Merge tag 'loongarch-kvm-6.19' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson into HEAD

LoongArch KVM changes for v6.19

1. Get VM PMU capability from HW GCFG register.
2. Add AVEC basic support.
3. Use 64-bit register definition for EIOINTC.
4. Add KVM timer test cases for tools/selftests.

Author: bonzini

arch/loongarch/include/asm/kvm_eiointc.h

@@ -10,10 +10,7 @@
 
 #define EIOINTC_IRQS			256
 #define EIOINTC_ROUTE_MAX_VCPUS		256
-#define EIOINTC_IRQS_U8_NUMS		(EIOINTC_IRQS / 8)
-#define EIOINTC_IRQS_U16_NUMS		(EIOINTC_IRQS_U8_NUMS / 2)
-#define EIOINTC_IRQS_U32_NUMS		(EIOINTC_IRQS_U8_NUMS / 4)
-#define EIOINTC_IRQS_U64_NUMS		(EIOINTC_IRQS_U8_NUMS / 8)
+#define EIOINTC_IRQS_U64_NUMS		(EIOINTC_IRQS / 64)
 /* map to ipnum per 32 irqs */
 #define EIOINTC_IRQS_NODETYPE_COUNT	16
 
@@ -64,54 +61,18 @@ struct loongarch_eiointc {
 	uint32_t status;
 
 	/* hardware state */
-	union nodetype {
-		u64 reg_u64[EIOINTC_IRQS_NODETYPE_COUNT / 4];
-		u32 reg_u32[EIOINTC_IRQS_NODETYPE_COUNT / 2];
-		u16 reg_u16[EIOINTC_IRQS_NODETYPE_COUNT];
-		u8 reg_u8[EIOINTC_IRQS_NODETYPE_COUNT * 2];
-	} nodetype;
+	u64 nodetype[EIOINTC_IRQS_NODETYPE_COUNT / 4];
 
 	/* one bit shows the state of one irq */
-	union bounce {
-		u64 reg_u64[EIOINTC_IRQS_U64_NUMS];
-		u32 reg_u32[EIOINTC_IRQS_U32_NUMS];
-		u16 reg_u16[EIOINTC_IRQS_U16_NUMS];
-		u8 reg_u8[EIOINTC_IRQS_U8_NUMS];
-	} bounce;
-
-	union isr {
-		u64 reg_u64[EIOINTC_IRQS_U64_NUMS];
-		u32 reg_u32[EIOINTC_IRQS_U32_NUMS];
-		u16 reg_u16[EIOINTC_IRQS_U16_NUMS];
-		u8 reg_u8[EIOINTC_IRQS_U8_NUMS];
-	} isr;
-	union coreisr {
-		u64 reg_u64[EIOINTC_ROUTE_MAX_VCPUS][EIOINTC_IRQS_U64_NUMS];
-		u32 reg_u32[EIOINTC_ROUTE_MAX_VCPUS][EIOINTC_IRQS_U32_NUMS];
-		u16 reg_u16[EIOINTC_ROUTE_MAX_VCPUS][EIOINTC_IRQS_U16_NUMS];
-		u8 reg_u8[EIOINTC_ROUTE_MAX_VCPUS][EIOINTC_IRQS_U8_NUMS];
-	} coreisr;
-	union enable {
-		u64 reg_u64[EIOINTC_IRQS_U64_NUMS];
-		u32 reg_u32[EIOINTC_IRQS_U32_NUMS];
-		u16 reg_u16[EIOINTC_IRQS_U16_NUMS];
-		u8 reg_u8[EIOINTC_IRQS_U8_NUMS];
-	} enable;
+	u64 bounce[EIOINTC_IRQS_U64_NUMS];
+	u64 isr[EIOINTC_IRQS_U64_NUMS];
+	u64 coreisr[EIOINTC_ROUTE_MAX_VCPUS][EIOINTC_IRQS_U64_NUMS];
+	u64 enable[EIOINTC_IRQS_U64_NUMS];
 
