Turns out we over-engineered Direct Mode for stimers a bit: unlike
traditional stimers where we may want to try to re-inject the message upon
EOI, Direct Mode stimers just set the irq in APIC and kvm_apic_set_irq()
fails only when APIC is disabled (see APIC_DM_FIXED case in
__apic_accept_irq()). Remove the redundant part.
Suggested-by: NRoman Kagan <rkagan@virtuozzo.com>
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

stimers_pending optimization only helps us to avoid multiple
kvm_make_request() calls. This doesn't happen very often and these
calls are very cheap in the first place, remove open-coded version of
stimer_mark_pending() from kvm_hv_notify_acked_sint().
Suggested-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: NRoman Kagan <rkagan@virtuozzo.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Turns out Hyper-V on KVM (as of 2016) will only use synthetic timers
if direct mode is available. With direct mode we notify the guest by
asserting APIC irq instead of sending a SynIC message.

The implementation uses existing vec_bitmap for letting lapic code
know that we're interested in the particular IRQ's EOI request. We assume
that the same APIC irq won't be used by the guest for both direct mode
stimer and as sint source (especially with AutoEOI semantics). It is
unclear how things should be handled if that's not true.

Direct mode is also somewhat less expensive; in my testing
stimer_send_msg() takes not less than 1500 cpu cycles and
stimer_notify_direct() can usually be done in 300-400. WS2016 without
Hyper-V, however, always sticks to non-direct version.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: NRoman Kagan <rkagan@virtuozzo.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

As a preparation to implementing Direct Mode for Hyper-V synthetic
timers switch to using stimer config definition from hyperv-tlfs.h.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

We implement Hyper-V SynIC and synthetic timers in KVM too so there's some
room for code sharing.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: NMichael Kelley <mikelley@microsoft.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

With every new Hyper-V Enlightenment we implement we're forced to add a
KVM_CAP_HYPERV_* capability. While this approach works it is fairly
inconvenient: the majority of the enlightenments we do have corresponding
CPUID feature bit(s) and userspace has to know this anyways to be able to
expose the feature to the guest.

Add KVM_GET_SUPPORTED_HV_CPUID ioctl (backed by KVM_CAP_HYPERV_CPUID, "one
cap to rule them all!") returning all Hyper-V CPUID feature leaves.

Using the existing KVM_GET_SUPPORTED_CPUID doesn't seem to be possible:
Hyper-V CPUID feature leaves intersect with KVM's (e.g. 0x40000000,
0x40000001) and we would probably confuse userspace in case we decide to
return these twice.

KVM_CAP_HYPERV_CPUID's number is interim: we're intended to drop
KVM_CAP_HYPERV_STIMER_DIRECT and use its number instead.
Suggested-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The upcoming KVM_GET_SUPPORTED_HV_CPUID ioctl will need to return
Enlightened VMCS version in HYPERV_CPUID_NESTED_FEATURES.EAX when
it was enabled.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

BIT(13) in HYPERV_CPUID_FEATURES.EBX is described as "ConfigureProfiler" in
TLFS v4.0 but starting 5.0 it is replaced with 'Reserved'. As we don't
currently us it in kernel it can just be dropped.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: NMichael Kelley <mikelley@microsoft.com>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

hyperv-tlfs.h is a bit messy: CPUID feature bits are not always sorted,
it's hard to get which CPUID they belong to, some items are duplicated
(e.g. HV_X64_MSR_CRASH_CTL_NOTIFY/HV_CRASH_CTL_CRASH_NOTIFY).

Do some housekeeping work. While on it, replace all (1 << X) with BIT(X)
macro.
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: NMichael Kelley <mikelley@microsoft.com>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The TLFS structures are used for hypervisor-guest communication and must
exactly meet the specification.

Compilers can add alignment padding to structures or reorder struct members
for randomization and optimization, which would break the hypervisor ABI.

Mark the structures as packed to prevent this. 'struct hv_vp_assist_page'
and 'struct hv_enlightened_vmcs' need to be properly padded to support the
change.
Suggested-by: NNadav Amit <nadav.amit@gmail.com>
Signed-off-by: NVitaly Kuznetsov <vkuznets@redhat.com>
Acked-by: NThomas Gleixner <tglx@linutronix.de>
Acked-by: NNadav Amit <nadav.amit@gmail.com>
Reviewed-by: NMichael Kelley <mikelley@microsoft.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The SynIC message delivery protocol allows the message originator to
request, should the message slot be busy, to be notified when it's free.

