public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Like Xu <like.xu.linux@gmail.com>
To: Sean Christopherson <seanjc@google.com>
Cc: kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	Xiaoyao Li <xiaoyao.li@intel.com>,
	Paolo Bonzini <pbonzini@redhat.com>
Subject: Re: [PATCH v2 08/21] KVM: selftests: Split PMU caps sub-tests to avoid writing MSR after KVM_RUN
Date: Tue, 14 Feb 2023 15:27:30 +0800	[thread overview]
Message-ID: <85b3d348-2e4a-43aa-0131-27e9fc375cf9@gmail.com> (raw)
In-Reply-To: <20230210003148.2646712-9-seanjc@google.com>

Nit, could this patch be applied before the relevant KVM modifications [1] so that
a CI bot doesn't generate an error report before this selftests patch is applied ?

[1] KVM: x86: Disallow writes to immutable feature MSRs after KVM_RUN

On 10/2/2023 8:31 am, Sean Christopherson wrote:
> Split the PERF_CAPABILITIES subtests into two parts so that the LBR format
> testcases don't execute after KVM_RUN.  Now that KVM disallows changing
> PERF_CAPABILITIES after KVM_RUN (same as guest CPUID), attempting to set
> the MSR after KVM_RUN will yield false positives and/or false negatives
> depending on what the test is trying to do.
> 
> Land the LBR format test in a more generic "immutable features" test in
> anticipation of expanding its scope to other immutable features.
> 
> Signed-off-by: Sean Christopherson <seanjc@google.com>
> ---
>   .../selftests/kvm/x86_64/vmx_pmu_caps_test.c  | 51 +++++++++++--------
>   1 file changed, 31 insertions(+), 20 deletions(-)
> 
> diff --git a/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c b/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c
> index c280ba1e6572..ac08c0fdd84d 100644
> --- a/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c
> +++ b/tools/testing/selftests/kvm/x86_64/vmx_pmu_caps_test.c
> @@ -41,24 +41,10 @@ static void guest_code(void)
>   	wrmsr(MSR_IA32_PERF_CAPABILITIES, PMU_CAP_LBR_FMT);
>   }
>   
> -int main(int argc, char *argv[])
> +static void test_fungible_perf_capabilities(union perf_capabilities host_cap)
>   {
> -	struct kvm_vm *vm;
>   	struct kvm_vcpu *vcpu;
> -	int ret;
> -	union perf_capabilities host_cap;
> -	uint64_t val;
> -
> -	host_cap.capabilities = kvm_get_feature_msr(MSR_IA32_PERF_CAPABILITIES);
> -	host_cap.capabilities &= (PMU_CAP_FW_WRITES | PMU_CAP_LBR_FMT);
> -
> -	/* Create VM */
> -	vm = vm_create_with_one_vcpu(&vcpu, guest_code);
> -
> -	TEST_REQUIRE(kvm_cpu_has(X86_FEATURE_PDCM));
> -
> -	TEST_REQUIRE(kvm_cpu_has_p(X86_PROPERTY_PMU_VERSION));
> -	TEST_REQUIRE(kvm_cpu_property(X86_PROPERTY_PMU_VERSION) > 0);
> +	struct kvm_vm *vm = vm_create_with_one_vcpu(&vcpu, guest_code);
>   
>   	/* testcase 1, set capabilities when we have PDCM bit */
>   	vcpu_set_msr(vcpu, MSR_IA32_PERF_CAPABILITIES, PMU_CAP_FW_WRITES);
> @@ -70,7 +56,16 @@ int main(int argc, char *argv[])
>   	vcpu_run(vcpu);
>   	ASSERT_EQ(vcpu_get_msr(vcpu, MSR_IA32_PERF_CAPABILITIES), PMU_CAP_FW_WRITES);
>   
> -	/* testcase 2, check valid LBR formats are accepted */
> +	kvm_vm_free(vm);
> +}
> +
> +static void test_immutable_perf_capabilities(union perf_capabilities host_cap)
> +{
> +	struct kvm_vcpu *vcpu;
> +	struct kvm_vm *vm = vm_create_with_one_vcpu(&vcpu, NULL);
> +	uint64_t val;
> +	int ret;
> +
>   	vcpu_set_msr(vcpu, MSR_IA32_PERF_CAPABILITIES, 0);
>   	ASSERT_EQ(vcpu_get_msr(vcpu, MSR_IA32_PERF_CAPABILITIES), 0);
>   
> @@ -78,8 +73,8 @@ int main(int argc, char *argv[])
>   	ASSERT_EQ(vcpu_get_msr(vcpu, MSR_IA32_PERF_CAPABILITIES), (u64)host_cap.lbr_format);
>   
>   	/*
> -	 * Testcase 3, check that an "invalid" LBR format is rejected.  Only an
> -	 * exact match of the host's format (and 0/disabled) is allowed.
> +	 * KVM only supports the host's native LBR format, as well as '0' (to
> +	 * disable LBR support).  Verify KVM rejects all other LBR formats.
>   	 */
>   	for (val = 1; val <= PMU_CAP_LBR_FMT; val++) {
>   		if (val == (host_cap.capabilities & PMU_CAP_LBR_FMT))
> @@ -88,7 +83,23 @@ int main(int argc, char *argv[])
>   		ret = _vcpu_set_msr(vcpu, MSR_IA32_PERF_CAPABILITIES, val);
>   		TEST_ASSERT(!ret, "Bad LBR FMT = 0x%lx didn't fail", val);
>   	}
> +	kvm_vm_free(vm);
> +}
> +
> +int main(int argc, char *argv[])
> +{
> +	union perf_capabilities host_cap;
> +
> +	TEST_REQUIRE(kvm_cpu_has(X86_FEATURE_PDCM));
> +
> +	TEST_REQUIRE(kvm_cpu_has_p(X86_PROPERTY_PMU_VERSION));
> +	TEST_REQUIRE(kvm_cpu_property(X86_PROPERTY_PMU_VERSION) > 0);
> +
> +	host_cap.capabilities = kvm_get_feature_msr(MSR_IA32_PERF_CAPABILITIES);
> +	host_cap.capabilities &= (PMU_CAP_FW_WRITES | PMU_CAP_LBR_FMT);
> +
> +	test_fungible_perf_capabilities(host_cap);
> +	test_immutable_perf_capabilities(host_cap);
>   
>   	printf("Completed perf capability tests.\n");
> -	kvm_vm_free(vm);
>   }

