linux-kernel - [PATCH v2 08/10] KVM: selftests: Test x2APIC ICR reserved bits

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]

Message-ID: <20240719235107.3023592-9-seanjc@google.com>
Date: Fri, 19 Jul 2024 16:51:05 -0700
From: Sean Christopherson <seanjc@...gle.com>
To: Sean Christopherson <seanjc@...gle.com>, Paolo Bonzini <pbonzini@...hat.com>
Cc: kvm@...r.kernel.org, linux-kernel@...r.kernel.org, 
	Maxim Levitsky <mlevitsk@...hat.com>, Suravee Suthikulpanit <suravee.suthikulpanit@....com>
Subject: [PATCH v2 08/10] KVM: selftests: Test x2APIC ICR reserved bits

Actually test x2APIC ICR reserved bits instead of deliberately skipping
them.  The behavior that is observed when IPI virtualization is enabled is
the architecturally correct behavior, KVM is the one who was wrong, i.e.
KVM was missing reserved bit checks.

Fixes: 4b88b1a518b3 ("KVM: selftests: Enhance handling WRMSR ICR register in x2APIC mode")
Signed-off-by: Sean Christopherson <seanjc@...gle.com>
---
 .../selftests/kvm/x86_64/xapic_state_test.c   | 23 ++++++++-----------
 1 file changed, 10 insertions(+), 13 deletions(-)

diff --git a/tools/testing/selftests/kvm/x86_64/xapic_state_test.c b/tools/testing/selftests/kvm/x86_64/xapic_state_test.c
index 928d65948c48..d701fe9dd686 100644
--- a/tools/testing/selftests/kvm/x86_64/xapic_state_test.c
+++ b/tools/testing/selftests/kvm/x86_64/xapic_state_test.c
@@ -31,6 +31,10 @@ static void xapic_guest_code(void)
 	}
 }
 
+#define X2APIC_RSVD_BITS_MASK  (GENMASK_ULL(31, 20) | \
+				GENMASK_ULL(17, 16) | \
+				GENMASK_ULL(13, 13))
+
 static void x2apic_guest_code(void)
 {
 	asm volatile("cli");
@@ -41,7 +45,10 @@ static void x2apic_guest_code(void)
 		uint64_t val = x2apic_read_reg(APIC_IRR) |
 			       x2apic_read_reg(APIC_IRR + 0x10) << 32;
 
-		x2apic_write_reg(APIC_ICR, val);
+		if (val & X2APIC_RSVD_BITS_MASK)
+			x2apic_write_reg_fault(APIC_ICR, val);
+		else
+			x2apic_write_reg(APIC_ICR, val);
 		GUEST_SYNC(val);
 	} while (1);
 }
@@ -72,24 +79,14 @@ static void ____test_icr(struct xapic_vcpu *x, uint64_t val)
 	      (u64)(*((u32 *)&xapic.regs[APIC_ICR2])) << 32;
 	if (!x->is_x2apic)
 		val &= (-1u | (0xffull << (32 + 24)));
+	else if (val & X2APIC_RSVD_BITS_MASK)
+		return;
 
 	TEST_ASSERT_EQ(icr, val & ~APIC_ICR_BUSY);
 }
 
-#define X2APIC_RSVED_BITS_MASK  (GENMASK_ULL(31,20) | \
-				 GENMASK_ULL(17,16) | \
-				 GENMASK_ULL(13,13))
-
 static void __test_icr(struct xapic_vcpu *x, uint64_t val)
 {
-	if (x->is_x2apic) {
-		/* Hardware writing vICR register requires reserved bits 31:20,
-		 * 17:16 and 13 kept as zero to avoid #GP exception. Data value
-		 * written to vICR should mask out those bits above.
-		 */
-		val &= ~X2APIC_RSVED_BITS_MASK;
-	}
-
 	/*
 	 * The BUSY bit is reserved on both AMD and Intel, but only AMD treats
 	 * it is as _must_ be zero.  Intel simply ignores the bit.  Don't test
-- 
2.45.2.1089.g2a221341d9-goog