Hyper-V TLFS allows for L0 and L1 hypervisors to collaborate on L2's
TLB flush hypercalls handling. With the correct setup, L2's TLB flush
hypercalls can be handled by L0 directly, without the need to exit to
L1.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220525115949.1294004-6-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM kind of supported "extended GVA ranges" (up to 4095 additional GFNs
per hypercall) since the implementation of Hyper-V PV TLB flush feature
(Linux-4.18) as regardless of the request, full TLB flush was always
performed. "Extended GVA ranges for TLB flush hypercalls" feature bit
wasn't exposed then. Now, as KVM gains support for fine-grained TLB
flush handling, exposing this feature starts making sense.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220525115949.1294004-5-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Hyper-V specification allows to pass parameters for certain hypercalls
using XMM registers ("XMM Fast Hypercall Input"). When the feature is
in use, it allows for faster hypercalls processing as KVM can avoid
reading guest's memory.
KVM supports the feature since v5.14.
Rename HV_HYPERCALL_{PARAMS_XMM_AVAILABLE -> XMM_INPUT_AVAILABLE} to
comply with KVM.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220525115949.1294004-4-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The newly introduced enlightenment allow L0 (KVM) and L1 (Hyper-V)
hypervisors to collaborate to avoid unnecessary updates to L2
MSR-Bitmap upon vmexits.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220525115949.1294004-3-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Previously, HV_CPUID_NESTED_FEATURES.EAX CPUID leaf was handled differently
as it was only used to encode the supported eVMCS version range. In fact,
there are also feature (e.g. Enlightened MSR-Bitmap) bits there. In
preparation to adding these features, move HV_CPUID_NESTED_FEATURES leaf
handling to hv_build_cpuid_leaf() and drop now-unneeded 'hyperv_nested'.
No functional change intended.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20220525115949.1294004-2-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Live migration can happen when Arch LBR LBREn bit is cleared,
e.g., when migration happens after guest entered SMM mode.
In this case, we still need to migrate Arch LBR MSRs.
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220517155024.33270-1-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In the first generation of Arch LBR, the max support
Arch LBR depth is 32, both host and guest use the value
to set depth MSR. This can simplify the implementation
of patch given the side-effect of mismatch of host/guest
depth MSR: XRSTORS will reset all recording MSRs to 0s
if the saved depth mismatches MSR_ARCH_LBR_DEPTH.
In most of the cases Arch LBR is not in active status,
so check the control bit before save/restore the big
chunck of Arch LBR MSRs.
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-7-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When try to get one msr from KVM, I found there's no such kind of
existing interface while kvm_put_one_msr() is there. So here comes
the patch. It'll remove redundant preparation code before finally
call KVM_GET_MSRS IOCTL.
No functional change intended.
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Message-Id: <20220215195258.29149-4-weijiang.yang@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Jon Doron <arilou@gmail.com>
Reviewed-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Message-Id: <20220216102500.692781-5-arilou@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SynDbg commands can come from two different flows:
1. Hypercalls, in this mode the data being sent is fully
encapsulated network packets.
2. SynDbg specific MSRs, in this mode only the data that needs to be
transfered is passed.
Signed-off-by: Jon Doron <arilou@gmail.com>
Reviewed-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Message-Id: <20220216102500.692781-4-arilou@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In the physical machine environment, when a SRAR error occurs,
the IA32_MCG_STATUS RIPV bit is set, but qemu does not set this
bit. When qemu injects an SRAR error into virtual machine, the
virtual machine kernel just call do_machine_check() to kill the
current task, but not call memory_failure() to isolate the faulty
page, which will cause the faulty page to be allocated and used
repeatedly. If used by the virtual machine kernel, it will cause
the virtual machine to crash
Signed-off-by: luofei <luofei@unicloud.com>
Message-Id: <20220120084634.131450-1-luofei@unicloud.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Fix vCPU hot-unplug related leak reported by Valgrind:
==132362== 4,096 bytes in 1 blocks are definitely lost in loss record 8,440 of 8,549
==132362== at 0x4C3B15F: memalign (vg_replace_malloc.c:1265)
==132362== by 0x4C3B288: posix_memalign (vg_replace_malloc.c:1429)
==132362== by 0xB41195: qemu_try_memalign (memalign.c:53)
==132362== by 0xB41204: qemu_memalign (memalign.c:73)
==132362== by 0x7131CB: kvm_init_xsave (kvm.c:1601)
==132362== by 0x7148ED: kvm_arch_init_vcpu (kvm.c:2031)
==132362== by 0x91D224: kvm_init_vcpu (kvm-all.c:516)
==132362== by 0x9242C9: kvm_vcpu_thread_fn (kvm-accel-ops.c:40)
==132362== by 0xB2EB26: qemu_thread_start (qemu-thread-posix.c:556)
==132362== by 0x7EB2159: start_thread (in /usr/lib64/libpthread-2.28.so)
==132362== by 0x9D45DD2: clone (in /usr/lib64/libc-2.28.so)
Reported-by: Mark Kanda <mark.kanda@oracle.com>
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Tested-by: Mark Kanda <mark.kanda@oracle.com>
Message-Id: <20220322120522.26200-1-philippe.mathieu.daude@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
KVM support for AMX includes a new system attribute, KVM_X86_XCOMP_GUEST_SUPP.
