Some read-only MSRs affect the behavior of ioctls such as
KVM_SET_NESTED_STATE. We can initialize them once and for all
right after the CPU is realized, since they will never be modified
by the guest.
Reported-by: Qingua Cheng <qcheng@redhat.com>
Cc: qemu-stable@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <1579544504-3616-2-git-send-email-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In ed9e923c3c ("x86: move SMM property to X86MachineState", 2019-12-17)
In v4.2.0-246-ged9e923c3c the SMM property was moved from PC
machine class to x86 machine class. Makes sense, but the change
was too aggressive: in target/i386/kvm.c:kvm_arch_init() it
altered check which sets SMRAM if given machine has SMM enabled.
The line that detects whether given machine object is class of
PC_MACHINE was removed from the check. This makes qemu try to
enable SMRAM for all machine types, which is not what we want.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Fixes: ed9e923c3c ("x86: move SMM property to X86MachineState", 2019-12-17)
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <7cc91bab3191bfd7e071bdd3fdf7fe2a2991deb0.1577692822.git.mprivozn@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Legacy PCI device assignment has been already removed in commit ab37bfc7d6
("pci-assign: Remove"), but some codes remain unused.
CC: qemu-trivial@nongnu.org
Signed-off-by: Eiichi Tsukata <devel@etsukata.com>
Message-Id: <20191209072932.313056-1-devel@etsukata.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
These are needed by microvm too, so move them outside of PC-specific files.
With this patch, microvm.c need not include pc.h anymore.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add it to microvm as well, it is a generic property of the x86
architecture.
Suggested-by: Sergio Lopez <slp@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The KVMState struct is opaque, so provide accessors for the fields
that will be moved from current_machine to the accelerator. For now
they just forward to the machine object, but this will change.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If kvm does not support VMX feature by nested=0, the kvm_vmx_basic
can't get the right value from MSR_IA32_VMX_BASIC register, which
make qemu coredump when qemu do KVM_SET_MSRS.
The coredump info:
error: failed to set MSR 0x480 to 0x0
kvm_put_msrs: Assertion `ret == cpu->kvm_msr_buf->nmsrs' failed.
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20191206071111.12128-1-yang.zhong@intel.com>
Reported-by: Catherine Ho <catherine.hecx@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The MSR_IA32_TSX_CTRL MSR can be used to hide TSX (also known as the
Trusty Side-channel Extension). By virtualizing the MSR, KVM guests
can disable TSX and avoid paying the price of mitigating TSX-based
attacks on microarchitectural side channels.
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
UMWAIT and TPAUSE instructions use 32bits IA32_UMWAIT_CONTROL at MSR
index E1H to determines the maximum time in TSC-quanta that the processor
can reside in either C0.1 or C0.2.
This patch is to Add support for save/load IA32_UMWAIT_CONTROL MSR in
guest.
Co-developed-by: Jingqi Liu <jingqi.liu@intel.com>
Signed-off-by: Jingqi Liu <jingqi.liu@intel.com>
Signed-off-by: Tao Xu <tao3.xu@intel.com>
Message-Id: <20191011074103.30393-3-tao3.xu@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
UMONITOR, UMWAIT and TPAUSE are a set of user wait instructions.
This patch adds support for user wait instructions in KVM. Availability
of the user wait instructions is indicated by the presence of the CPUID
feature flag WAITPKG CPUID.0x07.0x0:ECX[5]. User wait instructions may
be executed at any privilege level, and use IA32_UMWAIT_CONTROL MSR to
set the maximum time.
The patch enable the umonitor, umwait and tpause features in KVM.
Because umwait and tpause can put a (psysical) CPU into a power saving
state, by default we dont't expose it to kvm and enable it only when
guest CPUID has it. And use QEMU command-line "-overcommit cpu-pm=on"
(enable_cpu_pm is enabled), a VM can use UMONITOR, UMWAIT and TPAUSE
instructions. If the instruction causes a delay, the amount of time
delayed is called here the physical delay. The physical delay is first
computed by determining the virtual delay (the time to delay relative to
the VM’s timestamp counter). Otherwise, UMONITOR, UMWAIT and TPAUSE cause
an invalid-opcode exception(#UD).
