linux-kernel - Re: [PATCH v4 3/3] selftests: KVM: SVM: Add vmcall test

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 7 Feb 2020 15:06:54 +0100
From:   Auger Eric <eric.auger@...hat.com>
To:     Krish Sadhukhan <krish.sadhukhan@...cle.com>,
        eric.auger.pro@...il.com, linux-kernel@...r.kernel.org,
        kvm@...r.kernel.org, pbonzini@...hat.com, vkuznets@...hat.com
Cc:     thuth@...hat.com, drjones@...hat.com, wei.huang2@....com
Subject: Re: [PATCH v4 3/3] selftests: KVM: SVM: Add vmcall test

Hi Krish,

On 2/6/20 11:46 PM, Krish Sadhukhan wrote:
> 
> 
> On 02/06/2020 02:47 AM, Eric Auger wrote:
>> L2 guest calls vmcall and L1 checks the exit status does
>> correspond.
>>
>> Signed-off-by: Eric Auger <eric.auger@...hat.com>
>> Reviewed-by: Vitaly Kuznetsov <vkuznets@...hat.com>
>> Reviewed-by: Miaohe Lin <linmiaohe@...wei.com>
>>
>> ---
>>
>> v3 -> v4:
>> - remove useless includes
>> - collected Lin's R-b
>>
>> v2 -> v3:
>> - remove useless comment and add Vitaly's R-b
>> ---
>>   tools/testing/selftests/kvm/Makefile          |  1 +
>>   .../selftests/kvm/x86_64/svm_vmcall_test.c    | 79 +++++++++++++++++++
>>   2 files changed, 80 insertions(+)
>>   create mode 100644 tools/testing/selftests/kvm/x86_64/svm_vmcall_test.c
>>
>> diff --git a/tools/testing/selftests/kvm/Makefile
>> b/tools/testing/selftests/kvm/Makefile
>> index 2e770f554cae..b529d3b42c02 100644
>> --- a/tools/testing/selftests/kvm/Makefile
>> +++ b/tools/testing/selftests/kvm/Makefile
>> @@ -26,6 +26,7 @@ TEST_GEN_PROGS_x86_64 += x86_64/vmx_dirty_log_test
>>   TEST_GEN_PROGS_x86_64 += x86_64/vmx_set_nested_state_test
>>   TEST_GEN_PROGS_x86_64 += x86_64/vmx_tsc_adjust_test
>>   TEST_GEN_PROGS_x86_64 += x86_64/xss_msr_test
>> +TEST_GEN_PROGS_x86_64 += x86_64/svm_vmcall_test
>>   TEST_GEN_PROGS_x86_64 += clear_dirty_log_test
>>   TEST_GEN_PROGS_x86_64 += dirty_log_test
>>   TEST_GEN_PROGS_x86_64 += kvm_create_max_vcpus
>> diff --git a/tools/testing/selftests/kvm/x86_64/svm_vmcall_test.c
>> b/tools/testing/selftests/kvm/x86_64/svm_vmcall_test.c
>> new file mode 100644
>> index 000000000000..6d3565aab94e
>> --- /dev/null
>> +++ b/tools/testing/selftests/kvm/x86_64/svm_vmcall_test.c
>> @@ -0,0 +1,79 @@
>> +// SPDX-License-Identifier: GPL-2.0-only
>> +/*
>> + * svm_vmcall_test
>> + *
>> + * Copyright (C) 2020, Red Hat, Inc.
>> + *
>> + * Nested SVM testing: VMCALL
>> + */
>> +
>> +#include "test_util.h"
>> +#include "kvm_util.h"
>> +#include "processor.h"
>> +#include "svm_util.h"
>> +
>> +#define VCPU_ID        5
>> +
>> +static struct kvm_vm *vm;
>> +
>> +static inline void l2_vmcall(struct svm_test_data *svm)
>> +{
>> +    __asm__ __volatile__("vmcall");
> Is it possible to re-use the existing vmcall() function ?
well the function is declared in vmx header. Also vmx_tsc_adjust_test
does not use it for instance. For this test the above is simple and does
the job.

> Also, we should probably re-name the function to 'l2_guest_code' which
> is used in the existing code and also it matches with 'l1_guest_code'
> naming.
OK
>> +}
>> +
>> +static void l1_guest_code(struct svm_test_data *svm)
>> +{
>> +    #define L2_GUEST_STACK_SIZE 64
>> +    unsigned long l2_guest_stack[L2_GUEST_STACK_SIZE];
>> +    struct vmcb *vmcb = svm->vmcb;
>> +
>> +    /* Prepare for L2 execution. */
>> +    generic_svm_setup(svm, l2_vmcall,
>> +              &l2_guest_stack[L2_GUEST_STACK_SIZE]);
>> +
>> +    run_guest(vmcb, svm->vmcb_gpa);
>> +
>> +    GUEST_ASSERT(vmcb->control.exit_code == SVM_EXIT_VMMCALL);
>> +    GUEST_DONE();
>> +}
>> +
>> +int main(int argc, char *argv[])
>> +{
>> +    vm_vaddr_t svm_gva;
>> +
>> +    nested_svm_check_supported();
>> +
>> +    vm = vm_create_default(VCPU_ID, 0, (void *) l1_guest_code);
>> +    vcpu_set_cpuid(vm, VCPU_ID, kvm_get_supported_cpuid());
>> +
>> +    vcpu_alloc_svm(vm, &svm_gva);
>> +    vcpu_args_set(vm, VCPU_ID, 1, svm_gva);
>> +
>> +    for (;;) {
>> +        volatile struct kvm_run *run = vcpu_state(vm, VCPU_ID);
>> +        struct ucall uc;
>> +
>> +        vcpu_run(vm, VCPU_ID);
>> +        TEST_ASSERT(run->exit_reason == KVM_EXIT_IO,
>> +                "Got exit_reason other than KVM_EXIT_IO: %u (%s)\n",
>> +                run->exit_reason,
>> +                exit_reason_str(run->exit_reason));
>> +
>> +        switch (get_ucall(vm, VCPU_ID, &uc)) {
>> +        case UCALL_ABORT:
>> +            TEST_ASSERT(false, "%s",
>> +                    (const char *)uc.args[0]);
>> +            /* NOT REACHED */
>> +        case UCALL_SYNC:
>> +            break;
>> +        case UCALL_DONE:
>> +            goto done;
>> +        default:
>> +            TEST_ASSERT(false,
>> +                    "Unknown ucall 0x%x.", uc.cmd);
>> +        }
>> +    }
>> +done:
>> +    kvm_vm_free(vm);
>> +    return 0;
>> +}
> 
Thanks

Eric