kvm-unit-test: nVMX: Modify test_canonical() to process guest fields also

Signed-off-by: Krish Sadhukhkan <krish.sadhukhan@oracle.com>
Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Signed-off-

kvm-unit-test: nVMX: Modify test_canonical() to process guest fields also

Signed-off-by: Krish Sadhukhkan <krish.sadhukhan@oracle.com>
Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

Switch the order of the parameters in report() and report_xfail()

Commit c09c54c66b1df ("lib: use an argument which doesn't require
default argument promotion") fixed a warning that occurs with Clan

Switch the order of the parameters in report() and report_xfail()

Commit c09c54c66b1df ("lib: use an argument which doesn't require
default argument promotion") fixed a warning that occurs with Clang,
but introduced a regression: If the "pass" parameter is a value
which has only set the condition bits in the upper 32 bits of a
64 bit value, the condition is now false since the value is truncated
to "unsigned int" so that the upper bits are simply discarded.

We fixed it by reverting the commit, but that of course also means
trouble with Clang again. We can not use "bool" if it is the last
parameter before the variable argument list. The proper fix is to
swap the parameters around and make the format string the last
parameter.

This patch (except the changes in lib/libcflat.h and lib/report.c
and some rebase conflicts along the way) has basically been created
with following coccinelle script (with some additional manual tweaking
of long and disabled lines afterwards):

@@
expression fmt;
expression pass;
expression list args;
@@
report(
-fmt, pass
+pass, fmt
, args);

@@
expression fmt;
expression pass;
expression list args;
@@
report_xfail(
-fmt, xfail, pass
+xfail, pass, fmt
, args);

Tested-by: Andrew Jones <drjones@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20191206113102.14914-1-thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: Add RDTSC test

Verify that the difference between a guest RDTSC instruction and the
IA32_TIME_STAMP_COUNTER MSR value stored in the VMCS12's VM-exit
MSR-store list is less than 750 cycles, 99.9

x86: Add RDTSC test

Verify that the difference between a guest RDTSC instruction and the
IA32_TIME_STAMP_COUNTER MSR value stored in the VMCS12's VM-exit
MSR-store list is less than 750 cycles, 99.9% of the time.

Testcase for Linux kernel commit 662f1d1d1931 ("KVM: nVMX: Add support
for capturing highest observable L2 TSC”).

Signed-off-by: Aaron Lewis <aaronlewis@google.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Reviewed-by: Liran Alon <liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: vmx: Fix the check whether CMCI is supported

The logic of figuring out whether CMCI is supported is broken, causing
the CMCI accessing tests to fail on Skylake bare-metal.

Determine whether CM

x86: vmx: Fix the check whether CMCI is supported

The logic of figuring out whether CMCI is supported is broken, causing
the CMCI accessing tests to fail on Skylake bare-metal.

Determine whether CMCI is supported according to the maximum entries in
the LVT as encoded in the APIC version register.

Suggested-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Nadav Amit <namit@vmware.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: VMX: Add tests for nested "load IA32_PERF_GLOBAL_CTRL"

Tests to verify that KVM performs the correct checks on Host/Guest state
at VM-entry, as described in SDM 26.3.1.1 "Checks on Guest Contro

x86: VMX: Add tests for nested "load IA32_PERF_GLOBAL_CTRL"

Tests to verify that KVM performs the correct checks on Host/Guest state
at VM-entry, as described in SDM 26.3.1.1 "Checks on Guest Control
Registers, Debug Registers, and MSRs" and SDM 26.2.2 "Checks on Host
Control Registers and MSRs".

Test that KVM does the following:

If the "load IA32_PERF_GLOBAL_CTRL" VM-entry control is 1, the
reserved bits of the IA32_PERF_GLOBAL_CTRL MSR must be 0 in the
GUEST_IA32_PERF_GLOBAL_CTRL VMCS field. Otherwise, the VM-entry
should fail with an exit reason of "VM-entry failure due to invalid
guest state" (33). On a successful VM-entry, the correct value
should be observed when the nested VM performs an RDMSR on
IA32_PERF_GLOBAL_CTRL.

If the "load IA32_PERF_GLOBAL_CTRL" VM-exit control is 1, the
reserved bits of the IA32_PERF_GLOBAL_CTRL MSR must be 0 in the
HOST_IA32_PERF_GLOBAL_CTRL VMCS field. Otherwise, the VM-entry
should fail with a VM-instruction error of "VM entry with invalid
host-state field(s)" (8). On a successful VM-exit, the correct value
should be observed when L1 performs an RDMSR on
IA32_PERF_GLOBAL_CTRL.

Suggested-by: Jim Mattson <jmattson@google.com>
Co-developed-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Signed-off-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Signed-off-by: Oliver Upton <oupton@google.com>
Reviewed-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: VMX: Make guest_state_test_main() check state from nested VM

Introduce the vmx_state_area_test_data struct for sharing test
expectation data with the nested VM.

Signed-off-by: Oliver Upton <ou

x86: VMX: Make guest_state_test_main() check state from nested VM

Introduce the vmx_state_area_test_data struct for sharing test
expectation data with the nested VM.

Signed-off-by: Oliver Upton <oupton@google.com>
Reviewed-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: vmx: Verify pending LAPIC INIT event consume when exit on VMX_INIT

Intel SDM section 25.2 OTHER CAUSES OF VM EXITS specifies the following
on INIT signals: "Such exits do not modify register st

x86: vmx: Verify pending LAPIC INIT event consume when exit on VMX_INIT

Intel SDM section 25.2 OTHER CAUSES OF VM EXITS specifies the following
on INIT signals: "Such exits do not modify register state or clear pending
events as they would outside of VMX operation."

When commit 48adfb0f2e8e ("x86: vmx: Test INIT processing during various CPU VMX states")
was applied, I interepted above Intel SDM statement such that
VMX_INIT exit don’t consume the pending LAPIC INIT event.

However, when Nadav Amit run the unit-test on a bare-metal
machine, it turned out my interpetation was wrong. i.e. VMX_INIT
exit does consume the pending LAPIC INIT event.
(See: https://www.spinics.net/lists/kvm/msg196757.html)

Therefore, fix unit-test code to behave as observed on bare-metal.
i.e. End unit-test with the following steps:
1) Exit VMX operation and verify it still continues to run properly
as pending LAPIC INIT event should have been already consumed by
VMX_INIT exit.
2) Re-enter VMX operation and send another INIT signal to keep it
blocked until exit from VMX operation.
3) Exit VMX operation and verify that pending LAPIC INIT signal
is processed when exiting VMX operation.

Fixes: 48adfb0f2e8e ("x86: vmx: Test INIT processing during various CPU VMX states")
Reported-by: Nadav Amit <nadav.amit@gmail.com>
Reviewed-by: Mihai Carabas <mihai.carabas@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: vmx: Verify L2 modification to L1 LAPIC TPR works when L0 use TPR threshold

Test aims to verify that the issue fixed by commit ("KVM: nVMX: Update
vmcs01 TPR_THRESHOLD if L2 changed L1 TPR") is

x86: vmx: Verify L2 modification to L1 LAPIC TPR works when L0 use TPR threshold

Test aims to verify that the issue fixed by commit ("KVM: nVMX: Update
vmcs01 TPR_THRESHOLD if L2 changed L1 TPR") is indeed fixed.

Test performs the following steps:
1) Disable interrupts.
2) Raise TPR to high value and queue a pending interrupt in LAPIC by
issueing a self-IPI with lower priority.
3) Launch guest such that it is provided with passthrough access to
LAPIC.
4) Inside guest, disable interrupts and lower TPR to 0 and then exit guest.
5) Back on host, verify that indeed TPR was set to 0 and that enabling
interrupts indeed deliever pending interrupt in LAPIC.

Without above mentioned commit in L0, step (2) will cause L0 to raise
TPR-threshold to self-IPI vector priority and step (4) will *not* change
vmcs01 TPR-threshold to 0. This will result in infinite loop of VMExits
on TPR_BELOW_THRESHOLD every time L0 attempts to enter L1. Which will
cause test to hang and eventually fail on timeout.

Reviewed-by: Joao Martins <joao.m.martins@oracle.com>
Signed-off-by: Liran Alon <liran.alon@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: VMX: Mask advanced VM-exit info

Bits [9:11] are undefined in the VM-exit qualification when "advanced
VM-exit information for EPT violations" is not supported.

Mask these bits for now to avoid

x86: VMX: Mask advanced VM-exit info

Bits [9:11] are undefined in the VM-exit qualification when "advanced
VM-exit information for EPT violations" is not supported.

Mask these bits for now to avoid false failures. If KVM supports this
feature, the tests would need to be adapted, and the masking would need
to be removed.

Unfortunately, I do not have hardware that supports this feature
available for my use to make a better fix.

Signed-off-by: Nadav Amit <namit@vmware.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: Skip APIC-access address tests beyond mapped RAM

We no longer have any tests in vmx_tests.c that use
xfail_beyond_mapped_ram. However, an upcoming change to kvm will exit
to userspace with a kv

x86: Skip APIC-access address tests beyond mapped RAM

We no longer have any tests in vmx_tests.c that use
xfail_beyond_mapped_ram. However, an upcoming change to kvm will exit
to userspace with a kvm internal error whenever launching a nested VM
with the vmcs12 APIC-access address set to a non-cacheable address in
L1. Reuse the xfail_beyond_mapped_ram plumbing to support
skip_beyond_mapped_ram, and skip any APIC-access address tests that
use addresses beyond mapped RAM, so that the test won't induce a kvm
internal error.

Signed-off-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: vmx_tests: add GUEST_EFER tests

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

x86: vmx_tests: prepare for extending guest state area tests

Move the test setup and teardown from test_pat to vmx_guest_state_area_test,
so that we can add more tests after test_load_guest_pat.

Si

x86: vmx_tests: prepare for extending guest state area tests

Move the test setup and teardown from test_pat to vmx_guest_state_area_test,
so that we can add more tests after test_load_guest_pat.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: vmx_tests: extend HOST_EFER tests

Extend the tests to cover the host address size bit.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

x86: vmx_tests: gate load guest PAT tests on the correct bit

The tests should check the "load PAT" entry control,
not the exit control.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

x86: vmx: Test INIT processing during various CPU VMX states

Add vmx test to verify the functionality introduced by KVM commit:
4b9852f4f389 ("KVM: x86: Fix INIT signal handling in various CPU state

x86: vmx: Test INIT processing during various CPU VMX states

Add vmx test to verify the functionality introduced by KVM commit:
4b9852f4f389 ("KVM: x86: Fix INIT signal handling in various CPU states").

The test verifies the following functionality:
1) An INIT signal received when CPU is in VMX operation
is blocked until it exits VMX operation.
2) If there is an INIT signal pending when CPU is in
VMX non-root mode, it result in VMExit with (reason == 3).
3) Exit from VMX non-root mode on VMExit do not clear
pending INIT signal in LAPIC.
4) When CPU exits VMX operation, pending INIT signal in
LAPIC is processed.

Note: The test is excluded from the "vmx" test-suite as
it ruins the execution environment of CPU1 because
during the test it process an INIT signal.

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>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: nvmx: test max atomic switch MSRs

Excerise nested VMX's atomic MSR switch code (e.g., VM-entry MSR-load
list) at the maximum number of MSRs supported, as described in the SDM,
in the appendix c

x86: nvmx: test max atomic switch MSRs

Excerise nested VMX's atomic MSR switch code (e.g., VM-entry MSR-load
list) at the maximum number of MSRs supported, as described in the SDM,
in the appendix chapter titled "MISCELLANEOUS DATA".

Suggested-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Marc Orr <marcorr@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

kvm-unit-test: nVMX: Fix 95d6d2c32288 ("nVMX: Test Host Segment Registers and Descriptor Tables on vmentry of nested guests")

Commit 95d6d2c32288 added a test for the Segment Selector VMCS field. Th

kvm-unit-test: nVMX: Fix 95d6d2c32288 ("nVMX: Test Host Segment Registers and Descriptor Tables on vmentry of nested guests")

Commit 95d6d2c32288 added a test for the Segment Selector VMCS field. That test
sets the "host address-space size" VM-exit control to zero and as a result, on
VM-exit the guest exits as 32-bit. Since vmx tests are 64-bit, this results in
a hardware error.
This patch also cleans up a few other areas in commit 95d6d2c32288, including
replacing make_non_canonical() with NONCANONICAL.

Reported-by: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: Krish Sadhukhan <kris.sadhukhan@oracle.com>
Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

kvm-unit-test: nVMX: Check Host Address Space Size on vmentry of nested guests

According to section "Checks Related to Address-Space Size" in Intel SDM vol
3C, the following checks are performed on

kvm-unit-test: nVMX: Check Host Address Space Size on vmentry of nested guests

According to section "Checks Related to Address-Space Size" in Intel SDM vol
3C, the following checks are performed on vmentry of nested guests:

If the "host address-space size" VM-exit control is 0, the following must
hold:
- The "IA-32e mode guest" VM-entry control is 0.
- Bit 17 of the CR4 field (corresponding to CR4.PCIDE) is 0.
- Bits 63:32 in the RIP field are 0.

If the "host address-space size" VM-exit control is 1, the following must
hold:
- Bit 5 of the CR4 field (corresponding to CR4.PAE) is 1.
- The RIP field contains a canonical address.

Signed-off-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: VMX: INVEPT after modifying PA mapping in ept_untwiddle

ept_untwiddle modifies a PA mapping in the EPT paging structure.
According to the SDM 28.3.3.4, "Software should use the INVEPT
instructi

x86: VMX: INVEPT after modifying PA mapping in ept_untwiddle

ept_untwiddle modifies a PA mapping in the EPT paging structure.
According to the SDM 28.3.3.4, "Software should use the INVEPT
instruction with the "single-context" INVEPT type after making any of
the following changes to an EPT paging-structure entry ... Changing
the physical address in bits 51:12".

