mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Yosry Ahmed <yosry.ahmed@linux.dev>
To: Sean Christopherson <seanjc@google.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
	Jim Mattson <jmattson@google.com>,
	kvm@vger.kernel.org, linux-kernel@vger.kernel.org,
	Yosry Ahmed <yosryahmed@google.com>
Subject: [kvm-unit-tests 6/7] x86/svm: Generalize and improve selective CR0 write intercept test
Date: Fri, 24 Oct 2025 19:49:24 +0000	[thread overview]
Message-ID: <20251024194925.3201933-7-yosry.ahmed@linux.dev> (raw)
In-Reply-To: <20251024194925.3201933-1-yosry.ahmed@linux.dev>

From: Yosry Ahmed <yosryahmed@google.com>

In preparation for adding more test cases, make the test easier to
extend. Create a generic helper that sets an arbitrary bit in CR0,
optionally using FEP. The helper also stores the value to be written in
test->scratch, making it possible to double check if the write was
actually executed or not.

Use report_svm_guest() instead of report_fail() + exit().

Make test_sel_cr0_write_intercept() use the generic helper, and add
another test case that sets FEP to exercise the interception path in the
emulator.

Finally, in check_sel_cr0_intercept() also check that the write was not
executed by comparing CR0 value in the VMCB12 with the
value-to-be-written stored in test->scratch.

Signed-off-by: Yosry Ahmed <yosryahmed@google.com>
---
 x86/svm_tests.c | 38 +++++++++++++++++++++++++-------------
 1 file changed, 25 insertions(+), 13 deletions(-)

diff --git a/x86/svm_tests.c b/x86/svm_tests.c
index 61ab63db..71afb38a 100644
--- a/x86/svm_tests.c
+++ b/x86/svm_tests.c
@@ -133,27 +133,36 @@ static void prepare_sel_cr0_intercept(struct svm_test *test)
 	vmcb->control.intercept |= (1ULL << INTERCEPT_SELECTIVE_CR0);
 }
 
-static void test_sel_cr0_write_intercept(struct svm_test *test)
+static void __test_cr0_write_bit(struct svm_test *test, unsigned long bit,
+				 bool intercept, bool fep)
 {
 	unsigned long cr0;
 
-	/* read cr0, set CD, and write back */
-	cr0  = read_cr0();
-	cr0 |= X86_CR0_CD;
-	write_cr0(cr0);
+	cr0 = read_cr0();
+	cr0 |= bit;
+	test->scratch = cr0;
 
-	/*
-	 * If we are here the test failed, not sure what to do now because we
-	 * are not in guest-mode anymore so we can't trigger an intercept.
-	 * Trigger a tripple-fault for now.
-	 */
-	report_fail("sel_cr0 test. Can not recover from this - exiting");
-	exit(report_summary());
+	asm_conditional_fep_safe(fep, "mov %0,%%cr0", "r"(cr0));
+
+	/* This code should be unreachable when an intercept is expected */
+	report_svm_guest(!intercept, test, "Expected intercept on CR0 write");
+}
+
+/* MOV-to-CR0 updating CR0.CD is intercepted by the selective intercept */
+static void test_sel_cr0_write_intercept(struct svm_test *test)
+{
+	__test_cr0_write_bit(test, X86_CR0_CD, true, false);
+}
+
+static void test_sel_cr0_write_intercept_emul(struct svm_test *test)
+{
+	__test_cr0_write_bit(test, X86_CR0_CD, true, true);
 }
 
 static bool check_sel_cr0_intercept(struct svm_test *test)
 {
-	return vmcb->control.exit_code == SVM_EXIT_CR0_SEL_WRITE;
+	return vmcb->control.exit_code == SVM_EXIT_CR0_SEL_WRITE &&
+		vmcb->save.cr0 != test->scratch;
 }
 
 static void prepare_cr3_intercept(struct svm_test *test)
@@ -3461,6 +3470,9 @@ struct svm_test svm_tests[] = {
 	{ "sel cr0 write intercept", default_supported,
 	  prepare_sel_cr0_intercept, default_prepare_gif_clear,
 	  test_sel_cr0_write_intercept, default_finished, check_sel_cr0_intercept},
+	{ "sel cr0 write intercept emulate", fep_supported,
+	  prepare_sel_cr0_intercept, default_prepare_gif_clear,
+	  test_sel_cr0_write_intercept_emul, default_finished, check_sel_cr0_intercept},
 	{ "cr3 read intercept", default_supported,
 	  prepare_cr3_intercept, default_prepare_gif_clear,
 	  test_cr3_intercept, default_finished, check_cr3_intercept },
-- 
2.51.1.821.gb6fe4d2222-goog


  parent reply	other threads:[~2025-10-24 19:49 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-24 19:49 [kvm-unit-tests 0/7] More tests for selective CR0 intercept Yosry Ahmed
2025-10-24 19:49 ` [kvm-unit-tests 1/7] x86/svm: Cleanup selective cr0 write intercept test Yosry Ahmed
2025-10-24 19:49 ` [kvm-unit-tests 2/7] x86/svm: Move CR0 selective write intercept test near CR3 intercept Yosry Ahmed
2025-10-24 19:49 ` [kvm-unit-tests 3/7] x86/svm: Add FEP helpers for SVM tests Yosry Ahmed
2025-10-24 19:49 ` [kvm-unit-tests 4/7] x86/svm: Report unsupported " Yosry Ahmed
2025-10-24 19:49 ` [kvm-unit-tests 5/7] x86/svm: Move report_svm_guest() to the top of svm_tests.c Yosry Ahmed
2025-10-24 19:49 ` Yosry Ahmed [this message]
2025-10-24 19:49 ` [kvm-unit-tests 7/7] x86/svm: Add more selective CR0 write and LMSW test cases Yosry Ahmed
2025-10-24 20:02 ` [kvm-unit-tests 0/7] More tests for selective CR0 intercept Yosry Ahmed

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=20251024194925.3201933-7-yosry.ahmed@linux.dev \
    --to=yosry.ahmed@linux.dev \
    --cc=jmattson@google.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pbonzini@redhat.com \
    --cc=seanjc@google.com \
    --cc=yosryahmed@google.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

all inboxes | Powered by JetHome®