The release document ref below link:
https://software.intel.com/sites/default/files/\
managed/39/c5/325462-sdm-vol-1-2abcd-3abcd.pdf
Co-developed-by: Jingqi Liu <jingqi.liu@intel.com>
Signed-off-by: Jingqi Liu <jingqi.liu@intel.com>
Signed-off-by: Tao Xu <tao3.xu@intel.com>
Message-Id: <20191011074103.30393-2-tao3.xu@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Hyper-V TLFS specifies this enlightenment as:
"NoNonArchitecturalCoreSharing - Indicates that a virtual processor will never
share a physical core with another virtual processor, except for virtual
processors that are reported as sibling SMT threads. This can be used as an
optimization to avoid the performance overhead of STIBP".
However, STIBP is not the only implication. It was found that Hyper-V on
KVM doesn't pass MD_CLEAR bit to its guests if it doesn't see
NoNonArchitecturalCoreSharing bit.
KVM reports NoNonArchitecturalCoreSharing in KVM_GET_SUPPORTED_HV_CPUID to
indicate that SMT on the host is impossible (not supported of forcefully
disabled).
Implement NoNonArchitecturalCoreSharing support in QEMU as tristate:
'off' - the feature is disabled (default)
'on' - the feature is enabled. This is only safe if vCPUS are properly
pinned and correct topology is exposed. As CPU pinning is done outside
of QEMU the enablement decision will be made on a higher level.
'auto' - copy KVM setting. As during live migration SMT settings on the
source and destination host may differ this requires us to add a migration
blocker.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20191018163908.10246-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Patch logs MCE AO, AR messages injected to guest or taken by QEMU itself.
We print the QEMU address for guest MCEs, helps on hypervisors that have
another source of MCE logging like mce log, and when they go missing.
For example we found these QEMU logs:
September 26th 2019, 17:36:02.309 Droplet-153258224: Guest MCE Memory Error at qemu addr 0x7f8ce14f5000 and guest 3d6f5000 addr of type BUS_MCEERR_AR injected qemu-system-x86_64 amsN ams3nodeNNNN
September 27th 2019, 06:25:03.234 Droplet-153258224: Guest MCE Memory Error at qemu addr 0x7f8ce14f5000 and guest 3d6f5000 addr of type BUS_MCEERR_AR injected qemu-system-x86_64 amsN ams3nodeNNNN
The first log had a corresponding mce log entry, the second didnt (logging
thresholds) we can infer from second entry same PA and mce type.
Signed-off-by: Mario Smarduch <msmarduch@digitalocean.com>
Message-Id: <20191009164459.8209-3-msmarduch@digitalocean.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM has a 80-entry limit at KVM_SET_CPUID2. With the
introduction of CPUID[0x1F], it is now possible to hit this limit
with unusual CPU configurations, e.g.:
$ ./x86_64-softmmu/qemu-system-x86_64 \
-smp 1,dies=2,maxcpus=2 \
-cpu EPYC,check=off,enforce=off \
-machine accel=kvm
qemu-system-x86_64: kvm_init_vcpu failed: Argument list too long
This happens because QEMU adds a lot of all-zeroes CPUID entries
for unused CPUID leaves. In the example above, we end up
creating 48 all-zeroes CPUID entries.
KVM already returns all-zeroes when emulating the CPUID
instruction if an entry is missing, so the all-zeroes entries are
redundant. Skip those entries. This reduces the CPUID table
size by half while keeping CPUID output unchanged.