Commit 19db68ca68 ("x86: Grant AMX permission for guest", 2022-03-15) however
did not fully consider the behavior on older kernels. First, it warns
too aggressively. Second, it invokes the KVM_GET_DEVICE_ATTR ioctl
unconditionally and then uses the "bitmask" variable, which remains
uninitialized if the ioctl fails. Third, kvm_ioctl returns -errno rather
than -1 on errors.
While at it, explain why the ioctl is needed and KVM_GET_SUPPORTED_CPUID
is not enough.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
XFD(eXtended Feature Disable) allows to enable a
feature on xsave state while preventing specific
user threads from using the feature.
Support save and restore XFD MSRs if CPUID.D.1.EAX[4]
enumerate to be valid. Likewise migrate the MSRs and
related xsave state necessarily.
Signed-off-by: Zeng Guang <guang.zeng@intel.com>
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220217060434.52460-8-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When dynamic xfeatures (e.g. AMX) are used by the guest, the xsave
area would be larger than 4KB. KVM_GET_XSAVE2 and KVM_SET_XSAVE
under KVM_CAP_XSAVE2 works with a xsave buffer larger than 4KB.
Always use the new ioctls under KVM_CAP_XSAVE2 when KVM supports it.
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Zeng Guang <guang.zeng@intel.com>
Signed-off-by: Wei Wang <wei.w.wang@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220217060434.52460-7-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Add AMX primary feature bits XFD and AMX_TILE to
enumerate the CPU's AMX capability. Meanwhile, add
AMX TILE and TMUL CPUID leaf and subleaves which
exist when AMX TILE is present to provide the maximum
capability of TILE and TMUL.
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20220217060434.52460-6-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Kernel allocates 4K xstate buffer by default. For XSAVE features
which require large state component (e.g. AMX), Linux kernel
dynamically expands the xstate buffer only after the process has
acquired the necessary permissions. Those are called dynamically-
enabled XSAVE features (or dynamic xfeatures).
There are separate permissions for native tasks and guests.
Qemu should request the guest permissions for dynamic xfeatures
which will be exposed to the guest. This only needs to be done
once before the first vcpu is created.
KVM implemented one new ARCH_GET_XCOMP_SUPP system attribute API to
get host side supported_xcr0 and Qemu can decide if it can request
dynamically enabled XSAVE features permission.
https://lore.kernel.org/all/20220126152210.3044876-1-pbonzini@redhat.com/
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Signed-off-by: Jing Liu <jing2.liu@intel.com>
Message-Id: <20220217060434.52460-4-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We invoke the kvm_irqchip_commit_routes() for each addition to MSI route
table, which is not efficient if we are adding lots of routes in some cases.
This patch lets callers invoke the kvm_irqchip_commit_routes(), so the
callers can decide how to optimize.
[1] https://lists.gnu.org/archive/html/qemu-devel/2021-11/msg00967.html
Signed-off-by: Longpeng <longpeng2@huawei.com>
Message-Id: <20220222141116.2091-3-longpeng2@huawei.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Move the various memalign-related functions out of osdep.h and into
their own header, which we include only where they are used.