However, this is unnecessary and even undesirable for messages generated
by SynIC timers in periodic mode: if the period is short enough compared
to the time the guest spends in the timer interrupt handler, so the
timer ticks start piling up, the excessive interactions due to this
notification and retried message delivery only makes the things worse.

[This was observed, in particular, with Windows L2 guests setting
(temporarily) the periodic timer to 2 kHz, and spending hundreds of
microseconds in the timer interrupt handler due to several L2->L1 exits;
under some load in L0 this could exceed 500 us so the timer ticks
started to pile up and the guest livelocked.]

Relieve the situation somewhat by not retrying message delivery for
periodic SynIC timers.  This appears to remain within the "lazy" lost
ticks policy for SynIC timers as implemented in KVM.

Note that it doesn't solve the fundamental problem of livelocking the
guest with a periodic timer whose period is smaller than the time needed
to process a tick, but it makes it a bit less likely to be triggered.
Signed-off-by: NRoman Kagan <rkagan@virtuozzo.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

SynIC message delivery is somewhat overengineered: it pretends to follow
the ordering rules when grabbing the message slot, using atomic
operations and all that, but does it incorrectly and unnecessarily.

The correct order would be to first set .msg_pending, then atomically
replace .message_type if it was zero, and then clear .msg_pending if
the previous step was successful.  But this all is done in vcpu context
so the whole update looks atomic to the guest (it's assumed to only
access the message page from this cpu), and therefore can be done in
whatever order is most convenient (and is also the reason why the
incorrect order didn't trigger any bugs so far).

While at this, also switch to kvm_vcpu_{read,write}_guest_page, and drop
the no longer needed synic_clear_sint_msg_pending.
Signed-off-by: NRoman Kagan <rkagan@virtuozzo.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

In the previous code, the variable apic_sw_disabled influences
recalculate_apic_map. But in "KVM: x86: simplify kvm_apic_map"
(commit: 3b5a5ffa),
the access to apic_sw_disabled in recalculate_apic_map has been
deleted.
Signed-off-by: NPeng Hao <peng.hao2@zte.com.cn>
Reviewed-by: NRadim Krčmář <rkrcmar@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

structure svm_init_data is never used. So remove it.
Signed-off-by: NPeng Hao <peng.hao2@zte.com.cn>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Like IA32_STAR, IA32_LSTAR and IA32_FMASK only need to contain guest
values on VM-entry when the guest is in long mode and EFER.SCE is set.
Signed-off-by: NJim Mattson <jmattson@google.com>
Reviewed-by: NPeter Shier <pshier@google.com>
Reviewed-by: NMarc Orr <marcorr@google.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

SYSCALL raises #UD in compatibility mode on Intel CPUs, so it's
pointless to load the guest's IA32_CSTAR value into the hardware MSR.

IA32_CSTAR still provides 48 bits of storage on Intel CPUs that have
CPUID.80000001:EDX.LM[bit 29] set, so we cannot remove it from the
vmx_msr_index[] array.
Signed-off-by: NJim Mattson <jmattson@google.com>
Reviewed-by: NPeter Shier <pshier@google.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Add a comment explaining why MSR_STAR must be included in
vmx_msr_index[] even for i386 builds.

The elided comment has not been relevant since move_msr_up() was
introduced in commit a75beee6 ("KVM: VMX: Avoid saving and
restoring msrs on lightweight vmexit").
Signed-off-by: NJim Mattson <jmattson@google.com>
Reviewed-by: NPeter Shier <pshier@google.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

RDTSCP is supported in legacy mode as well as long mode. The
IA32_TSC_AUX MSR should be set to the correct guest value before
entering any guest that supports RDTSCP.

Fixes: 4e47c7a6 ("KVM: VMX: Add instruction rdtscp support for guest")
Signed-off-by: NJim Mattson <jmattson@google.com>
Reviewed-by: NPeter Shier <pshier@google.com>
Reviewed-by: NMarc Orr <marcorr@google.com>
Reviewed-by: NLiran Alon <liran.alon@oracle.com>
Reviewed-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

From a functional perspective, this is (supposed to be) a straight
copy-paste of code.  Code was moved piecemeal to nested.c as not all
code that could/should be moved was obviously nested-only.  The nested
code was then re-ordered as needed to compile, i.e. stats may not show
this is being a "pure" move despite there not being any intended changes
in functionality.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Exposing only the function allows @nested, i.e. the module param, to be
statically defined in vmx.c, ensuring we aren't unnecessarily checking
said variable in the nested code.  nested_vmx_allowed() is exposed due
to the need to verify nested support in vmx_{get,set}_nested_state().
The downside is that nested_vmx_allowed() likely won't be inlined in
vmx_{get,set}_nested_state(), but that should be a non-issue as they're
not a hot path.  Keeping vmx_{get,set}_nested_state() in vmx.c isn't a
viable option as they need access to several nested-only functions.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

