Merge tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm

Pull kvm fixes from Paolo Bonzini:
 "Bugfixes (arm and x86) and cleanups"

* tag 'for-linus' of git://git.kernel.org/pub/scm/virt/kvm/kvm:
  selftests: kvm: Adding config fragments
  KVM: selftests: Update gitignore file for latest changes
  kvm: remove unnecessary PageReserved check
  KVM: arm/arm64: vgic: Reevaluate level sensitive interrupts on enable
  KVM: arm: Don't write junk to CP15 registers on reset
  KVM: arm64: Don't write junk to sysregs on reset
  KVM: arm/arm64: Sync ICH_VMCR_EL2 back when about to block
  x86: kvm: remove useless calls to kvm_para_available
  KVM: no need to check return value of debugfs_create functions
  KVM: remove kvm_arch_has_vcpu_debugfs()
  KVM: Fix leak vCPU's VMCS value into other pCPU
  KVM: Check preempted_in_kernel for involuntary preemption
  KVM: LAPIC: Don't need to wakeup vCPU twice afer timer fire
  arm64: KVM: hyp: debug-sr: Mark expected switch fall-through
  KVM: arm64: Update kvm_arm_exception_class and esr_class_str for new EC
  KVM: arm: vgic-v3: Mark expected switch fall-through
  arm64: KVM: regmap: Fix unexpected switch fall-through
  KVM: arm/arm64: Introduce kvm_pmu_vcpu_init() to setup PMU counter index

arch/arm/kvm/coproc.c

@@ -651,13 +651,22 @@ int kvm_handle_cp14_64(struct kvm_vcpu *vcpu, struct kvm_run *run)
 }
 
 static void reset_coproc_regs(struct kvm_vcpu *vcpu,
-			      const struct coproc_reg *table, size_t num)
+			      const struct coproc_reg *table, size_t num,
+			      unsigned long *bmap)
 {
 	unsigned long i;
 
 	for (i = 0; i < num; i++)
-		if (table[i].reset)
+		if (table[i].reset) {
+			int reg = table[i].reg;
+
 			table[i].reset(vcpu, &table[i]);
+			if (reg > 0 && reg < NR_CP15_REGS) {
+				set_bit(reg, bmap);
+				if (table[i].is_64bit)
+					set_bit(reg + 1, bmap);
+			}
+		}
 }
 
 static struct coproc_params decode_32bit_hsr(struct kvm_vcpu *vcpu)
@@ -1432,17 +1441,15 @@ void kvm_reset_coprocs(struct kvm_vcpu *vcpu)
 {
 	size_t num;
 	const struct coproc_reg *table;
-
-	/* Catch someone adding a register without putting in reset entry. */
-	memset(vcpu->arch.ctxt.cp15, 0x42, sizeof(vcpu->arch.ctxt.cp15));
+	DECLARE_BITMAP(bmap, NR_CP15_REGS) = { 0, };
 
 	/* Generic chip reset first (so target could override). */
-	reset_coproc_regs(vcpu, cp15_regs, ARRAY_SIZE(cp15_regs));
+	reset_coproc_regs(vcpu, cp15_regs, ARRAY_SIZE(cp15_regs), bmap);
 
 	table = get_target_table(vcpu->arch.target, &num);
-	reset_coproc_regs(vcpu, table, num);
+	reset_coproc_regs(vcpu, table, num, bmap);
 
 	for (num = 1; num < NR_CP15_REGS; num++)
-		WARN(vcpu_cp15(vcpu, num) == 0x42424242,
+		WARN(!test_bit(num, bmap),
 		     "Didn't reset vcpu_cp15(vcpu, %zi)", num);
 }

arch/arm64/include/asm/kvm_arm.h

@@ -316,9 +316,10 @@
 
 #define kvm_arm_exception_class \
 	ECN(UNKNOWN), ECN(WFx), ECN(CP15_32), ECN(CP15_64), ECN(CP14_MR), \
-	ECN(CP14_LS), ECN(FP_ASIMD), ECN(CP10_ID), ECN(CP14_64), ECN(SVC64), \
-	ECN(HVC64), ECN(SMC64), ECN(SYS64), ECN(IMP_DEF), ECN(IABT_LOW), \
-	ECN(IABT_CUR), ECN(PC_ALIGN), ECN(DABT_LOW), ECN(DABT_CUR), \
+	ECN(CP14_LS), ECN(FP_ASIMD), ECN(CP10_ID), ECN(PAC), ECN(CP14_64), \
+	ECN(SVC64), ECN(HVC64), ECN(SMC64), ECN(SYS64), ECN(SVE), \
+	ECN(IMP_DEF), ECN(IABT_LOW), ECN(IABT_CUR), \
+	ECN(PC_ALIGN), ECN(DABT_LOW), ECN(DABT_CUR), \
 	ECN(SP_ALIGN), ECN(FP_EXC32), ECN(FP_EXC64), ECN(SERROR), \
 	ECN(BREAKPT_LOW), ECN(BREAKPT_CUR), ECN(SOFTSTP_LOW), \
 	ECN(SOFTSTP_CUR), ECN(WATCHPT_LOW), ECN(WATCHPT_CUR), \

arch/arm64/kernel/traps.c

@@ -733,6 +733,7 @@ static const char *esr_class_str[] = {
 	[ESR_ELx_EC_CP14_LS]		= "CP14 LDC/STC",
 	[ESR_ELx_EC_FP_ASIMD]		= "ASIMD",
 	[ESR_ELx_EC_CP10_ID]		= "CP10 MRC/VMRS",
+	[ESR_ELx_EC_PAC]		= "PAC",
 	[ESR_ELx_EC_CP14_64]		= "CP14 MCRR/MRRC",
 	[ESR_ELx_EC_ILL]		= "PSTATE.IL",
 	[ESR_ELx_EC_SVC32]		= "SVC (AArch32)",

arch/arm64/kvm/hyp/debug-sr.c

@@ -18,40 +18,70 @@
 #define save_debug(ptr,reg,nr)						\
 	switch (nr) {							\
 	case 15:	ptr[15] = read_debug(reg, 15);			\
+			/* Fall through */				\
 	case 14:	ptr[14] = read_debug(reg, 14);			\
+			/* Fall through */				\
 	case 13:	ptr[13] = read_debug(reg, 13);			\
+			/* Fall through */				\
 	case 12:	ptr[12] = read_debug(reg, 12);			\
+			/* Fall through */				\
 	case 11:	ptr[11] = read_debug(reg, 11);			\
+			/* Fall through */				\
 	case 10:	ptr[10] = read_debug(reg, 10);			\
+			/* Fall through */				\
 	case 9:		ptr[9] = read_debug(reg, 9);			\
+			/* Fall through */				\
 	case 8:		ptr[8] = read_debug(reg, 8);			\
+			/* Fall through */				\
 	case 7:		ptr[7] = read_debug(reg, 7);			\
+			/* Fall through */				\
 	case 6:		ptr[6] = read_debug(reg, 6);			\
+			/* Fall through */				\
 	case 5:		ptr[5] = read_debug(reg, 5);			\
+			/* Fall through */				\
 	case 4:		ptr[4] = read_debug(reg, 4);			\
+			/* Fall through */				\
 	case 3:		ptr[3] = read_debug(reg, 3);			\
+			/* Fall through */				\
 	case 2:		ptr[2] = read_debug(reg, 2);			\
+			/* Fall through */				\
 	case 1:		ptr[1] = read_debug(reg, 1);			\
+			/* Fall through */				\
 	default:	ptr[0] = read_debug(reg, 0);			\
 	}
 
 #define restore_debug(ptr,reg,nr)					\
 	switch (nr) {							\
 	case 15:	write_debug(ptr[15], reg, 15);			\
+			/* Fall through */				\
 	case 14:	write_debug(ptr[14], reg, 14);			\
+			/* Fall through */				\
 	case 13:	write_debug(ptr[13], reg, 13);			\
+			/* Fall through */				\
 	case 12:	write_debug(ptr[12], reg, 12);			\
+			/* Fall through */				\
 	case 11:	write_debug(ptr[11], reg, 11);			\
+			/* Fall through */				\
 	case 10:	write_debug(ptr[10], reg, 10);			\
+			/* Fall through */				\
 	case 9:		write_debug(ptr[9], reg, 9);			\
+			/* Fall through */				\
 	case 8:		write_debug(ptr[8], reg, 8);			\
+			/* Fall through */				\
 	case 7:		write_debug(ptr[7], reg, 7);			\
+			/* Fall through */				\
 	case 6:		write_debug(ptr[6], reg, 6);			\
+			/* Fall through */				\
 	case 5:		write_debug(ptr[5], reg, 5);			\
+			/* Fall through */				\
 	case 4:		write_debug(ptr[4], reg, 4);			\
+			/* Fall through */				\
 	case 3:		write_debug(ptr[3], reg, 3);			\
+			/* Fall through */				\
 	case 2:		write_debug(ptr[2], reg, 2);			\
+			/* Fall through */				\
 	case 1:		write_debug(ptr[1], reg, 1);			\
+			/* Fall through */				\
 	default:	write_debug(ptr[0], reg, 0);			\
 	}
 

arch/arm64/kvm/regmap.c

@@ -178,13 +178,18 @@ void vcpu_write_spsr32(struct kvm_vcpu *vcpu, unsigned long v)
 	switch (spsr_idx) {
 	case KVM_SPSR_SVC:
 		write_sysreg_el1(v, SYS_SPSR);
+		break;
 	case KVM_SPSR_ABT:
 		write_sysreg(v, spsr_abt);
+		break;
 	case KVM_SPSR_UND:
 		write_sysreg(v, spsr_und);
+		break;
 	case KVM_SPSR_IRQ:
 		write_sysreg(v, spsr_irq);
+		break;
 	case KVM_SPSR_FIQ:
 		write_sysreg(v, spsr_fiq);
+		break;
 	}
 }

arch/arm64/kvm/sys_regs.c

@@ -632,7 +632,7 @@ static void reset_pmcr(struct kvm_vcpu *vcpu, const struct sys_reg_desc *r)
 	 */
 	val = ((pmcr & ~ARMV8_PMU_PMCR_MASK)
 	       | (ARMV8_PMU_PMCR_MASK & 0xdecafbad)) & (~ARMV8_PMU_PMCR_E);
-	__vcpu_sys_reg(vcpu, PMCR_EL0) = val;
+	__vcpu_sys_reg(vcpu, r->reg) = val;
 }
 
 static bool check_pmu_access_disabled(struct kvm_vcpu *vcpu, u64 flags)
@@ -981,13 +981,13 @@ static bool access_pmuserenr(struct kvm_vcpu *vcpu, struct sys_reg_params *p,
 /* Silly macro to expand the DBG{BCR,BVR,WVR,WCR}n_EL1 registers in one go */
 #define DBG_BCR_BVR_WCR_WVR_EL1(n)					\
 	{ SYS_DESC(SYS_DBGBVRn_EL1(n)),					\
-	  trap_bvr, reset_bvr, n, 0, get_bvr, set_bvr },		\
+	  trap_bvr, reset_bvr, 0, 0, get_bvr, set_bvr },		\
 	{ SYS_DESC(SYS_DBGBCRn_EL1(n)),					\
-	  trap_bcr, reset_bcr, n, 0, get_bcr, set_bcr },		\
+	  trap_bcr, reset_bcr, 0, 0, get_bcr, set_bcr },		\
 	{ SYS_DESC(SYS_DBGWVRn_EL1(n)),					\
-	  trap_wvr, reset_wvr, n, 0,  get_wvr, set_wvr },		\
+	  trap_wvr, reset_wvr, 0, 0,  get_wvr, set_wvr },		\
 	{ SYS_DESC(SYS_DBGWCRn_EL1(n)),					\
-	  trap_wcr, reset_wcr, n, 0,  get_wcr, set_wcr }
+	  trap_wcr, reset_wcr, 0, 0,  get_wcr, set_wcr }
 
 /* Macro to expand the PMEVCNTRn_EL0 register */
 #define PMU_PMEVCNTR_EL0(n)						\
@@ -1540,7 +1540,7 @@ static const struct sys_reg_desc sys_reg_descs[] = {
 	{ SYS_DESC(SYS_CSSELR_EL1), access_csselr, reset_unknown, CSSELR_EL1 },
 	{ SYS_DESC(SYS_CTR_EL0), access_ctr },
 
-	{ SYS_DESC(SYS_PMCR_EL0), access_pmcr, reset_pmcr, },
+	{ SYS_DESC(SYS_PMCR_EL0), access_pmcr, reset_pmcr, PMCR_EL0 },
 	{ SYS_DESC(SYS_PMCNTENSET_EL0), access_pmcnten, reset_unknown, PMCNTENSET_EL0 },
 	{ SYS_DESC(SYS_PMCNTENCLR_EL0), access_pmcnten, NULL, PMCNTENSET_EL0 },
 	{ SYS_DESC(SYS_PMOVSCLR_EL0), access_pmovs, NULL, PMOVSSET_EL0 },
@@ -2254,13 +2254,19 @@ static int emulate_sys_reg(struct kvm_vcpu *vcpu,
 }
 
 static void reset_sys_reg_descs(struct kvm_vcpu *vcpu,
-			      const struct sys_reg_desc *table, size_t num)
+				const struct sys_reg_desc *table, size_t num,
+				unsigned long *bmap)
 {
 	unsigned long i;
 
 	for (i = 0; i < num; i++)
-		if (table[i].reset)
+		if (table[i].reset) {
+			int reg = table[i].reg;
+
 			table[i].reset(vcpu, &table[i]);
+			if (reg > 0 && reg < NR_SYS_REGS)
+				set_bit(reg, bmap);
+		}
 }
 
 /**
@@ -2774,18 +2780,16 @@ void kvm_reset_sys_regs(struct kvm_vcpu *vcpu)
 {
 	size_t num;
 	const struct sys_reg_desc *table;
-
-	/* Catch someone adding a register without putting in reset entry. */
-	memset(&vcpu->arch.ctxt.sys_regs, 0x42, sizeof(vcpu->arch.ctxt.sys_regs));
+	DECLARE_BITMAP(bmap, NR_SYS_REGS) = { 0, };
 
 	/* Generic chip reset first (so target could override). */
-	reset_sys_reg_descs(vcpu, sys_reg_descs, ARRAY_SIZE(sys_reg_descs));
+	reset_sys_reg_descs(vcpu, sys_reg_descs, ARRAY_SIZE(sys_reg_descs), bmap);
 
 	table = get_target_table(vcpu->arch.target, true, &num);
-	reset_sys_reg_descs(vcpu, table, num);
+	reset_sys_reg_descs(vcpu, table, num, bmap);
 
 	for (num = 1; num < NR_SYS_REGS; num++) {
-		if (WARN(__vcpu_sys_reg(vcpu, num) == 0x4242424242424242,
+		if (WARN(!test_bit(num, bmap),
 			 "Didn't reset __vcpu_sys_reg(%zi)\n", num))
 			break;
 	}

arch/mips/kvm/mips.c

@@ -150,16 +150,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	return 0;
 }
 
-bool kvm_arch_has_vcpu_debugfs(void)
-{
-	return false;
-}
-
-int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
-{
-	return 0;
-}
-
 void kvm_mips_free_vcpus(struct kvm *kvm)
 {
 	unsigned int i;

arch/powerpc/kvm/powerpc.c

@@ -50,6 +50,11 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *v)
 	return !!(v->arch.pending_exceptions) || kvm_request_pending(v);
 }
 
+bool kvm_arch_dy_runnable(struct kvm_vcpu *vcpu)
+{
+	return kvm_arch_vcpu_runnable(vcpu);
+}
+
 bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
 {
 	return false;
@@ -452,16 +457,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	return -EINVAL;
 }
 
-bool kvm_arch_has_vcpu_debugfs(void)
-{
-	return false;
-}
-
-int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
-{
-	return 0;
-}
-
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
 	unsigned int i;

arch/s390/kvm/kvm-s390.c

@@ -2516,16 +2516,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	return rc;
 }
 
-bool kvm_arch_has_vcpu_debugfs(void)
-{
-	return false;
-}
-
-int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
-{
-	return 0;
-}
-
 void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
 {
 	VCPU_EVENT(vcpu, 3, "%s", "free cpu");

arch/x86/include/asm/kvm_host.h

@@ -35,6 +35,8 @@
 #include <asm/kvm_vcpu_regs.h>
 #include <asm/hyperv-tlfs.h>
 
+#define __KVM_HAVE_ARCH_VCPU_DEBUGFS
+
 #define KVM_MAX_VCPUS 288
 #define KVM_SOFT_MAX_VCPUS 240
 #define KVM_MAX_VCPU_ID 1023
@@ -1175,6 +1177,7 @@ struct kvm_x86_ops {
 	int (*update_pi_irte)(struct kvm *kvm, unsigned int host_irq,
 			      uint32_t guest_irq, bool set);
 	void (*apicv_post_state_restore)(struct kvm_vcpu *vcpu);
+	bool (*dy_apicv_has_pending_interrupt)(struct kvm_vcpu *vcpu);
 
 	int (*set_hv_timer)(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc,
 			    bool *expired);

arch/x86/kernel/kvm.c

@@ -308,9 +308,6 @@ static notrace void kvm_guest_apic_eoi_write(u32 reg, u32 val)
 
 static void kvm_guest_cpu_init(void)
 {
-	if (!kvm_para_available())
-		return;
-
 	if (kvm_para_has_feature(KVM_FEATURE_ASYNC_PF) && kvmapf) {
 		u64 pa = slow_virt_to_phys(this_cpu_ptr(&apf_reason));
 
@@ -625,9 +622,6 @@ static void __init kvm_guest_init(void)
 {
 	int i;
 
-	if (!kvm_para_available())
-		return;
-
 	paravirt_ops_setup();
 	register_reboot_notifier(&kvm_pv_reboot_nb);
 	for (i = 0; i < KVM_TASK_SLEEP_HASHSIZE; i++)
@@ -848,8 +842,6 @@ asm(
  */
 void __init kvm_spinlock_init(void)
 {
-	if (!kvm_para_available())
-		return;
 	/* Does host kernel support KVM_FEATURE_PV_UNHALT? */
 	if (!kvm_para_has_feature(KVM_FEATURE_PV_UNHALT))
 		return;

arch/x86/kvm/debugfs.c

@@ -8,11 +8,6 @@
 #include <linux/debugfs.h>
 #include "lapic.h"
 
-bool kvm_arch_has_vcpu_debugfs(void)
-{
-	return true;
-}
-
 static int vcpu_get_timer_advance_ns(void *data, u64 *val)
 {
 	struct kvm_vcpu *vcpu = (struct kvm_vcpu *) data;
@@ -48,37 +43,22 @@ static int vcpu_get_tsc_scaling_frac_bits(void *data, u64 *val)
 
 DEFINE_SIMPLE_ATTRIBUTE(vcpu_tsc_scaling_frac_fops, vcpu_get_tsc_scaling_frac_bits, NULL, "%llu\n");
 
-int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
+void kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
 {
-	struct dentry *ret;
-
-	ret = debugfs_create_file("tsc-offset", 0444,
-							vcpu->debugfs_dentry,
-							vcpu, &vcpu_tsc_offset_fops);
-	if (!ret)
-		return -ENOMEM;
+	debugfs_create_file("tsc-offset", 0444, vcpu->debugfs_dentry, vcpu,
+			    &vcpu_tsc_offset_fops);
 
-	if (lapic_in_kernel(vcpu)) {
-		ret = debugfs_create_file("lapic_timer_advance_ns", 0444,
-								vcpu->debugfs_dentry,
-								vcpu, &vcpu_timer_advance_ns_fops);
-		if (!ret)
-			return -ENOMEM;
-	}
+	if (lapic_in_kernel(vcpu))
+		debugfs_create_file("lapic_timer_advance_ns", 0444,
+				    vcpu->debugfs_dentry, vcpu,
+				    &vcpu_timer_advance_ns_fops);
 
 	if (kvm_has_tsc_control) {
-		ret = debugfs_create_file("tsc-scaling-ratio", 0444,
-							vcpu->debugfs_dentry,
-							vcpu, &vcpu_tsc_scaling_fops);
-		if (!ret)
-			return -ENOMEM;
-		ret = debugfs_create_file("tsc-scaling-ratio-frac-bits", 0444,
-							vcpu->debugfs_dentry,
-							vcpu, &vcpu_tsc_scaling_frac_fops);
-		if (!ret)
-			return -ENOMEM;
-
+		debugfs_create_file("tsc-scaling-ratio", 0444,
+				    vcpu->debugfs_dentry, vcpu,
+				    &vcpu_tsc_scaling_fops);
+		debugfs_create_file("tsc-scaling-ratio-frac-bits", 0444,
+				    vcpu->debugfs_dentry, vcpu,
+				    &vcpu_tsc_scaling_frac_fops);
 	}
-
-	return 0;
 }

arch/x86/kvm/lapic.c

@@ -1548,7 +1548,6 @@ static void kvm_apic_inject_pending_timer_irqs(struct kvm_lapic *apic)
 static void apic_timer_expired(struct kvm_lapic *apic)
 {
 	struct kvm_vcpu *vcpu = apic->vcpu;
-	struct swait_queue_head *q = &vcpu->wq;
 	struct kvm_timer *ktimer = &apic->lapic_timer;
 
 	if (atomic_read(&apic->lapic_timer.pending))
@@ -1566,13 +1565,6 @@ static void apic_timer_expired(struct kvm_lapic *apic)
 
 	atomic_inc(&apic->lapic_timer.pending);
 	kvm_set_pending_timer(vcpu);
-
-	/*
-	 * For x86, the atomic_inc() is serialized, thus
-	 * using swait_active() is safe.
-	 */
-	if (swait_active(q))
-		swake_up_one(q);
 }
 
 static void start_sw_tscdeadline(struct kvm_lapic *apic)

arch/x86/kvm/svm.c

@@ -5190,6 +5190,11 @@ static void svm_deliver_avic_intr(struct kvm_vcpu *vcpu, int vec)
 		kvm_vcpu_wake_up(vcpu);
 }
 
+static bool svm_dy_apicv_has_pending_interrupt(struct kvm_vcpu *vcpu)
+{
+	return false;
+}
+
 static void svm_ir_list_del(struct vcpu_svm *svm, struct amd_iommu_pi_data *pi)
 {
 	unsigned long flags;
@@ -7314,6 +7319,7 @@ static struct kvm_x86_ops svm_x86_ops __ro_after_init = {
 
 	.pmu_ops = &amd_pmu_ops,
 	.deliver_posted_interrupt = svm_deliver_avic_intr,
+	.dy_apicv_has_pending_interrupt = svm_dy_apicv_has_pending_interrupt,
 	.update_pi_irte = svm_update_pi_irte,
 	.setup_mce = svm_setup_mce,
 

arch/x86/kvm/vmx/vmx.c

@@ -6117,6 +6117,11 @@ static int vmx_sync_pir_to_irr(struct kvm_vcpu *vcpu)
 	return max_irr;
 }
 
+static bool vmx_dy_apicv_has_pending_interrupt(struct kvm_vcpu *vcpu)
+{
+	return pi_test_on(vcpu_to_pi_desc(vcpu));
+}
+
 static void vmx_load_eoi_exitmap(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap)
 {
 	if (!kvm_vcpu_apicv_active(vcpu))
@@ -7726,6 +7731,7 @@ static struct kvm_x86_ops vmx_x86_ops __ro_after_init = {
 	.guest_apic_has_interrupt = vmx_guest_apic_has_interrupt,
 	.sync_pir_to_irr = vmx_sync_pir_to_irr,
 	.deliver_posted_interrupt = vmx_deliver_posted_interrupt,
+	.dy_apicv_has_pending_interrupt = vmx_dy_apicv_has_pending_interrupt,
 
 	.set_tss_addr = vmx_set_tss_addr,
 	.set_identity_map_addr = vmx_set_identity_map_addr,

arch/x86/kvm/x86.c

@@ -9698,6 +9698,22 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 	return kvm_vcpu_running(vcpu) || kvm_vcpu_has_events(vcpu);
 }
 
+bool kvm_arch_dy_runnable(struct kvm_vcpu *vcpu)
+{
+	if (READ_ONCE(vcpu->arch.pv.pv_unhalted))
+		return true;
+
+	if (kvm_test_request(KVM_REQ_NMI, vcpu) ||
+		kvm_test_request(KVM_REQ_SMI, vcpu) ||
+		 kvm_test_request(KVM_REQ_EVENT, vcpu))
+		return true;
+
+	if (vcpu->arch.apicv_active && kvm_x86_ops->dy_apicv_has_pending_interrupt(vcpu))
+		return true;
+
+	return false;
+}
+
 bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu)
 {
 	return vcpu->arch.preempted_in_kernel;

include/kvm/arm_pmu.h

@@ -34,6 +34,7 @@ struct kvm_pmu {
 u64 kvm_pmu_get_counter_value(struct kvm_vcpu *vcpu, u64 select_idx);
 void kvm_pmu_set_counter_value(struct kvm_vcpu *vcpu, u64 select_idx, u64 val);
 u64 kvm_pmu_valid_counter_mask(struct kvm_vcpu *vcpu);
+void kvm_pmu_vcpu_init(struct kvm_vcpu *vcpu);
 void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu);
 void kvm_pmu_vcpu_destroy(struct kvm_vcpu *vcpu);
 void kvm_pmu_disable_counter_mask(struct kvm_vcpu *vcpu, u64 val);
@@ -71,6 +72,7 @@ static inline u64 kvm_pmu_valid_counter_mask(struct kvm_vcpu *vcpu)
 {
 	return 0;
 }
+static inline void kvm_pmu_vcpu_init(struct kvm_vcpu *vcpu) {}
 static inline void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu) {}
 static inline void kvm_pmu_vcpu_destroy(struct kvm_vcpu *vcpu) {}
 static inline void kvm_pmu_disable_counter_mask(struct kvm_vcpu *vcpu, u64 val) {}

include/kvm/arm_vgic.h

@@ -350,6 +350,7 @@ int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu);
 
 void kvm_vgic_load(struct kvm_vcpu *vcpu);
 void kvm_vgic_put(struct kvm_vcpu *vcpu);
+void kvm_vgic_vmcr_sync(struct kvm_vcpu *vcpu);
 
 #define irqchip_in_kernel(k)	(!!((k)->arch.vgic.in_kernel))
 #define vgic_initialized(k)	((k)->arch.vgic.initialized)

include/linux/kvm_host.h

@@ -861,8 +861,9 @@ int kvm_arch_vcpu_setup(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_postcreate(struct kvm_vcpu *vcpu);
 void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu);
 
-bool kvm_arch_has_vcpu_debugfs(void);
-int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu);
+#ifdef __KVM_HAVE_ARCH_VCPU_DEBUGFS
+void kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu);
+#endif
 
 int kvm_arch_hardware_enable(void);
 void kvm_arch_hardware_disable(void);
@@ -872,6 +873,7 @@ int kvm_arch_check_processor_compat(void);
 int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu);
 bool kvm_arch_vcpu_in_kernel(struct kvm_vcpu *vcpu);
 int kvm_arch_vcpu_should_kick(struct kvm_vcpu *vcpu);
+bool kvm_arch_dy_runnable(struct kvm_vcpu *vcpu);
 
 #ifndef __KVM_HAVE_ARCH_VM_ALLOC
 /*

tools/testing/selftests/kvm/.gitignore

+/s390x/sync_regs_test
 /x86_64/cr4_cpuid_sync_test
 /x86_64/evmcs_test
 /x86_64/hyperv_cpuid
-/x86_64/kvm_create_max_vcpus
 /x86_64/mmio_warning_test
 /x86_64/platform_info_test
 /x86_64/set_sregs_test
@@ -13,3 +13,4 @@
 /x86_64/vmx_tsc_adjust_test
 /clear_dirty_log_test
 /dirty_log_test
+/kvm_create_max_vcpus

tools/testing/selftests/kvm/config

+CONFIG_KVM=y
+CONFIG_KVM_INTEL=y
+CONFIG_KVM_AMD=y

virt/kvm/arm/arm.c

@@ -144,11 +144,6 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	return ret;
 }
 
-bool kvm_arch_has_vcpu_debugfs(void)
-{
-	return false;
-}
-
 int kvm_arch_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
 {
 	return 0;
@@ -323,6 +318,17 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
 
 void kvm_arch_vcpu_blocking(struct kvm_vcpu *vcpu)
 {
+	/*
+	 * If we're about to block (most likely because we've just hit a
+	 * WFI), we need to sync back the state of the GIC CPU interface
+	 * so that we have the lastest PMR and group enables. This ensures
+	 * that kvm_arch_vcpu_runnable has up-to-date data to decide
+	 * whether we have pending interrupts.
+	 */
+	preempt_disable();
+	kvm_vgic_vmcr_sync(vcpu);
+	preempt_enable();
+
 	kvm_vgic_v4_enable_doorbell(vcpu);
 }
 
@@ -340,6 +346,8 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
 	/* Set up the timer */
 	kvm_timer_vcpu_init(vcpu);
 
+	kvm_pmu_vcpu_init(vcpu);
+
 	kvm_arm_reset_debug_ptr(vcpu);
 
 	return kvm_vgic_vcpu_init(vcpu);

virt/kvm/arm/hyp/vgic-v3-sr.c

@@ -349,8 +349,10 @@ void __hyp_text __vgic_v3_save_aprs(struct kvm_vcpu *vcpu)
 	case 7:
 		cpu_if->vgic_ap0r[3] = __vgic_v3_read_ap0rn(3);
 		cpu_if->vgic_ap0r[2] = __vgic_v3_read_ap0rn(2);
+		/* Fall through */
 	case 6:
 		cpu_if->vgic_ap0r[1] = __vgic_v3_read_ap0rn(1);
+		/* Fall through */
 	default:
 		cpu_if->vgic_ap0r[0] = __vgic_v3_read_ap0rn(0);
 	}
@@ -359,8 +361,10 @@ void __hyp_text __vgic_v3_save_aprs(struct kvm_vcpu *vcpu)
 	case 7:
 		cpu_if->vgic_ap1r[3] = __vgic_v3_read_ap1rn(3);
 		cpu_if->vgic_ap1r[2] = __vgic_v3_read_ap1rn(2);
+		/* Fall through */
 	case 6:
 		cpu_if->vgic_ap1r[1] = __vgic_v3_read_ap1rn(1);
+		/* Fall through */
 	default:
 		cpu_if->vgic_ap1r[0] = __vgic_v3_read_ap1rn(0);
 	}
@@ -382,8 +386,10 @@ void __hyp_text __vgic_v3_restore_aprs(struct kvm_vcpu *vcpu)
 	case 7:
 		__vgic_v3_write_ap0rn(cpu_if->vgic_ap0r[3], 3);
 		__vgic_v3_write_ap0rn(cpu_if->vgic_ap0r[2], 2);
+		/* Fall through */
 	case 6:
 		__vgic_v3_write_ap0rn(cpu_if->vgic_ap0r[1], 1);
+		/* Fall through */
 	default:
 		__vgic_v3_write_ap0rn(cpu_if->vgic_ap0r[0], 0);
 	}
@@ -392,8 +398,10 @@ void __hyp_text __vgic_v3_restore_aprs(struct kvm_vcpu *vcpu)
 	case 7:
 		__vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[3], 3);
 		__vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[2], 2);
+		/* Fall through */
 	case 6:
 		__vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[1], 1);
+		/* Fall through */
 	default:
 		__vgic_v3_write_ap1rn(cpu_if->vgic_ap1r[0], 0);
 	}

virt/kvm/arm/pmu.c

@@ -214,6 +214,20 @@ static void kvm_pmu_stop_counter(struct kvm_vcpu *vcpu, struct kvm_pmc *pmc)
 	kvm_pmu_release_perf_event(pmc);
 }
 
+/**
+ * kvm_pmu_vcpu_init - assign pmu counter idx for cpu
+ * @vcpu: The vcpu pointer
+ *
+ */
+void kvm_pmu_vcpu_init(struct kvm_vcpu *vcpu)
+{
+	int i;
+	struct kvm_pmu *pmu = &vcpu->arch.pmu;
+
+	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++)
+		pmu->pmc[i].idx = i;
+}
+
 /**
  * kvm_pmu_vcpu_reset - reset pmu state for cpu
  * @vcpu: The vcpu pointer
@@ -224,10 +238,8 @@ void kvm_pmu_vcpu_reset(struct kvm_vcpu *vcpu)
 	int i;
 	struct kvm_pmu *pmu = &vcpu->arch.pmu;
 
-	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++) {
+	for (i = 0; i < ARMV8_PMU_MAX_COUNTERS; i++)
 		kvm_pmu_stop_counter(vcpu, &pmu->pmc[i]);
-		pmu->pmc[i].idx = i;
-	}
 
 	bitmap_zero(vcpu->arch.pmu.chained, ARMV8_PMU_MAX_COUNTER_PAIRS);
 }

virt/kvm/arm/vgic/vgic-mmio.c

@@ -113,6 +113,22 @@ void vgic_mmio_write_senable(struct kvm_vcpu *vcpu,
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 
 		raw_spin_lock_irqsave(&irq->irq_lock, flags);
+		if (vgic_irq_is_mapped_level(irq)) {
+			bool was_high = irq->line_level;
+
+			/*
+			 * We need to update the state of the interrupt because
+			 * the guest might have changed the state of the device
+			 * while the interrupt was disabled at the VGIC level.
+			 */
+			irq->line_level = vgic_get_phys_line_level(irq);
+			/*
+			 * Deactivate the physical interrupt so the GIC will let
+			 * us know when it is asserted again.
+			 */
+			if (!irq->active && was_high && !irq->line_level)
+				vgic_irq_set_phys_active(irq, false);
+		}
 		irq->enabled = true;
 		vgic_queue_irq_unlock(vcpu->kvm, irq, flags);
 

virt/kvm/arm/vgic/vgic-v2.c

@@ -484,10 +484,17 @@ void vgic_v2_load(struct kvm_vcpu *vcpu)
 		       kvm_vgic_global_state.vctrl_base + GICH_APR);
 }
 
-void vgic_v2_put(struct kvm_vcpu *vcpu)
+void vgic_v2_vmcr_sync(struct kvm_vcpu *vcpu)
 {
 	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
 
 	cpu_if->vgic_vmcr = readl_relaxed(kvm_vgic_global_state.vctrl_base + GICH_VMCR);
+}
+
+void vgic_v2_put(struct kvm_vcpu *vcpu)
+{
+	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
+
+	vgic_v2_vmcr_sync(vcpu);
 	cpu_if->vgic_apr = readl_relaxed(kvm_vgic_global_state.vctrl_base + GICH_APR);
 }

virt/kvm/arm/vgic/vgic-v3.c

@@ -662,12 +662,17 @@ void vgic_v3_load(struct kvm_vcpu *vcpu)
 		__vgic_v3_activate_traps(vcpu);
 }
 
-void vgic_v3_put(struct kvm_vcpu *vcpu)
+void vgic_v3_vmcr_sync(struct kvm_vcpu *vcpu)
 {
 	struct vgic_v3_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v3;
 
 	if (likely(cpu_if->vgic_sre))
 		cpu_if->vgic_vmcr = kvm_call_hyp_ret(__vgic_v3_read_vmcr);
+}
+
+void vgic_v3_put(struct kvm_vcpu *vcpu)
+{
+	vgic_v3_vmcr_sync(vcpu);
 
 	kvm_call_hyp(__vgic_v3_save_aprs, vcpu);
 

virt/kvm/arm/vgic/vgic.c

@@ -919,6 +919,17 @@ void kvm_vgic_put(struct kvm_vcpu *vcpu)
 		vgic_v3_put(vcpu);
 }
 
+void kvm_vgic_vmcr_sync(struct kvm_vcpu *vcpu)
+{
+	if (unlikely(!irqchip_in_kernel(vcpu->kvm)))
+		return;
+
+	if (kvm_vgic_global_state.type == VGIC_V2)
+		vgic_v2_vmcr_sync(vcpu);
+	else
+		vgic_v3_vmcr_sync(vcpu);
+}
+
 int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu)
 {
 	struct vgic_cpu *vgic_cpu = &vcpu->arch.vgic_cpu;

virt/kvm/arm/vgic/vgic.h

@@ -193,6 +193,7 @@ int vgic_register_dist_iodev(struct kvm *kvm, gpa_t dist_base_address,
 void vgic_v2_init_lrs(void);
 void vgic_v2_load(struct kvm_vcpu *vcpu);
 void vgic_v2_put(struct kvm_vcpu *vcpu);
+void vgic_v2_vmcr_sync(struct kvm_vcpu *vcpu);
 
 void vgic_v2_save_state(struct kvm_vcpu *vcpu);
 void vgic_v2_restore_state(struct kvm_vcpu *vcpu);
@@ -223,6 +224,7 @@ bool vgic_v3_check_base(struct kvm *kvm);
 
 void vgic_v3_load(struct kvm_vcpu *vcpu);
 void vgic_v3_put(struct kvm_vcpu *vcpu);
+void vgic_v3_vmcr_sync(struct kvm_vcpu *vcpu);
 
 bool vgic_has_its(struct kvm *kvm);
 int kvm_vgic_register_its_device(void);

virt/kvm/kvm_main.c

@@ -1855,8 +1855,7 @@ void kvm_set_pfn_dirty(kvm_pfn_t pfn)
 	if (!kvm_is_reserved_pfn(pfn)) {
 		struct page *page = pfn_to_page(pfn);
 
-		if (!PageReserved(page))
-			SetPageDirty(page);
+		SetPageDirty(page);
 	}
 }
 EXPORT_SYMBOL_GPL(kvm_set_pfn_dirty);
@@ -2477,6 +2476,29 @@ static bool kvm_vcpu_eligible_for_directed_yield(struct kvm_vcpu *vcpu)
 #endif
 }
 
+/*
+ * Unlike kvm_arch_vcpu_runnable, this function is called outside
+ * a vcpu_load/vcpu_put pair.  However, for most architectures
+ * kvm_arch_vcpu_runnable does not require vcpu_load.
+ */
+bool __weak kvm_arch_dy_runnable(struct kvm_vcpu *vcpu)
+{
+	return kvm_arch_vcpu_runnable(vcpu);
+}
+
+static bool vcpu_dy_runnable(struct kvm_vcpu *vcpu)
+{
+	if (kvm_arch_dy_runnable(vcpu))
+		return true;
+
+#ifdef CONFIG_KVM_ASYNC_PF
+	if (!list_empty_careful(&vcpu->async_pf.done))
+		return true;
+#endif
+
+	return false;
+}
+
 void kvm_vcpu_on_spin(struct kvm_vcpu *me, bool yield_to_kernel_mode)
 {
 	struct kvm *kvm = me->kvm;
@@ -2506,9 +2528,10 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me, bool yield_to_kernel_mode)
 				continue;
 			if (vcpu == me)
 				continue;
-			if (swait_active(&vcpu->wq) && !kvm_arch_vcpu_runnable(vcpu))
+			if (swait_active(&vcpu->wq) && !vcpu_dy_runnable(vcpu))
 				continue;
-			if (yield_to_kernel_mode && !kvm_arch_vcpu_in_kernel(vcpu))
+			if (READ_ONCE(vcpu->preempted) && yield_to_kernel_mode &&
+				!kvm_arch_vcpu_in_kernel(vcpu))
 				continue;
 			if (!kvm_vcpu_eligible_for_directed_yield(vcpu))
 				continue;
@@ -2591,30 +2614,20 @@ static int create_vcpu_fd(struct kvm_vcpu *vcpu)
 	return anon_inode_getfd(name, &kvm_vcpu_fops, vcpu, O_RDWR | O_CLOEXEC);
 }
 
-static int kvm_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
+static void kvm_create_vcpu_debugfs(struct kvm_vcpu *vcpu)
 {
+#ifdef __KVM_HAVE_ARCH_VCPU_DEBUGFS
 	char dir_name[ITOA_MAX_LEN * 2];
-	int ret;
-
-	if (!kvm_arch_has_vcpu_debugfs())
-		return 0;
 
 	if (!debugfs_initialized())
-		return 0;
+		return;
 
 	snprintf(dir_name, sizeof(dir_name), "vcpu%d", vcpu->vcpu_id);
 	vcpu->debugfs_dentry = debugfs_create_dir(dir_name,
-								vcpu->kvm->debugfs_dentry);
-	if (!vcpu->debugfs_dentry)
-		return -ENOMEM;
-
-	ret = kvm_arch_create_vcpu_debugfs(vcpu);
-	if (ret < 0) {
-		debugfs_remove_recursive(vcpu->debugfs_dentry);
-		return ret;
-	}
+						  vcpu->kvm->debugfs_dentry);
 
-	return 0;
+	kvm_arch_create_vcpu_debugfs(vcpu);
+#endif
 }
 
 /*
@@ -2649,9 +2662,7 @@ static int kvm_vm_ioctl_create_vcpu(struct kvm *kvm, u32 id)
 	if (r)
 		goto vcpu_destroy;
 
-	r = kvm_create_vcpu_debugfs(vcpu);
-	if (r)
-		goto vcpu_destroy;
+	kvm_create_vcpu_debugfs(vcpu);
 
 	mutex_lock(&kvm->lock);
 	if (kvm_get_vcpu_by_id(kvm, id)) {
@@ -4205,7 +4216,7 @@ static void kvm_sched_in(struct preempt_notifier *pn, int cpu)
 {
 	struct kvm_vcpu *vcpu = preempt_notifier_to_vcpu(pn);
 
-	vcpu->preempted = false;
+	WRITE_ONCE(vcpu->preempted, false);
 	WRITE_ONCE(vcpu->ready, false);
 
 	kvm_arch_sched_in(vcpu, cpu);
@@ -4219,7 +4230,7 @@ static void kvm_sched_out(struct preempt_notifier *pn,
 	struct kvm_vcpu *vcpu = preempt_notifier_to_vcpu(pn);
 
 	if (current->state == TASK_RUNNING) {
-		vcpu->preempted = true;
+		WRITE_ONCE(vcpu->preempted, true);
 		WRITE_ONCE(vcpu->ready, true);
 	}
 	kvm_arch_vcpu_put(vcpu);