Merge tag 'kvm-x86-apic-6.20' of https://github.com/kvm-x86/linux into HEAD

KVM x86 APIC-ish changes for 6.20

 - Fix a benign bug where KVM could use the wrong memslots (ignored SMM) when
   creating a vCPU-specific mapping of guest memory.

 - Clean up KVM's handling of marking mapped vCPU pages dirty.

 - Drop a pile of *ancient* sanity checks hidden behind in KVM's unused
   ASSERT() macro, most of which could be trivially triggered by the guest
   and/or user, and all of which were useless.

 - Fold "struct dest_map" into its sole user, "struct rtc_status", to make it
   more obvious what the weird parameter is used for, and to allow burying the
   RTC shenanigans behind CONFIG_KVM_IOAPIC=y.

 - Bury all of ioapic.h and KVM_IRQCHIP_KERNEL behind CONFIG_KVM_IOAPIC=y.

 - Add a regression test for recent APICv update fixes.

 - Rework KVM's handling of VMCS updates while L2 is active to temporarily
   switch to vmcs01 instead of deferring the update until the next nested
   VM-Exit.  The deferred updates approach directly contributed to several
   bugs, was proving to be a maintenance burden due to the difficulty in
   auditing the correctness of deferred updates, and was polluting
   "struct nested_vmx" with a growing pile of booleans.

 - Handle "hardware APIC ISR", a.k.a. SVI, updates in kvm_apic_update_apicv()
   to consolidate the updates, and to co-locate SVI updates with the updates
   for KVM's own cache of ISR information.

 - Drop a dead function declaration.

3 files changed, 160 insertions, 0 deletions

diff --git a/tools/testing/selftests/kvm/Makefile.kvm b/tools/testing/selftests/kvm/Makefile.kvm
index c3c464513b4e..0166944fa998 100644
--- a/tools/testing/selftests/kvm/Makefile.kvm
+++ b/tools/testing/selftests/kvm/Makefile.kvm
@@ -118,6 +118,7 @@ TEST_GEN_PROGS_x86 += x86/ucna_injection_test
 TEST_GEN_PROGS_x86 += x86/userspace_io_test
 TEST_GEN_PROGS_x86 += x86/userspace_msr_exit_test
 TEST_GEN_PROGS_x86 += x86/vmx_apic_access_test
+TEST_GEN_PROGS_x86 += x86/vmx_apicv_updates_test
 TEST_GEN_PROGS_x86 += x86/vmx_exception_with_invalid_guest_state
 TEST_GEN_PROGS_x86 += x86/vmx_msrs_test
 TEST_GEN_PROGS_x86 += x86/vmx_invalid_nested_guest_state
diff --git a/tools/testing/selftests/kvm/include/x86/apic.h b/tools/testing/selftests/kvm/include/x86/apic.h
index e9b9aebaac97..5ca6bacbd70e 100644
--- a/tools/testing/selftests/kvm/include/x86/apic.h
+++ b/tools/testing/selftests/kvm/include/x86/apic.h
@@ -34,6 +34,7 @@
 #define	APIC_SPIV	0xF0
 #define		APIC_SPIV_FOCUS_DISABLED	(1 << 9)
 #define		APIC_SPIV_APIC_ENABLED		(1 << 8)
+#define	APIC_ISR	0x100
 #define APIC_IRR	0x200
 #define	APIC_ICR	0x300
 #define	APIC_LVTCMCI	0x2f0
@@ -71,6 +72,9 @@
 #define	APIC_TDCR	0x3E0
 #define	APIC_SELF_IPI	0x3F0
 
