kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Sean Christopherson <seanjc@google.com>
To: Paolo Bonzini <pbonzini@redhat.com>
Cc: kvm@vger.kernel.org, Sean Christopherson <seanjc@google.com>,
	Manali Shukla <manali.shukla@amd.com>
Subject: [kvm-unit-tests PATCH v3 9/9] nVMX: Move #NM test to generic exception test framework
Date: Wed,  5 Oct 2022 23:52:12 +0000	[thread overview]
Message-ID: <20221005235212.57836-10-seanjc@google.com> (raw)
In-Reply-To: <20221005235212.57836-1-seanjc@google.com>

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>
---
 x86/vmx_tests.c | 67 ++++++-------------------------------------------
 1 file changed, 8 insertions(+), 59 deletions(-)

diff --git a/x86/vmx_tests.c b/x86/vmx_tests.c
index 368ad43..2438022 100644
--- a/x86/vmx_tests.c
+++ b/x86/vmx_tests.c
@@ -8359,67 +8359,14 @@ static void vmx_cr4_osxsave_test(void)
 	TEST_ASSERT(this_cpu_has(X86_FEATURE_OSXSAVE));
 }
 
-static void vmx_nm_test_guest(void)
-{
-	write_cr0(read_cr0() | X86_CR0_TS);
-	asm volatile("fnop");
-}
-
-static void check_nm_exit(const char *test)
-{
-	u32 reason = vmcs_read(EXI_REASON);
-	u32 intr_info = vmcs_read(EXI_INTR_INFO);
-	const u32 expected = INTR_INFO_VALID_MASK | INTR_TYPE_HARD_EXCEPTION |
-		NM_VECTOR;
-
-	report(reason == VMX_EXC_NMI && intr_info == expected, "%s", test);
-}
-
 /*
- * This test checks that:
- *
- * (a) If L2 launches with CR0.TS clear, but later sets CR0.TS, then
- *     a subsequent #NM VM-exit is reflected to L1.
- *
- * (b) If L2 launches with CR0.TS clear and CR0.EM set, then a
- *     subsequent #NM VM-exit is reflected to L1.
+ * FNOP with both CR0.TS and CR0.EM clear should not generate #NM, and the L2
+ * guest should exit normally.
  */
-static void vmx_nm_test(void)
+static void vmx_no_nm_test(void)
 {
-	unsigned long cr0 = read_cr0();
-
-	test_set_guest(vmx_nm_test_guest);
-
-	/*
-	 * L1 wants to intercept #NM exceptions encountered in L2.
-	 */
-	vmcs_write(EXC_BITMAP, 1 << NM_VECTOR);
-
-	/*
-	 * Launch L2 with CR0.TS clear, but don't claim host ownership of
-	 * any CR0 bits. L2 will set CR0.TS and then try to execute fnop,
-	 * which will raise #NM. L0 should reflect the #NM VM-exit to L1.
-	 */
-	vmcs_write(CR0_MASK, 0);
-	vmcs_write(GUEST_CR0, cr0 & ~X86_CR0_TS);
-	enter_guest();
-	check_nm_exit("fnop with CR0.TS set in L2 triggers #NM VM-exit to L1");
-
-	/*
-	 * Re-enter L2 at the fnop instruction, with CR0.TS clear but
-	 * CR0.EM set. The fnop will still raise #NM, and L0 should
-	 * reflect the #NM VM-exit to L1.
-	 */
-	vmcs_write(GUEST_CR0, (cr0 & ~X86_CR0_TS) | X86_CR0_EM);
-	enter_guest();
-	check_nm_exit("fnop with CR0.EM set in L2 triggers #NM VM-exit to L1");
-
-	/*
-	 * Re-enter L2 at the fnop instruction, with both CR0.TS and
-	 * CR0.EM clear. There will be no #NM, and the L2 guest should
-	 * exit normally.
-	 */
-	vmcs_write(GUEST_CR0, cr0 & ~(X86_CR0_TS | X86_CR0_EM));
+	test_set_guest(fnop);
+	vmcs_write(GUEST_CR0, read_cr0() & ~(X86_CR0_TS | X86_CR0_EM));
 	enter_guest();
 }
 
@@ -10666,6 +10613,8 @@ struct vmx_exception_test vmx_exception_tests[] = {
 	{ BP_VECTOR, generate_bp },
 	{ AC_VECTOR, vmx_l2_ac_test },
 	{ OF_VECTOR, generate_of },
+	{ NM_VECTOR, generate_cr0_ts_nm },
+	{ NM_VECTOR, generate_cr0_em_nm },
 };
 
 static u8 vmx_exception_test_vector;
@@ -10804,7 +10753,7 @@ struct vmx_test vmx_tests[] = {
 	TEST(vmx_ldtr_test),
 	TEST(vmx_cr_load_test),
 	TEST(vmx_cr4_osxsave_test),
-	TEST(vmx_nm_test),
+	TEST(vmx_no_nm_test),
 	TEST(vmx_db_test),
 	TEST(vmx_nmi_window_test),
 	TEST(vmx_intr_window_test),
-- 
2.38.0.rc1.362.ged0d419d3c-goog


  parent reply	other threads:[~2022-10-05 23:52 UTC|newest]

Thread overview: 11+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-05 23:52 [kvm-unit-tests PATCH v3 0/9] x86: n{VMX,SVM} exception tests Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 1/9] nVMX: Add "nop" after setting EFLAGS.TF to guarantee single-step #DB Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 2/9] x86: Move helpers to generate misc exceptions to processor.h Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 3/9] nVMX: Move #OF test to generic exceptions test Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 4/9] nVMX: Drop one-off INT3=>#BP test Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 5/9] x86: nSVM: Add an exception test framework and tests Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 6/9] x86: nSVM: Move #BP test to exception test framework Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 7/9] x86: nSVM: Move #OF " Sean Christopherson
2022-10-05 23:52 ` [kvm-unit-tests PATCH v3 8/9] x86: nSVM: Move part of #NM " Sean Christopherson
2022-10-05 23:52 ` Sean Christopherson [this message]
2022-10-12  4:17 ` [kvm-unit-tests PATCH v3 0/9] x86: n{VMX,SVM} exception tests Manali Shukla

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=20221005235212.57836-10-seanjc@google.com \
    --to=seanjc@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=manali.shukla@amd.com \
    --cc=pbonzini@redhat.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;
as well as URLs for NNTP newsgroup(s).