nVMX: Assert that the test is configured for 64-bit mode

The nVMX tests are 64-bit only, assert as much to reduce the probability
of the tests not actually running due to some funky bug. It's easy

nVMX: Assert that the test is configured for 64-bit mode

The nVMX tests are 64-bit only, assert as much to reduce the probability
of the tests not actually running due to some funky bug. It's easy to
re-add an if-statement if the nVMX tests ever gain 32-bit support, whereas
most developers lack the psychic powers necessary to detect that a test
isn't doing anything useful.

Link: https://lore.kernel.org/r/20230901225004.3604702-4-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

nVMX: Assert CR4.PAE is set when testing 64-bit host

Assert that PAE is set in the original CR4 value when the host is setup to
VM-Exit to 64-bit mode in test_host_addr_size(). If CR4.PAE isn't set

nVMX: Assert CR4.PAE is set when testing 64-bit host

Assert that PAE is set in the original CR4 value when the host is setup to
VM-Exit to 64-bit mode in test_host_addr_size(). If CR4.PAE isn't set
then something is wildly broken and all bets are off as VM-Enter can't
possibly succeed.

Link: https://lore.kernel.org/r/20230901225004.3604702-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

nVMX: Test CR4.PCIDE can be set for 64-bit host iff PCID is supported

Check that PCID is actually supported before verifying that it can be set
in vmcs.HOST_CR4 for 64-bit hosts. PCID is all but ub

nVMX: Test CR4.PCIDE can be set for 64-bit host iff PCID is supported

Check that PCID is actually supported before verifying that it can be set
in vmcs.HOST_CR4 for 64-bit hosts. PCID is all but ubiquitous on modern
CPUs, but PCID may be hidden by the host to workaround a TLB flushing bug
on Alderlake and Raptorlake.

Link: https://lore.kernel.org/r/20230901225004.3604702-2-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

x86: Drop types.h

The file types.h never declared any types, only exception vectors which
are partially re-defined in processor.h.

Move the remaining vector definitions to processor.h and remove ty

x86: Drop types.h

The file types.h never declared any types, only exception vectors which
are partially re-defined in processor.h.

Move the remaining vector definitions to processor.h and remove types.h,
as all users already include processor.h

Signed-off-by: Mathias Krause <minipli@grsecurity.net>
Link: https://lore.kernel.org/r/20230413184219.36404-2-minipli@grsecurity.net
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

x86: introduce sti_nop() and sti_nop_cli()

Add functions that shorten the common usage of sti

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Emanuele Giuseppe Esposito <eesposit@r

x86: introduce sti_nop() and sti_nop_cli()

Add functions that shorten the common usage of sti

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Link: https://lore.kernel.org/r/20221122161152.293072-3-mlevitsk@redhat.com
[sean: massage comments]
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

x86: replace irq_{enable|disable}() with sti()/cli()

This removes a layer of indirection which is strictly
speaking not needed since its x86 code anyway.

Signed-off-by: Maxim Levitsky <mlevitsk@red

x86: replace irq_{enable|disable}() with sti()/cli()

This removes a layer of indirection which is strictly
speaking not needed since its x86 code anyway.

Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Reviewed-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Link: https://lore.kernel.org/r/20221122161152.293072-2-mlevitsk@redhat.com
[sean: move and reword IRQ shadow to be above and about safe_halt()]
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

x86/apic: Refactor x2APIC reg helper to provide exact semantics

Refactor x2apic_reg_reserved() into get_x2apic_reg_semantics() and have
it provide the semantics for all registers. The full semantic

x86/apic: Refactor x2APIC reg helper to provide exact semantics

Refactor x2apic_reg_reserved() into get_x2apic_reg_semantics() and have
it provide the semantics for all registers. The full semantics will be
used by the MSR test to verify KVM correctly emulates all x2APIC MSRs.

No functional change intended.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20230107011737.577244-3-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86/apic: Refactor x2APIC reg helper to provide exact semantics

Refactor x2apic_reg_reserved() into get_x2apic_reg_semantics() and have
it provide the semantics for all registers. The full semantic

x86/apic: Refactor x2APIC reg helper to provide exact semantics

Refactor x2apic_reg_reserved() into get_x2apic_reg_semantics() and have
it provide the semantics for all registers. The full semantics will be
used by the MSR test to verify KVM correctly emulates all x2APIC MSRs.

No functional change intended.

Link: https://lore.kernel.org/r/20230107011737.577244-3-seanjc@google.com
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

nVMX: Add forced emulation variant of #PF access test

Add a forced emulation variant of vmx_pf_exception_test to exercise KVM
emulation of L2 instructions and accesses. Like the non-nested version,

nVMX: Add forced emulation variant of #PF access test

Add a forced emulation variant of vmx_pf_exception_test to exercise KVM
emulation of L2 instructions and accesses. Like the non-nested version,
make the test nodefault as forcing KVM to emulate drastically increases
the runtime.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20230404165341.163500-10-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86/access: Add forced emulation support

Add support to force access tests to be handled by the emulator, if
supported by KVM. Due to emulation being rather slow, make the forced
emulation nodefaul

x86/access: Add forced emulation support

Add support to force access tests to be handled by the emulator, if
supported by KVM. Due to emulation being rather slow, make the forced
emulation nodefault, i.e. a manual-only testcase. Note, "manual" just
means the test runner needs to specify a specific group(s), i.e. the
test is still easy to enable in CI for compatible setups.