Reported-by: Yumei Huang <yuhuang@redhat.com>
Fixes: https://bugzilla.redhat.com/show_bug.cgi?id=1741508
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20190822225210.32541-1-ehabkost@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
When I run QEMU with KVM under Valgrind, I currently get this warning:
Syscall param ioctl(generic) points to uninitialised byte(s)
at 0x95BA45B: ioctl (in /usr/lib64/libc-2.28.so)
by 0x429DC3: kvm_ioctl (kvm-all.c:2365)
by 0x51B249: kvm_arch_get_supported_msr_feature (kvm.c:469)
by 0x4C2A49: x86_cpu_get_supported_feature_word (cpu.c:3765)
by 0x4C4116: x86_cpu_expand_features (cpu.c:5065)
by 0x4C7F8D: x86_cpu_realizefn (cpu.c:5242)
by 0x5961F3: device_set_realized (qdev.c:835)
by 0x7038F6: property_set_bool (object.c:2080)
by 0x707EFE: object_property_set_qobject (qom-qobject.c:26)
by 0x705814: object_property_set_bool (object.c:1338)
by 0x498435: pc_new_cpu (pc.c:1549)
by 0x49C67D: pc_cpus_init (pc.c:1681)
Address 0x1ffeffee74 is on thread 1's stack
in frame #2, created by kvm_arch_get_supported_msr_feature (kvm.c:445)
It's harmless, but a little bit annoying, so silence it by properly
initializing the whole structure with zeroes.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Some secondary controls are automatically enabled/disabled based on the CPUID
values that are set for the guest. However, they are still available at a
global level and therefore should be present when KVM_GET_MSRS is sent to
/dev/kvm.
Unfortunately KVM forgot to include those, so fix that.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add code to convert the VMX feature words back into MSR values,
allowing the user to enable/disable VMX features as they wish. The same
infrastructure enables support for limiting VMX features in named
CPU models.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
VMX requires 64-bit feature words for the IA32_VMX_EPT_VPID_CAP
and IA32_VMX_BASIC MSRs. (The VMX control MSRs are 64-bit wide but
actually have only 32 bits of information).
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Suggested-by: Samuel Ortiz <sameo@linux.intel.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190818225414.22590-3-philmd@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Allow guest reads CORE cstate when exposing host CPU power management capabilities
to the guest. PKG cstate is restricted to avoid a guest to get the whole package
information in multi-tenant scenario.
Cc: Eduardo Habkost <ehabkost@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Radim Krčmář <rkrcmar@redhat.com>
Signed-off-by: Wanpeng Li <wanpengli@tencent.com>
Message-Id: <1563154124-18579-1-git-send-email-wanpengli@tencent.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Intel CooperLake cpu adds AVX512_BF16 instruction, defining as
CPUID.(EAX=7,ECX=1):EAX[bit 05].
The patch adds a property for setting the subleaf of CPUID leaf 7 in
case that people would like to specify it.
The release spec link as follows,
https://software.intel.com/sites/default/files/managed/c5/15/\
architecture-instruction-set-extensions-programming-reference.pdf
Signed-off-by: Jing Liu <jing2.liu@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Not the whole structure is initialized before passing it to the KVM.
Reduce the number of Valgrind reports.
Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Message-Id: <1564502498-805893-4-git-send-email-andrey.shinkevich@virtuozzo.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Function 'kvm_get_supported_msrs' is only called once
now, get rid of the static variable 'kvm_supported_msrs'.
Signed-off-by: Li Qiang <liq3ea@163.com>
Message-Id: <20190725151639.21693-1-liq3ea@163.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add support for halt poll control MSR: save/restore, migration
and new feature name.
The purpose of this MSR is to allow the guest to disable
host halt poll.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Message-Id: <20190603230408.GA7938@amt.cnet>
[Do not enable by default, as pointed out by Mark Kanda. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
sysemu/sysemu.h is a rather unfocused dumping ground for stuff related
to the system-emulator. Evidence:
* It's included widely: in my "build everything" tree, changing
sysemu/sysemu.h still triggers a recompile of some 1100 out of 6600
objects (not counting tests and objects that don't depend on
qemu/osdep.h, down from 5400 due to the previous two commits).