...as they're used directly by the nested code.  This will allow
moving the bulk of the nested code out of vmx.c without concurrent
changes to vmx.h.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Exposed vmx_msr_index, vmx_return and host_efer via vmx.h so that the
nested code can be moved out of vmx.c.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

...so that the function doesn't need to be created when moving the
nested code out of vmx.c.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

...so that it doesn't need access to @nested. The only case where the
provided struct isn't already zeroed is the call from vmx_create_vcpu()
as setup_vmcs_config() zeroes the struct in the other use cases.  This
will allow @nested to be statically defined in vmx.c, i.e. this removes
the last direct reference from nested code.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

...in nested-specific code so that they can eventually be moved out of
vmx.c, e.g. into nested.c.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

...so that future patches can reference e.g. @kvm_vmx_exit_handlers
without having to simultaneously move a big chunk of code.  Speaking
from experience, resolving merge conflicts is an absolute nightmare
without pre-moving the code.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

...so that it can conditionally set by the VMX code, i.e. iff @nested is
true.  This will in turn allow it to be moved out of vmx.c and into a
nested-specified file.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Eventually this will allow us to move the nested VMX code out of vmx.c.
Note that this also effectively wraps @enable_shadow_vmcs with @nested
so that it too can be moved out of vmx.c.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

VMX has a few hundred lines of code just to wrap various VMX specific
instructions, e.g. VMWREAD, INVVPID, etc...  Move them to a dedicated
header so it's easier to find/isolate the boilerplate.

With this change, more inlines can be moved from vmx.c to vmx.h.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The header, evmcs.h, already exists and contains a fair amount of code,
but there are a few pieces in vmx.c that can be moved verbatim.  In
addition, move an array definition to evmcs.c to prepare for multiple
consumers of evmcs.h.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

vmcs12 is the KVM-defined struct used to track a nested VMCS, e.g. a
VMCS created by L1 for L2.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

This isn't intended to be a pure reflection of hardware, e.g. struct
loaded_vmcs and struct vmcs_host_state are KVM-defined constructs.
Similar to capabilities.h, this is a standalone file to avoid circular
dependencies between yet-to-be-created vmx.h and nested.h files.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Expose the variables associated with various module params that are
needed by the nested VMX code.  There is no ulterior logic for what
variables are/aren't exposed, this is purely "what's needed by the
nested code".

Note that @nested is intentionally not exposed.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Defining a separate capabilities.h as opposed to putting this code in
e.g. vmx.h avoids circular dependencies between (the yet-to-be-added)
vmx.h and nested.h.  The aforementioned circular dependencies are why
struct nested_vmx_msrs also resides in capabilities instead of e.g.
nested.h.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

...instead of referencing the global struct.  This will allow moving
setup_vmcs_config() to a separate file that may not have access to
the global variable.  Modify nested_vmx_setup_ctls_msrs() appropriately
since vmx_capability.ept may not be accurate when called by
vmx_check_processor_compat().

No functional change intended.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

EFER and PERF_GLOBAL_CTRL MSRs have dedicated VM Entry/Exit controls
that KVM dynamically toggles based on whether or not the guest's value
for each MSRs differs from the host.  Handle the dynamic behavior by
adding a helper that clears the dynamic bits so the bits aren't set
when initializing the VMCS field outside of the dynamic toggling flow.
This makes the handling consistent with similar behavior for other
controls, e.g. pin, exec and sec_exec.  More importantly, it eliminates
two global bools that are stealthily modified by setup_vmcs_config.

Opportunistically clean up a comment and print related to errata for
IA32_PERF_GLOBAL_CTRL.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

MSR_IA32_XSS has no relation to the VMCS whatsoever, it doesn't belong
in setup_vmcs_config() and its reference to host_xss prevents moving
setup_vmcs_config() to a dedicated file.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

VMX specific files now reside in a dedicated subdirectory, i.e. the
file name prefix is redundant.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

VMX specific files now reside in a dedicated subdirectory.  Drop the
"vmx" prefix, which is redundant, and add a "vmcs" prefix to clarify
that the file is referring to VMCS shadow fields.
Signed-off-by: NSean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>