  reply	other threads:[~2023-02-14  7:27 UTC|newest]

Thread overview: 39+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-02-10  0:31 [PATCH v2 00/21] KVM: x86: Disallow writes to feature MSRs post-KVM_RUN Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 01/21] KVM: x86: Rename kvm_init_msr_list() to clarify it inits multiple lists Sean Christopherson
2023-02-14  6:34   ` Xiaoyao Li
2023-02-10  0:31 ` [PATCH v2 02/21] KVM: x86: Add a helper to query whether or not a vCPU has ever run Sean Christopherson
2023-02-14  6:35   ` Xiaoyao Li
2023-02-10  0:31 ` [PATCH v2 03/21] KVM: x86: Add macros to track first...last VMX feature MSRs Sean Christopherson
2023-02-14 10:48   ` Like Xu
2023-02-14 18:51     ` Sean Christopherson
2023-02-15  3:20   ` Xiaoyao Li
2023-02-10  0:31 ` [PATCH v2 04/21] KVM: x86: Generate set of VMX feature MSRs using first/last definitions Sean Christopherson
2023-02-14  6:36   ` Xiaoyao Li
2023-02-10  0:31 ` [PATCH v2 05/21] KVM: x86: Disallow writes to immutable feature MSRs after KVM_RUN Sean Christopherson
2023-02-10 13:01   ` Yu Zhang
2023-02-10 16:31     ` Sean Christopherson
2023-02-14  6:39   ` Xiaoyao Li
2023-02-14 11:39   ` Like Xu
2023-02-15 22:36     ` Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 06/21] KVM: x86/pmu: WARN and bug the VM if PMU is refreshed after vCPU has run Sean Christopherson
2023-02-15 11:53   ` Like Xu
2023-02-15 15:20     ` Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 07/21] KVM: x86/pmu: Zero out LBR capabilities during PMU refresh Sean Christopherson
2023-02-15 12:00   ` Like Xu
2023-02-15 15:17     ` Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 08/21] KVM: selftests: Split PMU caps sub-tests to avoid writing MSR after KVM_RUN Sean Christopherson
2023-02-14  7:27   ` Like Xu [this message]
2023-02-14 17:42     ` Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 09/21] KVM: selftests: Move 0/initial value PERF_CAPS checks to dedicated sub-test Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 10/21] KVM: selftests: Assert that full-width PMC writes are supported if PDCM=1 Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 11/21] KVM: selftests: Print out failing MSR and value in vcpu_set_msr() Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 12/21] KVM: selftests: Verify KVM preserves userspace writes to "durable" MSRs Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 13/21] KVM: selftests: Drop now-redundant checks on PERF_CAPABILITIES writes Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 14/21] KVM: selftests: Test all fungible features in PERF_CAPABILITIES Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 15/21] KVM: selftests: Test all immutable non-format bits " Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 16/21] KVM: selftests: Expand negative testing of guest writes to PERF_CAPABILITIES Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 17/21] KVM: selftests: Test post-KVM_RUN " Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 18/21] KVM: selftests: Drop "all done!" printf() from PERF_CAPABILITIES test Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 19/21] KVM: selftests: Refactor LBR_FMT test to avoid use of separate macro Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 20/21] KVM: selftests: Add negative testcase for PEBS format in PERF_CAPABILITIES Sean Christopherson
2023-02-10  0:31 ` [PATCH v2 21/21] KVM: selftests: Verify LBRs are disabled if vPMU is disabled Sean Christopherson

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=85b3d348-2e4a-43aa-0131-27e9fc375cf9@gmail.com \
    --to=like.xu.linux@gmail.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=seanjc@google.com \
    --cc=xiaoyao.li@intel.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox