qemu_process.c | 34 ++--
src/qemu/qemu_validate.c | 13 +-
src/security/security_dac.c | 34 +++-
.../caps_9.1.0_x86_64.xml | 1 +
.../firmware/60-edk2-ovmf-x64-amdsev.json | 1 +
tests/qemumonitorjsontest.c | 65 +++++++-
...launch-security-sev-snp.x86_64-latest.args | 35 ++++
.../launch-security-sev-snp.x86_64-latest.xml | 1 +
.../launch-security-sev-snp.xml | 47 ++++++
tests/qemuxmlconftest.c | 2 +
28 files changed, 817 insertions(+), 127 deletions(-)
create mode 100644 tests/qemuxmlconfdata/launch-security-sev-snp.x86_64-latest.args
create mode 120000 tests/qemuxmlconfdata/launch-security-sev-snp.x86_64-latest.xml
create mode 100644 tests/qemuxmlconfdata/launch-security-sev-snp.xml
--
2.44.2
3 months
[PATCH 00/20] qemu: support mapped-ram+directio+mulitfd
by Jim Fehlig
This series is essentially V1 of a prior RFC [1] to support QEMU's
mapped-ram stream format [2] and migration capability. Along with
supporting mapped-ram, it implements a design approach we discussed
for supporting parallel save/restore [3]. In summary, the approach is
1. Add mapped-ram migration capability
2. Steal an element from save header 'unused' for a 'features' variable
and bump save version to 3.
3. Add /etc/libvirt/qemu.conf knob for the save format version,
defaulting to latest v3
4. Use v3 (aka mapped-ram) by default
5. Use mapped-ram with BYPASS_CACHE for v3, old approach for v2
6. include: Define constants for parallel save/restore
7. qemu: Add support for parallel save. Implies mapped-ram, reject if v2
8. qemu: Add support for parallel restore. Implies mapped-ram.
Reject if v2
9. tools: add parallel parameter to virsh save command
10. tools: add parallel parameter to virsh restore command
With this series, saving and restoring using mapped-ram is enabled by
default if the underlying QEMU advertises the mapped-ram migration
capability. It can be disabled by changing the 'save_image_version'
setting in qemu.conf.
To use mapped-ram with QEMU:
- The 'mapped-ram' migration capability must be set to true
- The 'multifd' migration capability must be set to true and
the 'multifd-channels' migration parameter must set to a
value >= 1
- QEMU must be provided an fdset containing the migration fd(s)
- The 'migrate' qmp command is invoked with a URI referencing the fdset
and an offset where to start reading or writing the data stream, e.g.
{"execute":"migrate",
"arguments":{"detach":true,"resume":false,
"uri":"file:/dev/fdset/0,offset=0x11921"}}
The mapped-ram stream, in conjunction with direct IO and multifd, can
significantly improve the time required to save VM memory state. The
following tables compare mapped-ram with the existing, sequential save
stream. In all cases, the save and restore operations are to/from a
block device comprised of two NVMe disks in RAID0 configuration with
xfs (~8600MiB/s). The values in the 'save time' and 'restore time'
columns were scraped from the 'real' time reported by time(1). The
'Size' and 'Blocks' columns were provided by the corresponding
outputs of stat(1).
VM: 32G RAM, 1 vcpu, idle (shortly after boot)
| save | restore |
| time | time | Size | Blocks
-----------------------+---------+---------+--------------+--------
legacy | 6.193s | 4.399s | 985744812 | 1925288
-----------------------+---------+---------+--------------+--------
mapped-ram | 5.109s | 1.176s | 34368554354 | 1774472
-----------------------+---------+---------+--------------+--------
legacy + direct IO | 5.725s | 4.512s | 985765251 | 1925328
-----------------------+---------+---------+--------------+--------
mapped-ram + direct IO | 4.627s | 1.490s | 34368554354 | 1774304
-----------------------+---------+---------+--------------+--------
mapped-ram + direct IO | | | |
+ multifd-channels=8 | 4.421s | 0.845s | 34368554318 | 1774312
-------------------------------------------------------------------
VM: 32G RAM, 30G dirty, 1 vcpu in tight loop dirtying memory
| save | restore |
| time | time | Size | Blocks
-----------------------+---------+---------+--------------+---------
legacy | 25.800s | 14.332s | 33154309983 | 64754512
-----------------------+---------+---------+--------------+---------
mapped-ram | 18.742s | 15.027s | 34368559228 | 64617160
-----------------------+---------+---------+--------------+---------
legacy + direct IO | 13.115s | 18.050s | 33154310496 | 64754520
-----------------------+---------+---------+--------------+---------
mapped-ram + direct IO | 13.623s | 15.959s | 34368557392 | 64662040
-----------------------+-------- +---------+--------------+---------
mapped-ram + direct IO | | | |
+ multifd-channels=8 | 6.994s | 6.470s | 34368554980 | 64665776
--------------------------------------------------------------------
As can be seen from the tables, one caveat of mapped-ram is the logical file
size of a saved image is basically equivalent to the VM memory size. Note
however that mapped-ram typically uses fewer blocks on disk.
Support for mapped-ram+direct-io only recently landed in upstream QEMU
and will first appear in the 9.1 release, which may complicate merging
support in libvirt. Specifically, I'm not sure how to detect if the
combination is supported by QEMU. Suggestions welcomed.
Similar to the RFC, V1 ignores compression. libvirt currently supports
compression by connecting the output of QEMU's save stream to the specified
compression program via a pipe. This approach is incompatible with mapped-ram
since the fd provided to QEMU must be seekable. In general, we can consider
mapped-ram and compression incompatible and document they cannot be used
together.
[1] https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/message/E...
[2] https://gitlab.com/qemu-project/qemu/-/blob/master/docs/devel/migration/m...
[3] https://lists.libvirt.org/archives/list/devel@lists.libvirt.org/message/K...
Claudio Fontana (2):
include: Define constants for parallel save/restore
tools: add parallel parameter to virsh restore command
Jim Fehlig (17):
lib: virDomainSaveParams: Ensure absolute save path
qemu_fd: Add function to retrieve fdset ID
qemu: Add function to check capability in migration params
qemu: Add function to get bool value from migration params
qemu: Add mapped-ram migration capability
qemu: Add function to get migration params for save
qemu: QEMU_SAVE_VERSION: Bump to version 3
qemu: conf: Add setting for save image version
qemu: Add helper function for creating save image fd
qemu: Add support for mapped-ram on save
qemu: Decompose qemuSaveImageOpen
qemu: Move creation of qemuProcessIncomingDef struct
qemu: Apply migration parameters in qemuMigrationDstRun
qemu: Add support for mapped-ram on restore
qemu: Support O_DIRECT with mapped-ram on save
qemu: Support O_DIRECT with mapped-ram on restore
qemu: Add support for parallel save and restore
Li Zhang (1):
tools: add parallel parameter to virsh save command
docs/manpages/virsh.rst | 9 +-
include/libvirt/libvirt-domain.h | 13 ++
src/libvirt-domain.c | 52 +++++--
src/qemu/libvirtd_qemu.aug | 1 +
src/qemu/qemu.conf.in | 6 +
src/qemu/qemu_conf.c | 16 +++
src/qemu/qemu_conf.h | 5 +
src/qemu/qemu_driver.c | 104 +++++++++-----
src/qemu/qemu_fd.c | 18 +++
src/qemu/qemu_fd.h | 3 +
src/qemu/qemu_migration.c | 192 +++++++++++++++++--------
src/qemu/qemu_migration.h | 9 +-
src/qemu/qemu_migration_params.c | 86 ++++++++++++
src/qemu/qemu_migration_params.h | 17 +++
src/qemu/qemu_monitor.c | 39 ++++++
src/qemu/qemu_monitor.h | 5 +
src/qemu/qemu_process.c | 120 +++++++++++-----
src/qemu/qemu_process.h | 19 ++-
src/qemu/qemu_saveimage.c | 216 ++++++++++++++++++++---------
src/qemu/qemu_saveimage.h | 35 +++--
src/qemu/qemu_snapshot.c | 26 ++--
src/qemu/test_libvirtd_qemu.aug.in | 1 +
tools/virsh-domain.c | 79 +++++++++--
23 files changed, 827 insertions(+), 244 deletions(-)
--
2.35.3
3 months, 1 week
[PATCH v3 0/5] ch: handle events from cloud-hypervisor
by Purna Pavan Chandra Aekkaladevi
changes from v2->v3:
* Remove patch 'utils: Implement virFileIsNamedPipe' as it is no more needed.
* Remove the eventmonitorpath only if it exists
* Added domain name as a prefix to logs from ch_events.c. This will make
debugging easier.
* Simplified event parsing logic by reserving a byte for null char.
changes from v1->v2:
* Rebase on latest master
* Use /* */ for comments
* Remove fifo file if already exists
* Address other comments from Praveen Paladugu
cloud-hypervisor raises various events, including VM lifecylce operations
such as boot, shutdown, pause, resume, etc. Libvirt will now read these
events and take the necessary actions, such as correctly updating the
domain state. A FIFO file is passed to `--event-monitor` option of
cloud-hypervisor. Libvirt creates a new thread that acts as the reader
of the fifo file and continuously monitors for new events. Currently,
shutdown events are handled by updating the domain state appropriately.
Purna Pavan Chandra Aekkaladevi (5):
ch: pass --event-monitor option to cloud-hypervisor
ch: start a new thread for handling ch events
ch: events: Read and parse cloud-hypervisor events
ch: events: facilitate lifecycle events handling
NEWS: Mention event handling support in ch driver
NEWS.rst | 7 +
po/POTFILES | 1 +
src/ch/ch_events.c | 329 ++++++++++++++++++++++++++++++++++++++++++++
src/ch/ch_events.h | 54 ++++++++
src/ch/ch_monitor.c | 52 ++++++-
src/ch/ch_monitor.h | 11 ++
src/ch/meson.build | 2 +
7 files changed, 449 insertions(+), 7 deletions(-)
create mode 100644 src/ch/ch_events.c
create mode 100644 src/ch/ch_events.h
--
2.34.1
4 months, 3 weeks