The Microsoft "Requirements for Implementing the Microsoft Hypervisor
Interface" document marks exactly two privileges in CPUID leaf
0x40000003 EAX as "Must be set": AccessHypercallMsrs (bit 5) and
AccessVpIndex (bit 6). Cloud Hypervisor advertised neither.
Without bit 5, Windows guests abort enlightened-mode initialization
before timer-API selection: HalpHvTimerApi is left NULL and every
QueryPerformanceCounter call falls back to reading
HV_X64_MSR_TIME_REF_COUNT (0x40000020), costing one VM exit per call.
The guest never writes HV_X64_MSR_REFERENCE_TSC (0x40000021) to enable
the reference TSC page, even though AccessPartitionReferenceTsc (bit 9)
is advertised.
With both bits set, Windows 10 22H2 and Windows 11 25H2 guests enable
the reference TSC page at boot. Measured QueryPerformanceCounter
throughput on a nested-KVM host went from ~71K calls/sec (14 us/call,
one MSR exit each) to ~1.3M calls/sec (free, no exits); on bare metal
from ~390K to ~1.9M calls/sec. Guest idle CPU and interrupt-service
time drop correspondingly.
Both MSR ranges are already handled in-kernel by KVM unconditionally,
so no backend change is needed. Bisection across the full delta to
QEMU's Hyper-V CPUID layout (vendor ID, max leaf, leaves 4-6 contents,
build number) shows bit 5 is the only load-bearing change; bit 6 is
included per the conformance document's mandate.
Signed-off-by: Tonic Li <tonic@simular.ai>
Signed-off-by: tonic <tonicbupt@gmail.com>
Import the std modules used in the crate instead of spelling the full
paths at every use site.
Signed-off-by: Henry Hrvoje Tonkovac <htonkovac@gmail.com>
Assisted-by: Claude:Opus-4.8
Cover the prior commit by constructing a Request directly and a stub
AsyncIo whose backend methods are unreachable, then submit a payload
with sector + num_sectors past u64::MAX and assert BadRequest.
Assisted-by: Claude:Opus-4.7
Signed-off-by: Anatol Belski <anbelski@linux.microsoft.com>
In Request::execute_async the WriteZeroes arm multiplied wz_sector
by SECTOR_SIZE before the checked_add of sector and num_sectors.
A wz_sector near u64::MAX overflows the multiplication.
Reorder the arm to run the checked_add and disk_nsectors check
first, matching the Discard arm above.
Signed-off-by: Anatol Belski <anbelski@linux.microsoft.com>
Import the std modules used in the crate instead of spelling the full
paths at every use site.
Signed-off-by: Henry Hrvoje Tonkovac <htonkovac@gmail.com>
Assisted-by: Claude:Opus-4.8
Import the std modules used in the disk-format handlers instead of
spelling the full paths at every use site.
Signed-off-by: Henry Hrvoje Tonkovac <htonkovac@gmail.com>
Assisted-by: Claude:Opus-4.8
During pause the block backend's async I/O path can have unfinished I/O
requests. A snapshot or migration RAM copy taken after pause returns can
then race with kernel writes and capture torn pages.
Since vCPUs are already paused, the VMM thread can stop new block
submissions and wait for the worker to drain before parking the worker
threads.
Assisted-by: Codex:GPT-5
Signed-off-by: Dylan Reid <dgreid@fb.com>
Remove stale #[allow]s whose lints no longer fire, convert the
unconditionally-firing ones to #[expect], and keep the conditional
ones as #[allow] (e.g. large_enum_variant only fires when both kvm
and mshv are enabled; a nonminimal_bool only on x86). The many
unreachable_patterns allows are feature-gated and left as #[allow].
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
Remove stale #[allow]s whose lints no longer fire (a module-level
non_camel_case_types in mpspec, and a too_many_arguments on a riscv64
configure_system that no longer exceeds the argument threshold), and
convert the still-needed ones to #[expect].
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
Convert the still-needed #[allow]s to #[expect] so they warn if the
lints stop firing.
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
Since igvm is a required feature of sev_snp and also sev_snp is x86-64
only the cfg attributes at build time can be consolidated & simplified.
Signed-off-by: Rob Bradford <rbradford@meta.com>
Validate that the PCI segment specified is a valid PCI segment ID (less
than the number of segments specified) defaulting to default if no
segments are specified because no there is no platform configuration.
See: #8376
Signed-off-by: Rob Bradford <rbradford@meta.com>
This test has a copy and paste error where the PCI segment ID was being
set with no extra segments configured.
Signed-off-by: Rob Bradford <rbradford@meta.com>
This test has a copy and paste error where the PCI segment ID was being
set with no extra segments configured.
Signed-off-by: Rob Bradford <rbradford@meta.com>
Validate that all files that are necessary for TLS encryption are in the
given folder. The knowledge which files are necessary is part of the TLS
module.
On-behalf-of: SAP sebastian.eydam@sap.com
Signed-off-by: Sebastian Eydam <sebastian.eydam@cyberus-technology.de>
As we now have more than one parameter for the receive migration call,
this commit also adds parsing and validation for those parameters. We
maintain backwards compatibility by also correctly parsing the case
where the caller only provides a URL.
On-behalf-of: SAP sebastian.eydam@sap.com
Signed-off-by: Sebastian Eydam <sebastian.eydam@cyberus-technology.de>
For TLS we have to parse the hostname from the given migration URL. For
that we have to make a few assumptions about the URL (e.g. it always has
a port). To catch problems early, we tighten the URL validation.
On-behalf-of: SAP sebastian.eydam@sap.com
Signed-off-by: Sebastian Eydam <sebastian.eydam@cyberus-technology.de>
Extend ReceiveListener with a TLS-backed listener variant for migration
receivers.
Store the TCP listener together with the server TLS configuration, wrap
accepted sockets in TlsStream::new_server(), and preserver the existing
listener cloning and fd polling behavior so receive-side migration code
can treat TLS listeners like the existing TCP and UNIX cases.
On-behalf-of: SAP sebastian.eydam@sap.com
Signed-off-by: Sebastian Eydam <sebastian.eydam@cyberus-technology.de>
Teach the migration transport to handle TLS-backed streams alongside
plain TCP and UNIX sockets.
Introduce a Tls variant in SocketStream and implement the necessary
traits.
Also updates the local-migration error path to reject any non-UNIX
transport, which now includes TLS-wrapped TCP connections.
On-behalf-of: SAP sebastian.eydam@sap.com
Signed-off-by: Sebastian Eydam <sebastian.eydam@cyberus-technology.de>
TLS connections have a TLS server (listens for incoming connections) and
a TLS client (initiates the connection). This commit adds the code for
the client side, which is the sender of a migration
On-behalf-of: SAP sebastian.eydam@sap.com
Signed-off-by: Sebastian Eydam <sebastian.eydam@cyberus-technology.de>
ReadVolatile already provides a default read_volatile_exact()
implementation, and WriteVolatile a default write_volatile_exact()
implementation. Overriding these functions adds no behavioral value, but
duplicates logic and needs to be updated whenever SocketStream gains or
changes a variant.
On-behalf-of: SAP sebastian.eydam@sap.com
Signed-off-by: Sebastian Eydam <sebastian.eydam@cyberus-technology.de>
When the host half-closed the socket this was wrongly interpreted as a
full shutdown preventing the guest from sending any more data. Instead
propagate the half-close by setting just `VSOCK_FLAGS_SHUTDOWN_SEND`,
leaving the connection alive so that guest-to-host writes are still
forwarded. The connection is only torn down once the guest also shuts
down its send side or a host write fails.
Fixes: #8300
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Rob Bradford <rbradford@meta.com>
When the guest did a half-close (shutting down only its send side) the
connection state was updated but the write half of the host Unix socket
was never closed so the host peer never saw an EOF. This caused issues
with newer systemd (v256+) as it now half closes its socket and waits
for the host side to react and fully close the connection.
Propagate the guest's half-close to the host by shutting down the write
half of the backing stream. This is deferred until any buffered guest
data has been flushed so that no data is lost, and the connection is
left open so that host-to-guest data keeps flowing.
Assisted-by: Claude:claude-opus-4-8
Signed-off-by: Rob Bradford <rbradford@meta.com>
Split `add_pci_device()` into two phases: `allocate_pci_bars()` which
only allocates BAR address space, and `commit_pci_device()` which
makes the device visible to the guest on the PCI bus.
All callers now follow the pattern: allocate BARs → perform device-
specific setup (ioeventfd, device_tree, mmio mapping) → commit device.
This eliminates a race window where the guest could discover a
partially-initialized device via `acpiphp_check_bridge()` during rapid
sequential hotplug, causing BAR reprogramming to fail because
ioeventfds and device_tree entries were not yet in place.
Signed-off-by: wuxinyue <wuxinyue.wxy@antgroup.com>
Assisted-by: Claude:Opus-4.6
`PciBus::register_mapping()` operates on `mmio_bus` and `io_bus`
which are passed in as external parameters and have nothing to do
with PciBus internal state. Move this logic into
`DeviceManager::register_bar_mapping()` where it belongs, and move
the `PioInsert`/`MmioInsert` error variants from `PciRootError` to
`DeviceManagerError` accordingly.
Signed-off-by: wuxinyue <wuxinyue.wxy@antgroup.com>
Assisted-by: Claude:Opus-4.6
Remove stale #[allow]s whose lints no longer fire, convert the
unconditionally-firing ones to #[expect], and keep the
arch/feature-conditional ones as #[allow]. Verified across kvm/mshv,
x86_64/aarch64, and --all-features.
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
Convert the still-needed #[allow]s to #[expect] so they warn if the
lints stop firing.
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
Drop stale #[allow]s whose lints no longer fire and convert the
still-needed ones to #[expect] so they warn if the lints stop firing.
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
Remove a stale #[allow(non_camel_case_types)] whose lint no longer
fires; the enum variants are already CamelCase.
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
Currently ejecting a device leaks its mapping keeping the container fd
open. Remove the mapping so the fd can be closed.
Assisted-by: Claude:Opus-4.8
Signed-off-by: Dylan Reid <dgreid@fb.com>
Document how live migration protocol versions are handled, including
the supported current/previous version window and the need to migrate
through an intermediate Cloud Hypervisor version for larger version
gaps.
On-behalf-of: SAP leander.kohler@sap.com
Signed-off-by: Leander Kohler <leander.kohler@cyberus-technology.de>
Print the supported vm-migration protocol version range in
cloud-hypervisor --version as an extra line:
vm-migration protocol versions v0-v1
This makes the currently supported compatibility window
visible without having to inspect the migration code.
On-behalf-of: SAP leander.kohler@sap.com
Signed-off-by: Leander Kohler <leander.kohler@cyberus-technology.de>
The cloud-hypervisor binary prints vm-migration protocol constants
directly, so it needs its own dependency on the vm-migration crate
instead of relying on vmm's transitive dependency.
On-behalf-of: SAP leander.kohler@sap.com
Signed-off-by: Leander Kohler <leander.kohler@cyberus-technology.de>
Validate the sender's migration protocol version when
handling the initial Start request.
Read the version from the Start command header, accept only
the supported version window n-1..=n, and reject unsupported
versions with Error. A rejected Start moves the receiver to
the aborted state.
This keeps compatibility one-way, from older protocol
versions to newer ones, and leaves later version-based
branching on the receiver side.
Log the protocol version on both sender and receiver to make
the active migration path visible.
On-behalf-of: SAP leander.kohler@sap.com
Signed-off-by: Leander Kohler <leander.kohler@cyberus-technology.de>
Add protocol-side support for migration protocol versioning.
Use the existing 6-byte Start command header, which was
previously zero padding, to carry the sender's migration
protocol version without changing the wire layout.
Store the version as a little-endian u16 in the first two
bytes and ignore the remaining four bytes. A zeroed command
header continues to mean a legacy v0 sender.
This keeps the message flow unchanged for rollout:
Start is still followed by plain OK or Error (Aborted), and no new
command is needed.
On-behalf-of: SAP leander.kohler@sap.com
Signed-off-by: Leander Kohler <leander.kohler@cyberus-technology.de>
Import the modules used in the crate instead of spelling the
fully-qualified paths at every use site, and collapse Result<T,
io::Error> into io::Result<T>. This covers the feature-gated modules
(fw_cfg, ivshmem, pvmemcontrol) as well, leaving the whole crate free
of clippy::absolute_paths warnings.
Signed-off-by: Henry Hrvoje Tonkovac <htonkovac@gmail.com>
Assisted-by: Claude:Opus-4.8
Drop stale #[allow]s whose lints no longer fire and convert the
rest to #[expect], which warns if they ever stop being needed.
Part of #8326.
Signed-off-by: Tushar Khatri <hello@tusharkhatri.in>
During migration send one final set of changed memory after capturing
the snapshot/state. This captures any memory changed as a side effect of
capturing that state. In particular with vhost-user capturing the device
state can lead to inflight requests being drained/flushed which could
change memory. As this is related to the snapshot account for this
memory transfer in the snapshot metrics.
No equivalent change is needed for snapshot as the memory is written
after the state is snapshotted.
Signed-off-by: Rob Bradford <rbradford@meta.com>
Query and save the dirty log before shutting down the vhost-user device.
This allows any final dirty memory ranges to be recorded before it
becomes impossible to do that as the vhost-user handle has been closed.
This is required to ensure that all memory writes have been correctly
recorded that may be triggered by inflight I/O drains from vhost-user
device state capture.
One small implementation wrinkle: with local migrations there is no
dirty logging (since we just pass the memory FD over the socket) so
calling dirty_log() would generate an error. As there is no clean way to
query if dirty logging has been started add a boolean to track if its
active.
Signed-off-by: Rob Bradford <rbradford@meta.com>
Rather than each device implementation holding a reference to the guest
memory move this to VhostUserCommon. This refactoring simplifies the
function signatures but also allows for methods that act on
VhostUserCommon that don't have the memory available to them.
Signed-off-by: Rob Bradford <rbradford@meta.com>
Currently both kvm and mshv require an IGVM file to boot a SEV-SNP VM.
This is already configured in the top-level cloud-hypervisor
Cargo.toml where sev_snp depends on igvm.
Add a similar dependency in the vmm crate which helps simplify some of
the in-code cfg blocks by removing the ones that are within a sev_snp
cfg block.
Signed-off-by: Ruben Hakobyan <hruben@meta.com>
Import the modules used in the crate instead of spelling the
fully-qualified paths at every use site. This covers std paths along
with a few crate-internal and external-crate paths, leaving pci free of
clippy::absolute_paths warnings.
Signed-off-by: Henry Hrvoje Tonkovac <htonkovac@gmail.com>
Assisted-by: Claude:Opus-4.8
Re-enable the fw_cfg integration tests for MSHV now that port string I/O
is handled by the hypervisor backend.
Signed-off-by: Wei Liu <liuwe@microsoft.com>
Assisted-by: Copilot:GPT-5.5