+#define APIC_VECTOR_TO_BIT_NUMBER(v) ((unsigned int)(v) % 32)
+#define APIC_VECTOR_TO_REG_OFFSET(v) ((unsigned int)(v) / 32 * 0x10)
+
 void apic_disable(void);
 void xapic_enable(void);
 void x2apic_enable(void);
diff --git a/tools/testing/selftests/kvm/x86/vmx_apicv_updates_test.c b/tools/testing/selftests/kvm/x86/vmx_apicv_updates_test.c
new file mode 100644
index 000000000000..337c53fddeff
--- /dev/null
+++ b/tools/testing/selftests/kvm/x86/vmx_apicv_updates_test.c
@@ -0,0 +1,155 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include "test_util.h"
+#include "kvm_util.h"
+#include "processor.h"
+#include "vmx.h"
+
+#define GOOD_IPI_VECTOR 0xe0
+#define BAD_IPI_VECTOR 0xf0
+
+static volatile int good_ipis_received;
+
+static void good_ipi_handler(struct ex_regs *regs)
+{
+	good_ipis_received++;
+}
+
+static void bad_ipi_handler(struct ex_regs *regs)
+{
+	GUEST_FAIL("Received \"bad\" IPI; ICR MMIO write should have been ignored");
+}
+
+static void l2_guest_code(void)
+{
+	x2apic_enable();
+	vmcall();
+
+	xapic_enable();
+	xapic_write_reg(APIC_ID, 1 << 24);
+	vmcall();
+}
+
+static void l1_guest_code(struct vmx_pages *vmx_pages)
+{
+#define L2_GUEST_STACK_SIZE 64
+	unsigned long l2_guest_stack[L2_GUEST_STACK_SIZE];
+	uint32_t control;
+
+	GUEST_ASSERT(prepare_for_vmx_operation(vmx_pages));
+	GUEST_ASSERT(load_vmcs(vmx_pages));
+
+	/* Prepare the VMCS for L2 execution. */
+	prepare_vmcs(vmx_pages, l2_guest_code, &l2_guest_stack[L2_GUEST_STACK_SIZE]);
+	control = vmreadz(CPU_BASED_VM_EXEC_CONTROL);
+	control |= CPU_BASED_USE_MSR_BITMAPS;
+	vmwrite(CPU_BASED_VM_EXEC_CONTROL, control);
+
+	/* Modify APIC ID to coerce KVM into inhibiting APICv. */
+	xapic_enable();
+	xapic_write_reg(APIC_ID, 1 << 24);
+
+	/*
+	 * Generate+receive an IRQ without doing EOI to get an IRQ set in vISR
+	 * but not SVI.  APICv should be inhibited due to running with a
+	 * modified APIC ID.
+	 */
+	xapic_write_reg(APIC_ICR, APIC_DEST_SELF | APIC_DM_FIXED | GOOD_IPI_VECTOR);
+	GUEST_ASSERT_EQ(xapic_read_reg(APIC_ID), 1 << 24);
+
+	/* Enable IRQs and verify the IRQ was received. */
+	sti_nop();
+	GUEST_ASSERT_EQ(good_ipis_received, 1);
+
+	/*
+	 * Run L2 to switch to x2APIC mode, which in turn will uninhibit APICv,
+	 * as KVM should force the APIC ID back to its default.
+	 */
+	GUEST_ASSERT(!vmlaunch());
+	GUEST_ASSERT(vmreadz(VM_EXIT_REASON) == EXIT_REASON_VMCALL);
+	vmwrite(GUEST_RIP, vmreadz(GUEST_RIP) + vmreadz(VM_EXIT_INSTRUCTION_LEN));
+	GUEST_ASSERT(rdmsr(MSR_IA32_APICBASE) & MSR_IA32_APICBASE_EXTD);
+
+	/*
+	 * Scribble the APIC access page to verify KVM disabled xAPIC
+	 * virtualization in vmcs01, and to verify that KVM flushes L1's TLB
+	 * when L2 switches back to accelerated xAPIC mode.
+	 */
+	xapic_write_reg(APIC_ICR2, 0xdeadbeefu);
+	xapic_write_reg(APIC_ICR, APIC_DEST_SELF | APIC_DM_FIXED | BAD_IPI_VECTOR);
+
+	/*
+	 * Verify the IRQ is still in-service and emit an EOI to verify KVM
+	 * propagates the highest vISR vector to SVI when APICv is activated
+	 * (and does so even if APICv was uninhibited while L2 was active).
+	 */
+	GUEST_ASSERT_EQ(x2apic_read_reg(APIC_ISR + APIC_VECTOR_TO_REG_OFFSET(GOOD_IPI_VECTOR)),
+			BIT(APIC_VECTOR_TO_BIT_NUMBER(GOOD_IPI_VECTOR)));
+	x2apic_write_reg(APIC_EOI, 0);
+	GUEST_ASSERT_EQ(x2apic_read_reg(APIC_ISR + APIC_VECTOR_TO_REG_OFFSET(GOOD_IPI_VECTOR)), 0);
+
+	/*
+	 * Run L2 one more time to switch back to xAPIC mode to verify that KVM
+	 * handles the x2APIC => xAPIC transition and inhibits APICv while L2
+	 * is active.
+	 */
+	GUEST_ASSERT(!vmresume());
+	GUEST_ASSERT(vmreadz(VM_EXIT_REASON) == EXIT_REASON_VMCALL);
+	GUEST_ASSERT(!(rdmsr(MSR_IA32_APICBASE) & MSR_IA32_APICBASE_EXTD));
+
+	xapic_write_reg(APIC_ICR, APIC_DEST_SELF | APIC_DM_FIXED | GOOD_IPI_VECTOR);
+	/* Re-enable IRQs, as VM-Exit clears RFLAGS.IF. */
+	sti_nop();
+	GUEST_ASSERT_EQ(good_ipis_received, 2);
+
+	GUEST_ASSERT_EQ(xapic_read_reg(APIC_ISR + APIC_VECTOR_TO_REG_OFFSET(GOOD_IPI_VECTOR)),
+			BIT(APIC_VECTOR_TO_BIT_NUMBER(GOOD_IPI_VECTOR)));
+	xapic_write_reg(APIC_EOI, 0);
+	GUEST_ASSERT_EQ(xapic_read_reg(APIC_ISR + APIC_VECTOR_TO_REG_OFFSET(GOOD_IPI_VECTOR)), 0);
+	GUEST_DONE();
+}
+
+int main(int argc, char *argv[])
+{
+	vm_vaddr_t vmx_pages_gva;
+	struct vmx_pages *vmx;
+	struct kvm_vcpu *vcpu;
+	struct kvm_vm *vm;
+	struct ucall uc;
+
+	TEST_REQUIRE(kvm_cpu_has(X86_FEATURE_VMX));
+
+	vm = vm_create_with_one_vcpu(&vcpu, l1_guest_code);
+
+	vmx = vcpu_alloc_vmx(vm, &vmx_pages_gva);
+	prepare_virtualize_apic_accesses(vmx, vm);
+	vcpu_args_set(vcpu, 1, vmx_pages_gva);
+
+	virt_pg_map(vm, APIC_DEFAULT_GPA, APIC_DEFAULT_GPA);
+	vm_install_exception_handler(vm, BAD_IPI_VECTOR, bad_ipi_handler);
+	vm_install_exception_handler(vm, GOOD_IPI_VECTOR, good_ipi_handler);
+
+	vcpu_run(vcpu);
+	TEST_ASSERT_KVM_EXIT_REASON(vcpu, KVM_EXIT_IO);
+
+	switch (get_ucall(vcpu, &uc)) {
+	case UCALL_ABORT:
+		REPORT_GUEST_ASSERT(uc);
+		/* NOT REACHED */
+	case UCALL_DONE:
+		break;
+	default:
+		TEST_FAIL("Unexpected ucall %lu", uc.cmd);
+	}
+
+	/*
+	 * Verify at least two IRQs were injected.  Unfortunately, KVM counts
+	 * re-injected IRQs (e.g. if delivering the IRQ hits an EPT violation),
+	 * so being more precise isn't possible given the current stats.
+	 */
+	TEST_ASSERT(vcpu_get_stat(vcpu, irq_injections) >= 2,
+		    "Wanted at least 2 IRQ injections, got %lu\n",
+		    vcpu_get_stat(vcpu, irq_injections));
+
+	kvm_vm_free(vm);
+	return 0;
+}