Suggested-by: Peter Shier <pshier@google.com>
Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: nVMX: Fix wrong reserved bits of error-code

The SDM indeed says that "If deliver-error-code is 1, bits 31:15 of the
VM-entry exception error-code field are 0." However, the SDM is wrong,
and bi

x86: nVMX: Fix wrong reserved bits of error-code

The SDM indeed says that "If deliver-error-code is 1, bits 31:15 of the
VM-entry exception error-code field are 0." However, the SDM is wrong,
and bits that need to be zeroed are 31:16.

Our engineers confirmed that the SDM is wrong with Intel. Fix the test.

Note that KVM should be fixed as well.

Fixes: 8d2cdb35a07a ("x86: Add test for nested VM entry prereqs")
Cc: Marc Orr <marcorr@google.com>
Signed-off-by: Nadav Amit <namit@vmware.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: nVMX: Do not use test_skip() when multiple tests are run

Using test_skip() when multiple tests are run causes all the following
tests to be skipped. Instead, just print a message and return.

F

x86: nVMX: Do not use test_skip() when multiple tests are run

Using test_skip() when multiple tests are run causes all the following
tests to be skipped. Instead, just print a message and return.

Fixes: 47cc3d85c2fe ("nVMX x86: Check PML and EPT on vmentry of L2 guests")
Fixes: 7fd449f2ed2e ("nVMX x86: Check VPID value on vmentry of L2 guests")
Fixes: 181219bfd76b ("x86: Add test for checking NMI controls on vmentry of L2 guests")
Fixes: 1d70eb823e12 ("nVMX x86: Check EPTP on vmentry of L2 guests")
Cc: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Signed-off-by: Nadav Amit <namit@vmware.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: Skip EPT tests that involve unrestricted guest when URG is disabled

Cc: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Fixes: f749ddc19bb2d ("nVMX x86: Check enable-EPT on vmentry of L2 guests")

x86: Skip EPT tests that involve unrestricted guest when URG is disabled

Cc: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Fixes: f749ddc19bb2d ("nVMX x86: Check enable-EPT on vmentry of L2 guests")
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: Support environments without test-devices

Enable to run the tests when test-device is not present (e.g.,
bare-metal). Users can provide the number of CPUs and ram size through
kernel parameters

x86: Support environments without test-devices

Enable to run the tests when test-device is not present (e.g.,
bare-metal). Users can provide the number of CPUs and ram size through
kernel parameters.

On Ubuntu that uses grub, for example, the tests can be run by copying a
test to the boot directory (/boot) and adding a menu-entry to grub
(e.g., by editing /etc/grub.d/40_custom):

menuentry 'idt_test' {
set root='[ROOT]'
multiboot [BOOT_RELATIVE]/[TEST].flat [PARAMETERS]
module params.initrd
}

Replace:
* [ROOT] with `grub-probe --target=bios_hints /boot`
* [BOOT_RELATIVE] with `grub-mkrelpath /boot`
* [TEST] with the test executed
* [PARAMETERS] with the test parameters

params.initrd, which would be located on the boot directory should
describe the machine and tell the test infrastructure that a test
device is not present and boot-loader was used (the bootloader and qemu
deliver test . For example for a 4 core machines with 4GB of
memory:

NR_CPUS=4
MEMSIZE=4096
TEST_DEVICE=0
BOOTLOADER=1

Since we do not really use E820, using more than 4GB is likely to fail
due to holes.

Finally, do not forget to run update-grub. Remember that the output goes
to the serial port.

Cc: Andrew Jones <drjones@redhat.com>
Signed-off-by: Nadav Amit <nadav.amit@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

x86: vmx: Fix comment typo for ept_access_paddr()

the comment for ept_access_paddr() referred to a nonexistent
ept_access_data. Fixing the comment to correctly refer to
ept_access_test_data.

Signed

x86: vmx: Fix comment typo for ept_access_paddr()

the comment for ept_access_paddr() referred to a nonexistent
ept_access_data. Fixing the comment to correctly refer to
ept_access_test_data.

Signed-off-by: Oliver Upton <oupton@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...

kvm-unit-test: x86: Replace cpuid/cpuid_indexed calls with this_cpu_has()

Signed-off-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Signed-off-b

kvm-unit-test: x86: Replace cpuid/cpuid_indexed calls with this_cpu_has()

Signed-off-by: Krish Sadhukhan <krish.sadhukhan@oracle.com>
Reviewed-by: Karl Heubaum <karl.heubaum@oracle.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

show more ...