While we're doing this, add some brief documentation comments.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20220226180723.1706285-10-peter.maydell@linaro.org
This is unnecessary, because the interrupt would be retrieved and queued
anyway by KVM_GET_VCPU_EVENTS and KVM_SET_VCPU_EVENTS respectively,
and it makes the flow more similar to the one for KVM_GET/SET_SREGS2.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This allows to make PDPTRs part of the migration
stream and thus not reload them after migration which
is against X86 spec.
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Message-Id: <20211101132300.192584-2-mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
QEMU coding style mandates to not use Linux kernel internal
types for scalars types. Replace __u32 by uint32_t.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20211116193955.2793171-1-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
"sysemu/sev.h" is only used from x86-specific files. Let's move it
to include/hw/i386, and merge it with target/i386/sev.h.
Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20211007161716.453984-16-philmd@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SEV is a x86 specific feature, and the "sev_i386.h" header
is already in target/i386/. Rename it as "sev.h" to simplify.
Patch created mechanically using:
$ git mv target/i386/sev_i386.h target/i386/sev.h
$ sed -i s/sev_i386.h/sev.h/ $(git grep -l sev_i386.h)
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20211007161716.453984-15-philmd@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Currently, we hardcode Hyper-V version id (CPUID 0x40000002) to
WS2008R2 and it is known that certain tools in Windows check this. It
seems useful to provide some flexibility by making it possible to change
this info at will. CPUID information is defined in TLFS as:
EAX: Build Number
EBX Bits 31-16: Major Version
Bits 15-0: Minor Version
ECX Service Pack
EDX Bits 31-24: Service Branch
Bits 23-0: Service Number
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210902093530.345756-8-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The enlightenment allows to use Hyper-V SynIC with hardware APICv/AVIC
enabled. Normally, Hyper-V SynIC disables these hardware features and
suggests the guest to use paravirtualized AutoEOI feature. Linux-4.15
gains support for conditional APICv/AVIC disablement, the feature
stays on until the guest tries to use AutoEOI feature with SynIC. With
'HV_DEPRECATING_AEOI_RECOMMENDED' bit exposed, modern enough Windows/
Hyper-V versions should follow the recommendation and not use the
(unwanted) feature.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210902093530.345756-7-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In preparation to enabling Hyper-V + APICv/AVIC move
HV_APIC_ACCESS_RECOMMENDED setting out of kvm_hyperv_properties[]: the
'real' feature bit for the vAPIC features is HV_APIC_ACCESS_AVAILABLE,
HV_APIC_ACCESS_RECOMMENDED is a recommendation to use the feature which
we may not always want to give.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210902093530.345756-6-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
By default, KVM allows the guest to use all currently supported Hyper-V
enlightenments when Hyper-V CPUID interface was exposed, regardless of if
some features were not announced in guest visible CPUIDs. hv-enforce-cpuid
feature alters this behavior and only allows the guest to use exposed
Hyper-V enlightenments. The feature is supported by Linux >= 5.14 and is
not enabled by default in QEMU.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210902093530.345756-5-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
By default, KVM allows the guest to use all currently supported PV features
even when they were not announced in guest visible CPUIDs. Introduce a new
"kvm-pv-enforce-cpuid" flag to limit the supported feature set to the
exposed features. The feature is supported by Linux >= 5.10 and is not
enabled by default in QEMU.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210902093530.345756-4-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Provide a name field for all the memory listeners. It can be used to identify
which memory listener is which.
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20210817013553.30584-2-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The SGX sub-leafs are enumerated at CPUID 0x12. Indices 0 and 1 are
always present when SGX is supported, and enumerate SGX features and
capabilities. Indices >=2 are directly correlated with the platform's
EPC sections. Because the number of EPC sections is dynamic and user
defined, the number of SGX sub-leafs is "NULL" terminated.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-15-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
If the guest want to fully use SGX, the guest needs to be able to
access provisioning key. Add a new KVM_CAP_SGX_ATTRIBUTE to KVM to
support provisioning key to KVM guests.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-14-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
SGX adds multiple flags to FEATURE_CONTROL to enable SGX and Flexible
Launch Control.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-12-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
On real hardware, on systems that supports SGX Launch Control, those
MSRs are initialized to digest of Intel's signing key; on systems that
don't support SGX Launch Control, those MSRs are not available but
hardware always uses digest of Intel's signing key in EINIT.
KVM advertises SGX LC via CPUID if and only if the MSRs are writable.
Unconditionally initialize those MSRs to digest of Intel's signing key
when CPU is realized and reset to reflect the fact. This avoids
potential bug in case kvm_arch_put_registers() is called before
kvm_arch_get_registers() is called, in which case guest's virtual
SGX_LEPUBKEYHASH MSRs will be set to 0, although KVM initializes those
to digest of Intel's signing key by default, since KVM allows those MSRs
to be updated by Qemu to support live migration.
Save/restore the SGX Launch Enclave Public Key Hash MSRs if SGX Launch
Control (LC) is exposed to the guest. Likewise, migrate the MSRs if they
are writable by the guest.
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Kai Huang <kai.huang@intel.com>
Signed-off-by: Yang Zhong <yang.zhong@intel.com>
Message-Id: <20210719112136.57018-11-yang.zhong@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Most callers check the return value. Some check whether it set an
error. Functionally equivalent, but the former tends to be easier on
the eyes, so do that everywhere.
Prior art: commit c6ecec43b2 "qemu-option: Check return value instead
of @err where convenient".
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20210720125408.387910-10-armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
invtsc_mig_blocker has static storage duration. When a CPU with
certain features is initialized, and invtsc_mig_blocker is still null,
we add a migration blocker and store it in invtsc_mig_blocker.
The object is freed when migrate_add_blocker() fails, leaving
invtsc_mig_blocker dangling. It is not freed on later failures.
Same for hv_passthrough_mig_blocker and hv_no_nonarch_cs_mig_blocker.
All failures are actually fatal, so whether we free or not doesn't
really matter, except as bad examples to be copied / imitated.
Clean this up in a minimal way: never free these blocker objects.
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Marcelo Tosatti <mtosatti@redhat.com>
Cc: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20210720125408.387910-7-armbru@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Coverity reports potential NULL pointer dereference in
get_supported_hv_cpuid_legacy() when 'cs->kvm_state' is NULL. While
'cs->kvm_state' can indeed be NULL in hv_cpuid_get_host(),
kvm_hyperv_expand_features() makes sure that it only happens when
KVM_CAP_SYS_HYPERV_CPUID is supported and KVM_CAP_SYS_HYPERV_CPUID
implies KVM_CAP_HYPERV_CPUID so get_supported_hv_cpuid_legacy() is
never really called. Add asserts to strengthen the protection against
broken KVM behavior.
Coverity: CID 1458243
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210716115852.418293-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When Hyper-V SynIC is enabled, we may need to allow Windows guests to make
hypercalls (POST_MESSAGES/SIGNAL_EVENTS). No issue is currently observed
because KVM is very permissive, allowing these hypercalls regarding of
guest visible CPUid bits.
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210608120817.1325125-9-vkuznets@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
According to TLFS, Hyper-V guest is supposed to check
HV_HYPERCALL_AVAILABLE privilege bit before accessing
HV_X64_MSR_GUEST_OS_ID/HV_X64_MSR_HYPERCALL MSRs but at least some
Windows versions ignore that. As KVM is very permissive and allows
accessing these MSRs unconditionally, no issue is observed. We may,
however, want to tighten the checks eventually. Conforming to the
spec is probably also a good idea.
Enable HV_HYPERCALL_AVAILABLE bit unconditionally.
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210608120817.1325125-8-vkuznets@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
hv_cpuid_check_and_set() does too much:
- Checks if the feature is supported by KVM;
- Checks if all dependencies are enabled;
- Sets the feature bit in cpu->hyperv_features for 'passthrough' mode.
To reduce the complexity, move all the logic except for dependencies
check out of it. Also, in 'passthrough' mode we don't really need to
check dependencies because KVM is supposed to provide a consistent
set anyway.
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210608120817.1325125-7-vkuznets@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
To make Hyper-V features appear in e.g. QMP query-cpu-model-expansion we
need to expand and set the corresponding CPUID leaves early. Modify
x86_cpu_get_supported_feature_word() to call newly intoduced Hyper-V
specific kvm_hv_get_supported_cpuid() instead of
kvm_arch_get_supported_cpuid(). We can't use kvm_arch_get_supported_cpuid()
as Hyper-V specific CPUID leaves intersect with KVM's.