 	/* use one byte to config ipmap for 32 irqs at once */
-	union ipmap {
-		u64 reg_u64;
-		u32 reg_u32[EIOINTC_IRQS_U32_NUMS / 4];
-		u16 reg_u16[EIOINTC_IRQS_U16_NUMS / 4];
-		u8 reg_u8[EIOINTC_IRQS_U8_NUMS / 4];
-	} ipmap;
+	u64 ipmap;
 	/* use one byte to config coremap for one irq */
-	union coremap {
-		u64 reg_u64[EIOINTC_IRQS / 8];
-		u32 reg_u32[EIOINTC_IRQS / 4];
-		u16 reg_u16[EIOINTC_IRQS / 2];
-		u8 reg_u8[EIOINTC_IRQS];
-	} coremap;
+	u64 coremap[EIOINTC_IRQS / 8];
 
 	DECLARE_BITMAP(sw_coreisr[EIOINTC_ROUTE_MAX_VCPUS][LOONGSON_IP_NUM], EIOINTC_IRQS);
 	uint8_t  sw_coremap[EIOINTC_IRQS];

arch/loongarch/include/asm/kvm_host.h

@@ -126,6 +126,8 @@ struct kvm_arch {
 	struct kvm_phyid_map  *phyid_map;
 	/* Enabled PV features */
 	unsigned long pv_features;
+	/* Supported KVM features */
+	unsigned long kvm_features;
 
 	s64 time_offset;
 	struct kvm_context __percpu *vmcs;
@@ -293,6 +295,12 @@ static inline int kvm_get_pmu_num(struct kvm_vcpu_arch *arch)
 	return (arch->cpucfg[6] & CPUCFG6_PMNUM) >> CPUCFG6_PMNUM_SHIFT;
 }
 
+/* Check whether KVM support this feature (VMM may disable it) */
+static inline bool kvm_vm_support(struct kvm_arch *arch, int feature)
+{
+	return !!(arch->kvm_features & BIT_ULL(feature));
+}
+
 bool kvm_arch_pmi_in_guest(struct kvm_vcpu *vcpu);
 
 /* Debug: dump vcpu state */

arch/loongarch/include/asm/kvm_vcpu.h

@@ -15,6 +15,7 @@
 #define CPU_PMU				(_ULCAST_(1) << 10)
 #define CPU_TIMER			(_ULCAST_(1) << 11)
 #define CPU_IPI				(_ULCAST_(1) << 12)
+#define CPU_AVEC                        (_ULCAST_(1) << 14)
 
 /* Controlled by 0x52 guest exception VIP aligned to estat bit 5~12 */
 #define CPU_IP0				(_ULCAST_(1))

arch/loongarch/include/asm/loongarch.h

@@ -511,6 +511,8 @@
 #define  CSR_GCFG_GPERF_SHIFT		24
 #define  CSR_GCFG_GPERF_WIDTH		3
 #define  CSR_GCFG_GPERF			(_ULCAST_(0x7) << CSR_GCFG_GPERF_SHIFT)
+#define  CSR_GCFG_GPMP_SHIFT		23
+#define  CSR_GCFG_GPMP			(_ULCAST_(0x1) << CSR_GCFG_GPMP_SHIFT)
 #define  CSR_GCFG_GCI_SHIFT		20
 #define  CSR_GCFG_GCI_WIDTH		2
 #define  CSR_GCFG_GCI			(_ULCAST_(0x3) << CSR_GCFG_GCI_SHIFT)

arch/loongarch/include/uapi/asm/kvm.h

@@ -104,6 +104,7 @@ struct kvm_fpu {
 #define  KVM_LOONGARCH_VM_FEAT_PV_IPI		6
 #define  KVM_LOONGARCH_VM_FEAT_PV_STEALTIME	7
 #define  KVM_LOONGARCH_VM_FEAT_PTW		8
+#define  KVM_LOONGARCH_VM_FEAT_MSGINT		9
 
 /* Device Control API on vcpu fd */
 #define KVM_LOONGARCH_VCPU_CPUCFG	0

arch/loongarch/kvm/intc/eiointc.c

@@ -13,19 +13,19 @@ static void eiointc_set_sw_coreisr(struct loongarch_eiointc *s)
 	struct kvm_vcpu *vcpu;
 
 	for (irq = 0; irq < EIOINTC_IRQS; irq++) {
-		ipnum = s->ipmap.reg_u8[irq / 32];
+		ipnum = (s->ipmap >> (irq / 32 * 8)) & 0xff;
 		if (!(s->status & BIT(EIOINTC_ENABLE_INT_ENCODE))) {
 			ipnum = count_trailing_zeros(ipnum);
 			ipnum = (ipnum >= 0 && ipnum < 4) ? ipnum : 0;
 		}
 
-		cpuid = s->coremap.reg_u8[irq];
+		cpuid = ((u8 *)s->coremap)[irq];
 		vcpu = kvm_get_vcpu_by_cpuid(s->kvm, cpuid);
 		if (!vcpu)
 			continue;
 
 		cpu = vcpu->vcpu_id;
-		if (test_bit(irq, (unsigned long *)s->coreisr.reg_u32[cpu]))
+		if (test_bit(irq, (unsigned long *)s->coreisr[cpu]))
 			__set_bit(irq, s->sw_coreisr[cpu][ipnum]);
 		else
 			__clear_bit(irq, s->sw_coreisr[cpu][ipnum]);
@@ -38,7 +38,7 @@ static void eiointc_update_irq(struct loongarch_eiointc *s, int irq, int level)
 	struct kvm_vcpu *vcpu;
 	struct kvm_interrupt vcpu_irq;
 
-	ipnum = s->ipmap.reg_u8[irq / 32];
+	ipnum = (s->ipmap >> (irq / 32 * 8)) & 0xff;
 	if (!(s->status & BIT(EIOINTC_ENABLE_INT_ENCODE))) {
 		ipnum = count_trailing_zeros(ipnum);
 		ipnum = (ipnum >= 0 && ipnum < 4) ? ipnum : 0;
@@ -53,13 +53,13 @@ static void eiointc_update_irq(struct loongarch_eiointc *s, int irq, int level)
 
 	if (level) {
 		/* if not enable return false */
-		if (!test_bit(irq, (unsigned long *)s->enable.reg_u32))
+		if (!test_bit(irq, (unsigned long *)s->enable))
 			return;
-		__set_bit(irq, (unsigned long *)s->coreisr.reg_u32[cpu]);
+		__set_bit(irq, (unsigned long *)s->coreisr[cpu]);
 		found = find_first_bit(s->sw_coreisr[cpu][ipnum], EIOINTC_IRQS);
 		__set_bit(irq, s->sw_coreisr[cpu][ipnum]);
 	} else {
-		__clear_bit(irq, (unsigned long *)s->coreisr.reg_u32[cpu]);
+		__clear_bit(irq, (unsigned long *)s->coreisr[cpu]);
 		__clear_bit(irq, s->sw_coreisr[cpu][ipnum]);
 		found = find_first_bit(s->sw_coreisr[cpu][ipnum], EIOINTC_IRQS);
 	}
@@ -94,7 +94,7 @@ static inline void eiointc_update_sw_coremap(struct loongarch_eiointc *s,
 		if (s->sw_coremap[irq + i] == cpu)
 			continue;
 
-		if (notify && test_bit(irq + i, (unsigned long *)s->isr.reg_u8)) {
+		if (notify && test_bit(irq + i, (unsigned long *)s->isr)) {
 			/* lower irq at old cpu and raise irq at new cpu */
 			eiointc_update_irq(s, irq + i, 0);
 			s->sw_coremap[irq + i] = cpu;
@@ -108,7 +108,7 @@ static inline void eiointc_update_sw_coremap(struct loongarch_eiointc *s,
 void eiointc_set_irq(struct loongarch_eiointc *s, int irq, int level)
 {
 	unsigned long flags;
-	unsigned long *isr = (unsigned long *)s->isr.reg_u8;
+	unsigned long *isr = (unsigned long *)s->isr;
 