Manually check for FEP support in the test instead of using the "check"
option to query the module param, as any non-zero force_emulation_prefix
value enables FEP (see KVM commit d500e1ed3dc8 ("KVM: x86: Allow clearing
RFLAGS.RF on forced emulation to test code #DBs") and scripts/runtime.bash
only supports checking for a single value.

Defer enabling FEP for the nested VMX variants to a future patch. KVM has
a bug related to emulating NOP (yes, NOP) for L2, and the nVMX varaints
will require additional massaging to propagate the "allow emulation" flag.

Signed-off-by: Mathias Krause <minipli@grsecurity.net>
[sean: make generic fep testcase nodefault instead of impossible]
Signed-off-by: Sean Christopherson <seanjc@google.com>
Message-Id: <20230404165341.163500-6-seanjc@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86/pmu: Add lib/x86/pmu.[c.h] and move common code to header files

Given all the PMU stuff coming in, we need e.g. lib/x86/pmu.h to hold all
of the hardware-defined stuff, e.g. #defines, accessors,

x86/pmu: Add lib/x86/pmu.[c.h] and move common code to header files

Given all the PMU stuff coming in, we need e.g. lib/x86/pmu.h to hold all
of the hardware-defined stuff, e.g. #defines, accessors, helpers and structs
that are dictated by hardware. This will greatly help with code reuse and
reduce unnecessary vm-exit.

Opportunistically move lbr msrs definition to header processor.h.

Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Like Xu <likexu@tencent.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221102225110.3023543-14-seanjc@google.com

show more ...

nVMX: Move #NM test to generic exception test framework

Move the #NM test cases to the generic exception test framework, and
rename the dedicated test to note that it tests the "no #NM" case.

Signe

nVMX: Move #NM test to generic exception test framework

Move the #NM test cases to the generic exception test framework, and
rename the dedicated test to note that it tests the "no #NM" case.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221005235212.57836-10-seanjc@google.com

show more ...

nVMX: Drop one-off INT3=>#BP test

Drop the dedicated INT3=>#BP test, it's a subset of what is covered by
the INT3=>#BP case in the generic exceptions test.

Signed-off-by: Sean Christopherson <seanj

nVMX: Drop one-off INT3=>#BP test

Drop the dedicated INT3=>#BP test, it's a subset of what is covered by
the INT3=>#BP case in the generic exceptions test.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221005235212.57836-5-seanjc@google.com

show more ...

nVMX: Move #OF test to generic exceptions test

Move the INTO=>#OF test, along with its more precise checking of the
exit interrupt info, to the generic nVMX exceptions test.

Move the helper that g

nVMX: Move #OF test to generic exceptions test

Move the INTO=>#OF test, along with its more precise checking of the
exit interrupt info, to the generic nVMX exceptions test.

Move the helper that generates #OF to processor.h so that it can be
reused by nSVM for an identical test.

Note, this effectively adds new checks for all other vectors, i.e.
affects more vectors than just #OF.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221005235212.57836-4-seanjc@google.com

show more ...

x86: Move helpers to generate misc exceptions to processor.h

Move nested VMX's helpers to generate miscellaenous exceptions, e.g. #DE,
to processor.h so that they can be used for nearly-identical ne

x86: Move helpers to generate misc exceptions to processor.h

Move nested VMX's helpers to generate miscellaenous exceptions, e.g. #DE,
to processor.h so that they can be used for nearly-identical nested SVM
tests.

No functional change intended.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221005235212.57836-3-seanjc@google.com

show more ...

nVMX: Add "nop" after setting EFLAGS.TF to guarantee single-step #DB

Add a "nop" after enabling single-stepping in vmx_l2_db_test() to ensure
a #DB is generated in the scope of the function even if

nVMX: Add "nop" after setting EFLAGS.TF to guarantee single-step #DB

Add a "nop" after enabling single-stepping in vmx_l2_db_test() to ensure
a #DB is generated in the scope of the function even if the helper is
inlined. Enabling single-step #DBs have a one-instruction delay, e.g. if
the function were inlined and the caller generated a "terminal" VM-Exit
immediately after invoking the helper, then no #DB would be generated.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20221005235212.57836-2-seanjc@google.com

show more ...