Note, early expansion will only happen when KVM supports system wide
KVM_GET_SUPPORTED_HV_CPUID ioctl (KVM_CAP_SYS_HYPERV_CPUID).
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20210608120817.1325125-6-vkuznets@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Currently, the only eVMCS version, supported by KVM (and described in TLFS)
is '1'. When Enlightened VMCS feature is enabled, QEMU takes the supported
eVMCS version range (from KVM_CAP_HYPERV_ENLIGHTENED_VMCS enablement) and
puts it to guest visible CPUIDs. When (and if) eVMCS ver.2 appears a
problem on migration is expected: it doesn't seem to be possible to migrate
from a host supporting eVMCS ver.2 to a host, which only support eVMCS
ver.1.
Hardcode eVMCS ver.1 as the result of 'hv-evmcs' enablement for now. Newer
eVMCS versions will have to have their own enablement options (e.g.
'hv-evmcs=2').
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20210608120817.1325125-4-vkuznets@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Rather than relying on the X86XSaveArea structure definition,
determine the offset of XSAVE state areas using CPUID leaf 0xd where
possible (KVM and HVF).
Signed-off-by: David Edmondson <david.edmondson@oracle.com>
Message-Id: <20210705104632.2902400-8-david.edmondson@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
In preparation for removing assumptions about XSAVE area offsets, pass
a buffer pointer and buffer length to the XSAVE helper functions.
Signed-off-by: David Edmondson <david.edmondson@oracle.com>
Message-Id: <20210705104632.2902400-5-david.edmondson@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Rather than having similar but different checks in cpu.h and kvm.c,
move them all to cpu.h.
Message-Id: <20210705104632.2902400-3-david.edmondson@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Linux 5.14 will add support for nested TSC scaling. Add the
corresponding feature in QEMU; to keep support for existing kernels,
do not add it to any processor yet.
The handling of the VMCS enumeration MSR is ugly; once we have more than
one case, we may want to add a table to check VMX features against.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A bus lock is acquired through either split locked access to writeback
(WB) memory or any locked access to non-WB memory. It is typically >1000
cycles slower than an atomic operation within a cache and can also
disrupts performance on other cores.
Virtual Machines can exploit bus locks to degrade the performance of
system. To address this kind of performance DOS attack coming from the
VMs, bus lock VM exit is introduced in KVM and it can report the bus
locks detected in guest. If enabled in KVM, it would exit to the
userspace to let the user enforce throttling policies once bus locks
acquired in VMs.
The availability of bus lock VM exit can be detected through the
KVM_CAP_X86_BUS_LOCK_EXIT. The returned bitmap contains the potential
policies supported by KVM. The field KVM_BUS_LOCK_DETECTION_EXIT in
bitmap is the only supported strategy at present. It indicates that KVM
will exit to userspace to handle the bus locks.
This patch adds a ratelimit on the bus locks acquired in guest as a
mitigation policy.
Introduce a new field "bus_lock_ratelimit" to record the limited speed
of bus locks in the target VM. The user can specify it through the
"bus-lock-ratelimit" as a machine property. In current implementation,
the default value of the speed is 0 per second, which means no
restrictions on the bus locks.
As for ratelimit on detected bus locks, simply set the ratelimit
interval to 1s and restrict the quota of bus lock occurence to the value
of "bus_lock_ratelimit". A potential alternative is to introduce the
time slice as a property which can help the user achieve more precise
control.
The detail of bus lock VM exit can be found in spec:
https://software.intel.com/content/www/us/en/develop/download/intel-architecture-instruction-set-extensions-programming-reference.html
Signed-off-by: Chenyi Qiang <chenyi.qiang@intel.com>
Message-Id: <20210521043820.29678-1-chenyi.qiang@intel.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
There is no need to use vCPU-specific kvm state in hyperv_enabled() check
and we need to do that when feature expansion happens early, before vCPU
specific KVM state is created.
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20210422161130.652779-15-vkuznets@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Vitaly Kuznetsov
Yang Weijiang
Jon Doron
luofei
Philippe Mathieu-Daudé
Paolo Bonzini
Zeng Guang
Jing Liu
Yang Zhong
Longpeng(Mike)
Peter Maydell
Maxim Levitsky
Philippe Mathieu-Daudé
Peter Xu
Sean Christopherson
Markus Armbruster
David Edmondson
Chenyi Qiang