Merge tag 'kvm-x86-fixes-6.4' of https://github.com/kvm-x86/linux into HEAD

KVM x86 fixes for 6.4

 - Fix a memslot lookup bug in the NX recovery thread that could
   theoretically let userspace bypass the NX hugepage mitigation

 - Fix a s/BLOCKING/PENDING bug in SVM's vNMI support

 - Account exit stats for fastpath VM-Exits that never leave the super
   tight run-loop

 - Fix an out-of-bounds bug in the optimized APIC map code, and add a
   regression test for the race.

6 files changed, 101 insertions, 4 deletions

diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index e542cf285b51..3c300a196bdf 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -229,6 +229,23 @@ static int kvm_recalculate_phys_map(struct kvm_apic_map *new,
 	u32 physical_id;
 
 	/*
+	 * For simplicity, KVM always allocates enough space for all possible
+	 * xAPIC IDs.  Yell, but don't kill the VM, as KVM can continue on
+	 * without the optimized map.
+	 */
+	if (WARN_ON_ONCE(xapic_id > new->max_apic_id))
+		return -EINVAL;
+
+	/*
+	 * Bail if a vCPU was added and/or enabled its APIC between allocating
+	 * the map and doing the actual calculations for the map.  Note, KVM
+	 * hardcodes the x2APIC ID to vcpu_id, i.e. there's no TOCTOU bug if
+	 * the compiler decides to reload x2apic_id after this check.
+	 */
+	if (x2apic_id > new->max_apic_id)
+		return -E2BIG;
+
+	/*
 	 * Deliberately truncate the vCPU ID when detecting a mismatched APIC
 	 * ID to avoid false positives if the vCPU ID, i.e. x2APIC ID, is a
 	 * 32-bit value.  Any unwanted aliasing due to truncation results will
@@ -253,8 +270,7 @@ static int kvm_recalculate_phys_map(struct kvm_apic_map *new,
 	 */
 	if (vcpu->kvm->arch.x2apic_format) {
 		/* See also kvm_apic_match_physical_addr(). */
-		if ((apic_x2apic_mode(apic) || x2apic_id > 0xff) &&
-			x2apic_id <= new->max_apic_id)
+		if (apic_x2apic_mode(apic) || x2apic_id > 0xff)
 			new->phys_map[x2apic_id] = apic;
 
 		if (!apic_x2apic_mode(apic) && !new->phys_map[xapic_id])
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index c8961f45e3b1..6eaa3d6994ae 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -7091,7 +7091,10 @@ static void kvm_recover_nx_huge_pages(struct kvm *kvm)
 		 */
 		slot = NULL;
 		if (atomic_read(&kvm->nr_memslots_dirty_logging)) {
-			slot = gfn_to_memslot(kvm, sp->gfn);
+			struct kvm_memslots *slots;
+
+			slots = kvm_memslots_for_spte_role(kvm, sp->role);
+			slot = __gfn_to_memslot(slots, sp->gfn);
 			WARN_ON_ONCE(!slot);
 		}
 
diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c
index ca32389f3c36..54089f990c8f 100644
--- a/arch/x86/kvm/svm/svm.c
+++ b/arch/x86/kvm/svm/svm.c
@@ -3510,7 +3510,7 @@ static bool svm_is_vnmi_pending(struct kvm_vcpu *vcpu)
 	if (!is_vnmi_enabled(svm))
 		return false;
 
-	return !!(svm->vmcb->control.int_ctl & V_NMI_BLOCKING_MASK);
+	return !!(svm->vmcb->control.int_ctl & V_NMI_PENDING_MASK);
 }
 
 static bool svm_set_vnmi_pending(struct kvm_vcpu *vcpu)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index c0778ca39650..04b57a336b34 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -10758,6 +10758,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
 			exit_fastpath = EXIT_FASTPATH_EXIT_HANDLED;
 			break;
 		}
+
+		/* Note, VM-Exits that go down the "slow" path are accounted below. */
+		++vcpu->stat.exits;
 	}
 
