hdr.vmx.flags is meant for future extensions to the ABI, rejecting
invalid flags is necessary to avoid broken half-loads of the
nVMX state.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
A missing VMCS12 was not causing -EINVAL (it was just read with
copy_from_user, so it is not a security issue, but it is still
wrong). Test for VMCS12 validity and reject the nested state
if a VMCS12 is required but not present.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Setting KVM_STATE_NESTED_GUEST_MODE enables various consistency checks
on VMCS12 and therefore causes KVM_SET_NESTED_STATE to fail spuriously
with -EINVAL. Do not set the flag so that we're sure to cover the
conditions included by the test, and cover the case where VMCS12 is
set and KVM_SET_NESTED_STATE is called with invalid VMCS12 contents.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It was reported that older GCCs compile smm_test in a way that breaks
it completely:
kvm_exit: reason EXIT_CPUID rip 0x4014db info 0 0
func 7ffffffd idx 830 rax 0 rbx 0 rcx 0 rdx 0, cpuid entry not found
...
kvm_exit: reason EXIT_MSR rip 0x40abd9 info 0 0
kvm_msr: msr_read 487 = 0x0 (#GP)
...
Note, '7ffffffd' was supposed to be '80000001' as we're checking for
SVM. Dropping '-O2' from compiler flags help. Turns out, asm block in
sync_with_host() is wrong. We us 'in 0xe, %%al' instruction to sync
with the host and in 'AL' register we actually pass the parameter
(stage) but after sync 'AL' gets written to but GCC thinks the value
is still there and uses it to compute 'EAX' for 'cpuid'.
smm_test can't fully use standard ucall() framework as we need to
write a very simple SMI handler there. Fix the immediate issue by
making RAX input/output operand. While on it, make sync_with_host()
static inline.
Reported-by: Marcelo Bandeira Condotta <mcondotta@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200610164116.770811-1-vkuznets@redhat.com>
Reviewed-by: Jim Mattson <jmattson@google.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Explicitly set the VA width to 48 bits for the x86_64-only PXXV48_4K VM
mode instead of asserting the guest VA width is 48 bits. The fact that
KVM supports 5-level paging is irrelevant unless the selftests opt-in to
5-level paging by setting CR4.LA57 for the guest. The overzealous
assert prevents running the selftests on a kernel with 5-level paging
enabled.
Incorporate LA57 into the assert instead of removing the assert entirely
as a sanity check of KVM's CPUID output.
Fixes: 567a9f1e9d ("KVM: selftests: Introduce VM_MODE_PXXV48_4K")
Reported-by: Sergio Perez Gonzalez <sergio.perez.gonzalez@intel.com>
Cc: Adriana Cervantes Jimenez <adriana.cervantes.jimenez@intel.com>
Cc: Peter Xu <peterx@redhat.com>
Signed-off-by: Sean Christopherson <sean.j.christopherson@intel.com>
Message-Id: <20200528021530.28091-1-sean.j.christopherson@intel.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Marcelo reports that kvm selftests fail to build with
"make ARCH=x86_64":
gcc -Wall -Wstrict-prototypes -Wuninitialized -O2 -g -std=gnu99
-fno-stack-protector -fno-PIE -I../../../../tools/include
-I../../../../tools/arch/x86_64/include -I../../../../usr/include/
-Iinclude -Ilib -Iinclude/x86_64 -I.. -c lib/kvm_util.c
-o /var/tmp/20200604202744-bin/lib/kvm_util.o
In file included from lib/kvm_util.c:11:
include/x86_64/processor.h:14:10: fatal error: asm/msr-index.h: No such
file or directory
#include <asm/msr-index.h>
^~~~~~~~~~~~~~~~~
compilation terminated.
"make ARCH=x86", however, works. The problem is that arch specific headers
for x86_64 live in 'tools/arch/x86/include', not in
'tools/arch/x86_64/include'.
Fixes: 66d69e081b ("selftests: fix kvm relocatable native/cross builds and installs")
Reported-by: Marcelo Bandeira Condotta <mcondotta@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200605142028.550068-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
vmx_tsc_adjust_test fails with:
IA32_TSC_ADJUST is -4294969448 (-1 * TSC_ADJUST_VALUE + -2152).
IA32_TSC_ADJUST is -4294969448 (-1 * TSC_ADJUST_VALUE + -2152).
IA32_TSC_ADJUST is 281470681738540 (65534 * TSC_ADJUST_VALUE + 4294962476).
==== Test Assertion Failure ====
x86_64/vmx_tsc_adjust_test.c:153: false
pid=19738 tid=19738 - Interrupted system call
1 0x0000000000401192: main at vmx_tsc_adjust_test.c:153
2 0x00007fe1ef8583d4: ?? ??:0
3 0x0000000000401201: _start at ??:?
Failed guest assert: (adjust <= max)
The problem is that is 'tsc_val' should be u64, not u32 or the reading
gets truncated.