* It pulls in more than a dozen additional headers.
Split stuff related to run state management into its own header
sysemu/runstate.h.
Touching sysemu/sysemu.h now recompiles some 850 objects. qemu/uuid.h
also drops from 1100 to 850, and qapi/qapi-types-run-state.h from 4400
to 4200. Touching new sysemu/runstate.h recompiles some 500 objects.
Since I'm touching MAINTAINERS to add sysemu/runstate.h anyway, also
add qemu/main-loop.h.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190812052359.30071-30-armbru@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
[Unbreak OS-X build]
In my "build everything" tree, changing qemu/main-loop.h triggers a
recompile of some 5600 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h). It includes block/aio.h,
which in turn includes qemu/event_notifier.h, qemu/notify.h,
qemu/processor.h, qemu/qsp.h, qemu/queue.h, qemu/thread-posix.h,
qemu/thread.h, qemu/timer.h, and a few more.
Include qemu/main-loop.h only where it's needed. Touching it now
recompiles only some 1700 objects. For block/aio.h and
qemu/event_notifier.h, these numbers drop from 5600 to 2800. For the
others, they shrink only slightly.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190812052359.30071-21-armbru@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
In my "build everything" tree, changing sysemu/reset.h triggers a
recompile of some 2600 out of 6600 objects (not counting tests and
objects that don't depend on qemu/osdep.h).
The main culprit is hw/hw.h, which supposedly includes it for
convenience.
Include sysemu/reset.h only where it's needed. Touching it now
recompiles less than 200 objects.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190812052359.30071-9-armbru@redhat.com>
Writing the nested state e.g. after a vmport access can invalidate
important parts of the kernel-internal state, and it is not needed as
well. So leave this out from KVM_PUT_RUNTIME_STATE.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Message-Id: <bdd53f40-4e60-f3ae-7ec6-162198214953@siemens.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Do not allocate env->nested_state unless we later need to migrate the
nested virtualization state.
With this change, nested_state_needed() will return false if the
VMX flag is not included in the virtual machine. KVM_GET/SET_NESTED_STATE
is also disabled for SVM which is safer (we know that at least the NPT
root and paging mode have to be saved/loaded), and thus the corresponding
subsection can go away as well.
Inspired by a patch from Liran Alon.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Previous to this change, a vCPU exposed with VMX running on a kernel
without KVM_CAP_NESTED_STATE or KVM_CAP_EXCEPTION_PAYLOAD resulted in
adding a migration blocker. This was because when the code was written
it was thought there is no way to reliably know if a vCPU is utilising
VMX or not at runtime. However, it turns out that this can be known to
some extent:
In order for a vCPU to enter VMX operation it must have CR4.VMXE set.
Since it was set, CR4.VMXE must remain set as long as the vCPU is in
VMX operation. This is because CR4.VMXE is one of the bits set
in MSR_IA32_VMX_CR4_FIXED1.
There is one exception to the above statement when vCPU enters SMM mode.
When a vCPU enters SMM mode, it temporarily exits VMX operation and
may also reset CR4.VMXE during execution in SMM mode.
When the vCPU exits SMM mode, vCPU state is restored to be in VMX operation
and CR4.VMXE is restored to its original state of being set.
Therefore, when the vCPU is not in SMM mode, we can infer whether
VMX is being used by examining CR4.VMXE. Otherwise, we cannot
know for certain but assume the worse that vCPU may utilise VMX.
Summaring all the above, a vCPU may have enabled VMX in case
CR4.VMXE is set or vCPU is in SMM mode.
Therefore, remove migration blocker and check before migration
(cpu_pre_save()) if the vCPU may have enabled VMX. If true, only then
require relevant kernel capabilities.
While at it, demand KVM_CAP_EXCEPTION_PAYLOAD only when the vCPU is in
guest-mode and there is a pending/injected exception. Otherwise, this
kernel capability is not required for proper migration.
Reviewed-by: Joao Martins <joao.m.martins@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Reviewed-by: Maran Wilson <maran.wilson@oracle.com>
Tested-by: Maran Wilson <maran.wilson@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.22 (GNU/Linux)
iQEcBAABAgAGBQJdH7FgAAoJEL/70l94x66DdUUH+gLr/ZdjLIdfYy9cjcnevf4E
cJlxdaW9KvUsK2uVgqQ/3b1yF+GCGk10n6n8ZTIbClhs+6NpqEMz5O3FA/Na6FGA
48M2DwJaJ2H9AG/lQBlSBNUZfLsEJ9rWy7DHvNut5XMJFuWGwdtF/jRhUm3KqaRq
vAaOgcQHbzHU9W8r1NJ7l6pnPebeO7S0JQV+82T/ITTz2gEBDUkJ36boO6fedkVQ
jLb9nZyG3CJXHm2WlxGO4hkqbLFzURnCi6imOh2rMdD8BCu1eIVl59tD1lC/A0xv
Pp3xXnv9SgJXsV4/I/N3/nU85ZhGVMPQZXkxaajHPtJJ0rQq7FAG8PJMEj9yPe8=
=poke
-----END PGP SIGNATURE-----
Merge remote-tracking branch 'remotes/bonzini/tags/for-upstream' into staging
Bugfixes.
# gpg: Signature made Fri 05 Jul 2019 21:21:52 BST
# gpg: using RSA key BFFBD25F78C7AE83
# gpg: Good signature from "Paolo Bonzini <bonzini@gnu.org>" [full]
# gpg: aka "Paolo Bonzini <pbonzini@redhat.com>" [full]
# Primary key fingerprint: 46F5 9FBD 57D6 12E7 BFD4 E2F7 7E15 100C CD36 69B1
# Subkey fingerprint: F133 3857 4B66 2389 866C 7682 BFFB D25F 78C7 AE83
* remotes/bonzini/tags/for-upstream:
ioapic: use irq number instead of vector in ioapic_eoi_broadcast
hw/i386: Fix linker error when ISAPC is disabled
Makefile: generate header file with the list of devices enabled
target/i386: kvm: Fix when nested state is needed for migration
minikconf: do not include variables from MINIKCONF_ARGS in config-all-devices.mak
target/i386: fix feature check in hyperv-stub.c
ioapic: clear irq_eoi when updating the ioapic redirect table entry
intel_iommu: Fix unexpected unmaps during global unmap
intel_iommu: Fix incorrect "end" for vtd_address_space_unmap
i386/kvm: Fix build with -m32
checkpatch: do not warn for multiline parenthesized returned value
pc: fix possible NULL pointer dereference in pc_machine_get_device_memory_region_size()
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
find_next_bit() takes a pointer of type "const unsigned long *", but the
first argument passed here is a "uint64_t *". These types are
incompatible when compiling qemu with -m32.
Just use ctz64() instead.
Fixes: c686193072
Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20190624193913.28343-1-mreitz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The CPUID.1F as Intel V2 Extended Topology Enumeration Leaf would be
exposed if guests want to emulate multiple software-visible die within
each package. Per Intel's SDM, the 0x1f is a superset of 0xb, thus they
can be generated by almost same code as 0xb except die_offset setting.
If the number of dies per package is greater than 1, the cpuid_min_level
would be adjusted to 0x1f regardless of whether the host supports CPUID.1F.
Likewise, the CPUID.1F wouldn't be exposed if env->nr_dies < 2.
Suggested-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Like Xu <like.xu@linux.intel.com>
Message-Id: <20190620054525.37188-2-like.xu@linux.intel.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Previous commits have added support for migration of nested virtualization
workloads. This was done by utilising two new KVM capabilities:
KVM_CAP_NESTED_STATE and KVM_CAP_EXCEPTION_PAYLOAD. Both which are
required in order to correctly migrate such workloads.
Therefore, change code to add a migration blocker for vCPUs exposed with
Intel VMX or AMD SVM in case one of these kernel capabilities is
missing.
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Reviewed-by: Maran Wilson <maran.wilson@oracle.com>
Message-Id: <20190619162140.133674-11-liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Kernel commit c4f55198c7c2 ("kvm: x86: Introduce KVM_CAP_EXCEPTION_PAYLOAD")
introduced a new KVM capability which allows userspace to correctly
distinguish between pending and injected exceptions.
This distinguish is important in case of nested virtualization scenarios
because a L2 pending exception can still be intercepted by the L1 hypervisor
while a L2 injected exception cannot.
Furthermore, when an exception is attempted to be injected by QEMU,
QEMU should specify the exception payload (CR2 in case of #PF or
DR6 in case of #DB) instead of having the payload already delivered in
the respective vCPU register. Because in case exception is injected to
L2 guest and is intercepted by L1 hypervisor, then payload needs to be
reported to L1 intercept (VMExit handler) while still preserving
respective vCPU register unchanged.
This commit adds support for QEMU to properly utilise this new KVM
capability (KVM_CAP_EXCEPTION_PAYLOAD).
Reviewed-by: Nikita Leshenko <nikita.leshchenko@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Message-Id: <20190619162140.133674-10-liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Kernel commit 8fcc4b5923af ("kvm: nVMX: Introduce KVM_CAP_NESTED_STATE")
introduced new IOCTLs to extract and restore vCPU state related to
Intel VMX & AMD SVM.
Utilize these IOCTLs to add support for migration of VMs which are
running nested hypervisors.
Reviewed-by: Nikita Leshenko <nikita.leshchenko@oracle.com>
Reviewed-by: Maran Wilson <maran.wilson@oracle.com>
Tested-by: Maran Wilson <maran.wilson@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Message-Id: <20190619162140.133674-9-liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit d98f26073b ("target/i386: kvm: add VMX migration blocker")
added a migration blocker for vCPU exposed with Intel VMX.
However, migration should also be blocked for vCPU exposed with
AMD SVM.
Both cases should be blocked because QEMU should extract additional
vCPU state from KVM that should be migrated as part of vCPU VMState.
E.g. Whether vCPU is running in guest-mode or host-mode.
Fixes: d98f26073b ("target/i386: kvm: add VMX migration blocker")
Reviewed-by: Maran Wilson <maran.wilson@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Message-Id: <20190619162140.133674-6-liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If userspace (QEMU) debug guest, when #DB is raised in guest and
intercepted by KVM, KVM forwards information on #DB to userspace
instead of injecting #DB to guest.
While doing so, KVM don't update vCPU DR6 but instead report the #DB DR6
value to userspace for further handling.
See KVM's handle_exception() DB_VECTOR handler.
QEMU handler for this case is kvm_handle_debug(). This handler basically
checks if #DB is related to one of user set hardware breakpoints and if
not, it re-inject #DB into guest.
The re-injection is done by setting env->exception_injected to #DB which
will later be passed as events.exception.nr to KVM_SET_VCPU_EVENTS ioctl
by kvm_put_vcpu_events().
However, in case userspace re-injects #DB, KVM expects userspace to set
vCPU DR6 as reported to userspace when #DB was intercepted! Otherwise,
KVM_REQ_EVENT handler will inject #DB with wrong DR6 to guest.
Fix this issue by updating vCPU DR6 appropriately when re-inject #DB to
guest.
Reviewed-by: Nikita Leshenko <nikita.leshchenko@oracle.com>
Reviewed-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Message-Id: <20190619162140.133674-5-liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Simiar to how kvm_init_vcpu() calls kvm_arch_init_vcpu() to perform
arch-dependent initialisation, introduce kvm_arch_destroy_vcpu()
to be called from kvm_destroy_vcpu() to perform arch-dependent
destruction.
This was added because some architectures (Such as i386)
currently do not free memory that it have allocated in
kvm_arch_init_vcpu().
Suggested-by: Maran Wilson <maran.wilson@oracle.com>
Reviewed-by: Maran Wilson <maran.wilson@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Message-Id: <20190619162140.133674-3-liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Commit d98f26073b ("target/i386: kvm: add VMX migration blocker")
added migration blocker for vCPU exposed with Intel VMX because QEMU
doesn't yet contain code to support migration of nested virtualization
workloads.
However, that commit missed adding deletion of the migration blocker in
case init of vCPU failed. Similar to invtsc_mig_blocker. This commit fix
that issue.
Fixes: d98f26073b ("target/i386: kvm: add VMX migration blocker")
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Reviewed-by: Maran Wilson <maran.wilson@oracle.com>
Message-Id: <20190619162140.133674-2-liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
MSR IA32_CORE_CAPABILITY is a feature-enumerating MSR, which only
enumerates the feature split lock detection (via bit 5) by now.
The existence of MSR IA32_CORE_CAPABILITY is enumerated by CPUID.7_0:EDX[30].
The latest kernel patches about them can be found here:
https://lkml.org/lkml/2019/4/24/1909
Signed-off-by: Xiaoyao Li <xiaoyao.li@linux.intel.com>
Message-Id: <20190617153654.916-1-xiaoyao.li@linux.intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Hyper-V on KVM can only use Synthetic timers with Direct Mode (opting for
an interrupt instead of VMBus message). This new capability is only
announced in KVM_GET_SUPPORTED_HV_CPUID.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20190517141924.19024-10-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Enlightened VMCS is enabled by writing to a field in VP assist page and
these require virtual APIC.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20190517141924.19024-9-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Synthetic timers operate in hv-time time and Windows won't use these
without SynIC.
Add .dependencies field to kvm_hyperv_properties[] and a generic mechanism
to check dependencies between features.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20190517141924.19024-7-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In many case we just want to give Windows guests all currently supported
Hyper-V enlightenments and that's where this new mode may come handy. We
pass through what was returned by KVM_GET_SUPPORTED_HV_CPUID.
hv_cpuid_check_and_set() is modified to also set cpu->hyperv_* flags as
we may want to check them later (and we actually do for hv_runtime,
hv_synic,...).
'hv-passthrough' is a development only feature, a migration blocker is
added to prevent issues while migrating between hosts with different
feature sets.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20190517141924.19024-6-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Let's consolidate Hyper-V features handling in hyperv_handle_properties().
The change is necessary to support 'hv-passthrough' mode as we'll be just
copying CPUIDs from KVM instead of filling them in.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20190517141924.19024-4-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM now supports reporting supported Hyper-V features through CPUID
(KVM_GET_SUPPORTED_HV_CPUID ioctl). Going forward, this is going to be
the only way to announce new functionality and this has already happened
with Direct Mode stimers.
While we could just support KVM_GET_SUPPORTED_HV_CPUID for new features,
it seems to be beneficial to use it for all Hyper-V enlightenments when
possible. This way we can implement 'hv-all' pass-through mode giving the
guest all supported Hyper-V features even when QEMU knows nothing about
them.
Implementation-wise we create a new kvm_hyperv_properties structure
defining Hyper-V features, get_supported_hv_cpuid()/
get_supported_hv_cpuid_legacy() returning the supported CPUID set and
a bit over-engineered hv_cpuid_check_and_set() which we will also be
used to set cpu->hyperv_* properties for 'hv-all' mode.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20190517141924.19024-3-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Representing Hyper-V properties as bits will allow us to check features
and dependencies between them in a natural way.
Suggested-by: Roman Kagan <rkagan@virtuozzo.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20190517141924.19024-2-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