 	/*
diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile
index 7a5ff646e7e7..4761b768b773 100644
--- a/tools/testing/selftests/kvm/Makefile
+++ b/tools/testing/selftests/kvm/Makefile
@@ -116,6 +116,7 @@ TEST_GEN_PROGS_x86_64 += x86_64/sev_migrate_tests
 TEST_GEN_PROGS_x86_64 += x86_64/amx_test
 TEST_GEN_PROGS_x86_64 += x86_64/max_vcpuid_cap_test
 TEST_GEN_PROGS_x86_64 += x86_64/triple_fault_event_test
+TEST_GEN_PROGS_x86_64 += x86_64/recalc_apic_map_test
 TEST_GEN_PROGS_x86_64 += access_tracking_perf_test
 TEST_GEN_PROGS_x86_64 += demand_paging_test
 TEST_GEN_PROGS_x86_64 += dirty_log_test
diff --git a/tools/testing/selftests/kvm/x86_64/recalc_apic_map_test.c b/tools/testing/selftests/kvm/x86_64/recalc_apic_map_test.c
new file mode 100644
index 000000000000..4c416ebe7d66
--- /dev/null
+++ b/tools/testing/selftests/kvm/x86_64/recalc_apic_map_test.c
@@ -0,0 +1,74 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Test edge cases and race conditions in kvm_recalculate_apic_map().
+ */
+
+#include <sys/ioctl.h>
+#include <pthread.h>
+#include <time.h>
+
+#include "processor.h"
+#include "test_util.h"
+#include "kvm_util.h"
+#include "apic.h"
+
+#define TIMEOUT		5	/* seconds */
+
+#define LAPIC_DISABLED	0
+#define LAPIC_X2APIC	(MSR_IA32_APICBASE_ENABLE | X2APIC_ENABLE)
+#define MAX_XAPIC_ID	0xff
+
+static void *race(void *arg)
+{
+	struct kvm_lapic_state lapic = {};
+	struct kvm_vcpu *vcpu = arg;
+
+	while (1) {
+		/* Trigger kvm_recalculate_apic_map(). */
+		vcpu_ioctl(vcpu, KVM_SET_LAPIC, &lapic);
+		pthread_testcancel();
+	}
+
+	return NULL;
+}
+
+int main(void)
+{
+	struct kvm_vcpu *vcpus[KVM_MAX_VCPUS];
+	struct kvm_vcpu *vcpuN;
+	struct kvm_vm *vm;
+	pthread_t thread;
+	time_t t;
+	int i;
+
+	kvm_static_assert(KVM_MAX_VCPUS > MAX_XAPIC_ID);
+
+	/*
+	 * Create the max number of vCPUs supported by selftests so that KVM
+	 * has decent amount of work to do when recalculating the map, i.e. to
+	 * make the problematic window large enough to hit.
+	 */
+	vm = vm_create_with_vcpus(KVM_MAX_VCPUS, NULL, vcpus);
+
+	/*
+	 * Enable x2APIC on all vCPUs so that KVM doesn't bail from the recalc
+	 * due to vCPUs having aliased xAPIC IDs (truncated to 8 bits).
+	 */
+	for (i = 0; i < KVM_MAX_VCPUS; i++)
+		vcpu_set_msr(vcpus[i], MSR_IA32_APICBASE, LAPIC_X2APIC);
+
+	ASSERT_EQ(pthread_create(&thread, NULL, race, vcpus[0]), 0);
+
+	vcpuN = vcpus[KVM_MAX_VCPUS - 1];
+	for (t = time(NULL) + TIMEOUT; time(NULL) < t;) {
+		vcpu_set_msr(vcpuN, MSR_IA32_APICBASE, LAPIC_X2APIC);
+		vcpu_set_msr(vcpuN, MSR_IA32_APICBASE, LAPIC_DISABLED);
+	}
+
+	ASSERT_EQ(pthread_cancel(thread), 0);
+	ASSERT_EQ(pthread_join(thread, NULL), 0);
+
+	kvm_vm_free(vm);
+
+	return 0;
+}