Fixes: 8d7fbf01f9 ("KVM: selftests: VMX preemption timer migration test")
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Message-Id: <20200601154726.261868-1-vkuznets@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The test is similar to the existing one for VMX, but simpler because we
don't have to test shadow VMCS or vmptrld/vmptrst/vmclear.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When KVM_EXIT_DEBUG is raised for the disabled-breakpoints case (DR7.GD),
DR6 was incorrectly copied from the value in the VM. Instead,
DR6.BD should be set in order to catch this case.
On AMD this does not need any special code because the processor triggers
a #DB exception that is intercepted. However, the testcase would fail
without the previous patch because both DR6.BS and DR6.BD would be set.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Pull kselftest fixes from Shuah Khan:
"ftrace test fixes and a fix to kvm Makefile for relocatable
native/cross builds and installs"
* tag 'linux-kselftest-5.7-rc5' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
selftests: fix kvm relocatable native/cross builds and installs
selftests/ftrace: Make XFAIL green color
ftrace/selftest: make unresolved cases cause failure if --fail-unresolved set
ftrace/selftests: workaround cgroup RT scheduling issues
I got this error when building kvm selftests:
/usr/bin/ld: /home/xz/git/linux/tools/testing/selftests/kvm/libkvm.a(vmx.o):/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:222: multiple definition of `current_evmcs'; /tmp/cco1G48P.o:/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:222: first defined here
/usr/bin/ld: /home/xz/git/linux/tools/testing/selftests/kvm/libkvm.a(vmx.o):/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:223: multiple definition of `current_vp_assist'; /tmp/cco1G48P.o:/home/xz/git/linux/tools/testing/selftests/kvm/include/evmcs.h:223: first defined here
I think it's because evmcs.h is included both in a test file and a lib file so
the structs have multiple declarations when linking. After all it's not a good
habit to declare structs in the header files.
Cc: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <20200504220607.99627-1-peterx@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
kvm test Makefile doesn't fully support cross-builds and installs.
UNAME_M = $(shell uname -m) variable is used to define the target
programs and libraries to be built from arch specific sources in
sub-directories.
For cross-builds to work, UNAME_M has to map to ARCH and arch specific
directories and targets in this Makefile.
UNAME_M variable to used to run the compiles pointing to the right arch
directories and build the right targets for these supported architectures.
TEST_GEN_PROGS and LIBKVM are set using UNAME_M variable.
LINUX_TOOL_ARCH_INCLUDE is set using ARCH variable.
x86_64 targets are named to include x86_64 as a suffix and directories
for includes are in x86_64 sub-directory. s390x and aarch64 follow the
same convention. "uname -m" doesn't result in the correct mapping for
s390x and aarch64. Fix it to set UNAME_M correctly for s390x and aarch64
cross-builds.
In addition, Makefile doesn't create arch sub-directories in the case of
relocatable builds and test programs under s390x and x86_64 directories
fail to build. This is a problem for native and cross-builds. Fix it to
create all necessary directories keying off of TEST_GEN_PROGS.
The following use-cases work with this change:
Native x86_64:
make O=/tmp/kselftest -C tools/testing/selftests TARGETS=kvm install \
INSTALL_PATH=$HOME/x86_64
arm64 cross-build:
make O=$HOME/arm64_build/ ARCH=arm64 HOSTCC=gcc \
CROSS_COMPILE=aarch64-linux-gnu- defconfig
make O=$HOME/arm64_build/ ARCH=arm64 HOSTCC=gcc \
CROSS_COMPILE=aarch64-linux-gnu- all
make kselftest-install TARGETS=kvm O=$HOME/arm64_build ARCH=arm64 \
HOSTCC=gcc CROSS_COMPILE=aarch64-linux-gnu-
s390x cross-build:
make O=$HOME/s390x_build/ ARCH=s390 HOSTCC=gcc \
CROSS_COMPILE=s390x-linux-gnu- defconfig
make O=$HOME/s390x_build/ ARCH=s390 HOSTCC=gcc \
CROSS_COMPILE=s390x-linux-gnu- all
make kselftest-install TARGETS=kvm O=$HOME/s390x_build/ ARCH=s390 \
HOSTCC=gcc CROSS_COMPILE=s390x-linux-gnu- all
No regressions in the following use-cases:
make -C tools/testing/selftests TARGETS=kvm
make kselftest-all TARGETS=kvm
Signed-off-by: Shuah Khan <skhan@linuxfoundation.org>
On AMD, the state of the VMCB is undefined after a shutdown VMEXIT. KVM
takes a very conservative approach to that and resets the guest altogether
when that happens. This causes the set_memory_region_test to fail
because the RIP is 0xfff0 (the reset vector). Restrict the RIP test
to KVM_EXIT_INTERNAL_ERROR in order to fix this.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