Merge tag 'for_paolo' of https://github.com/sean-jc/kvm-unit-tests into HEAD

x86 fixes, cleanups, and new sub-tests:

- Bug fix for the VMX-preemption timer expiration test
- Refactor SVM tests

Merge tag 'for_paolo' of https://github.com/sean-jc/kvm-unit-tests into HEAD

x86 fixes, cleanups, and new sub-tests:

- Bug fix for the VMX-preemption timer expiration test
- Refactor SVM tests to split out NPT tests
- Add tests for MCE banks to MSR test
- Add SMP Support for x86 UEFI tests
- x86: nVMX: Add VMXON #UD test (and exception cleanup)
- PMU cleanup and related nVMX bug fixes

show more ...

x86: Skip perf related tests when platform cannot support

Add helpers to check whether MSR_CORE_PERF_GLOBAL_CTRL and rdpmc are
supported in KVM. When pmu is disabled with enable_pmu=0, reading
MSR_C

x86: Skip perf related tests when platform cannot support

Add helpers to check whether MSR_CORE_PERF_GLOBAL_CTRL and rdpmc are
supported in KVM. When pmu is disabled with enable_pmu=0, reading
MSR_CORE_PERF_GLOBAL_CTRL or executing rdpmc leads to #GP, so skip
related tests in this case to avoid test failure.

Opportunistically hoist mwait support check function as helper and
change related code.

Suggested-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Link: https://lore.kernel.org/r/20220711041841.126648-4-weijiang.yang@intel.com
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

nVMX: Rename monitor_support() to this_cpu_has_mwait(), drop #define

Rename VMX's callback wrapper for checking MONITOR/MWAIT support to align
with X86_FEATURE_MWAIT, and drop a related #define that

nVMX: Rename monitor_support() to this_cpu_has_mwait(), drop #define

Rename VMX's callback wrapper for checking MONITOR/MWAIT support to align
with X86_FEATURE_MWAIT, and drop a related #define that's no longer used.

No functional change intended.

Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

nVMX: Move wrappers of this_cpu_has() to nVMX's VM-Exit test

Move wrappers of this_cpu_has() whose sole purpose is to be queried as a
callback in VM-Exit tests into vmxexit.c in order to discourage

nVMX: Move wrappers of this_cpu_has() to nVMX's VM-Exit test

Move wrappers of this_cpu_has() whose sole purpose is to be queried as a
callback in VM-Exit tests into vmxexit.c in order to discourage general
use, i.e. force tests to use this_cpu_has().

No functional change intended.

Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

x86: Drop cpuid_osxsave(), just use this_cpu_has(X86_FEATURE_OSXSAVE)

Drop cpuid_osxsave(), which is just an open coded implementation of
this_cpu_has(X86_FEATURE_OSXSAVE).

Signed-off-by: Sean Chri

x86: Drop cpuid_osxsave(), just use this_cpu_has(X86_FEATURE_OSXSAVE)

Drop cpuid_osxsave(), which is just an open coded implementation of
this_cpu_has(X86_FEATURE_OSXSAVE).

Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

x86: nVMX: Use report_skip() to print messages when VMX tests are skipped

Use report_skip() for VMX tests so that the message contains the "SKIP:"
prefix, and so that the test is counted as skipped.

x86: nVMX: Use report_skip() to print messages when VMX tests are skipped

Use report_skip() for VMX tests so that the message contains the "SKIP:"
prefix, and so that the test is counted as skipped.

Opportunistically unify the message format for report_skip().

Signed-off-by: Yang Weijiang <weijiang.yang@intel.com>
Link: https://lore.kernel.org/r/20220711041841.126648-2-weijiang.yang@intel.com
[sean: add a space after the colon, fix more reports]
Signed-off-by: Sean Christopherson <seanjc@google.com>

show more ...

nVMX: Wrap VMXON in ASM_TRY(), a.k.a. in exception fixup

Gracefully handle and return faults on VMXON instead of letting VMXON
explode. The primary motivation is to be able to reuse the helper in
t

nVMX: Wrap VMXON in ASM_TRY(), a.k.a. in exception fixup

Gracefully handle and return faults on VMXON instead of letting VMXON
explode. The primary motivation is to be able to reuse the helper in
tests that verify VMXON faults when it's supposed to, but printing a nice
error message on fault is also nice.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20220608235238.3881916-9-seanjc@google.com

show more ...

nVMX: Check result of VMXON in INIT/SIPI tests

Assert that VMXON succeeds in the INIT/SIPI tests, _vmx_on() doesn't
check the result, i.e. doesn't guarantee success.

Signed-off-by: Sean Christopher

nVMX: Check result of VMXON in INIT/SIPI tests

Assert that VMXON succeeds in the INIT/SIPI tests, _vmx_on() doesn't
check the result, i.e. doesn't guarantee success.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20220608235238.3881916-8-seanjc@google.com

show more ...

x86: Use "safe" terminology instead of "checking"

Rename all helpers that eat (and return) exceptions to use "safe" instead
of "checking". This aligns KUT with the kernel and KVM selftests.

Signed

x86: Use "safe" terminology instead of "checking"

Rename all helpers that eat (and return) exceptions to use "safe" instead
of "checking". This aligns KUT with the kernel and KVM selftests.

Signed-off-by: Sean Christopherson <seanjc@google.com>
Link: https://lore.kernel.org/r/20220608235238.3881916-4-seanjc@google.com

show more ...