 	spin_lock_irqsave(&s->lock, flags);
 	level ? __set_bit(irq, isr) : __clear_bit(irq, isr);
@@ -127,27 +127,27 @@ static int loongarch_eiointc_read(struct kvm_vcpu *vcpu, struct loongarch_eioint
 	switch (offset) {
 	case EIOINTC_NODETYPE_START ... EIOINTC_NODETYPE_END:
 		index = (offset - EIOINTC_NODETYPE_START) >> 3;
-		data = s->nodetype.reg_u64[index];
+		data = s->nodetype[index];
 		break;
 	case EIOINTC_IPMAP_START ... EIOINTC_IPMAP_END:
 		index = (offset - EIOINTC_IPMAP_START) >> 3;
-		data = s->ipmap.reg_u64;
+		data = s->ipmap;
 		break;
 	case EIOINTC_ENABLE_START ... EIOINTC_ENABLE_END:
 		index = (offset - EIOINTC_ENABLE_START) >> 3;
-		data = s->enable.reg_u64[index];
+		data = s->enable[index];
 		break;
 	case EIOINTC_BOUNCE_START ... EIOINTC_BOUNCE_END:
 		index = (offset - EIOINTC_BOUNCE_START) >> 3;
-		data = s->bounce.reg_u64[index];
+		data = s->bounce[index];
 		break;
 	case EIOINTC_COREISR_START ... EIOINTC_COREISR_END:
 		index = (offset - EIOINTC_COREISR_START) >> 3;
-		data = s->coreisr.reg_u64[vcpu->vcpu_id][index];
+		data = s->coreisr[vcpu->vcpu_id][index];
 		break;
 	case EIOINTC_COREMAP_START ... EIOINTC_COREMAP_END:
 		index = (offset - EIOINTC_COREMAP_START) >> 3;
-		data = s->coremap.reg_u64[index];
+		data = s->coremap[index];
 		break;
 	default:
 		ret = -EINVAL;
@@ -223,26 +223,26 @@ static int loongarch_eiointc_write(struct kvm_vcpu *vcpu,
 	switch (offset) {
 	case EIOINTC_NODETYPE_START ... EIOINTC_NODETYPE_END:
 		index = (offset - EIOINTC_NODETYPE_START) >> 3;
-		old = s->nodetype.reg_u64[index];
-		s->nodetype.reg_u64[index] = (old & ~mask) | data;
+		old = s->nodetype[index];
+		s->nodetype[index] = (old & ~mask) | data;
 		break;
 	case EIOINTC_IPMAP_START ... EIOINTC_IPMAP_END:
 		/*
 		 * ipmap cannot be set at runtime, can be set only at the beginning
 		 * of irqchip driver, need not update upper irq level
 		 */
-		old = s->ipmap.reg_u64;
-		s->ipmap.reg_u64 = (old & ~mask) | data;
+		old = s->ipmap;
+		s->ipmap = (old & ~mask) | data;
 		break;
 	case EIOINTC_ENABLE_START ... EIOINTC_ENABLE_END:
 		index = (offset - EIOINTC_ENABLE_START) >> 3;
-		old = s->enable.reg_u64[index];
-		s->enable.reg_u64[index] = (old & ~mask) | data;
+		old = s->enable[index];
+		s->enable[index] = (old & ~mask) | data;
 		/*
 		 * 1: enable irq.
 		 * update irq when isr is set.
 		 */
-		data = s->enable.reg_u64[index] & ~old & s->isr.reg_u64[index];
+		data = s->enable[index] & ~old & s->isr[index];
 		while (data) {
 			irq = __ffs(data);
 			eiointc_update_irq(s, irq + index * 64, 1);
@@ -252,7 +252,7 @@ static int loongarch_eiointc_write(struct kvm_vcpu *vcpu,
 		 * 0: disable irq.
 		 * update irq when isr is set.
 		 */
-		data = ~s->enable.reg_u64[index] & old & s->isr.reg_u64[index];
+		data = ~s->enable[index] & old & s->isr[index];
 		while (data) {
 			irq = __ffs(data);
 			eiointc_update_irq(s, irq + index * 64, 0);
@@ -262,16 +262,16 @@ static int loongarch_eiointc_write(struct kvm_vcpu *vcpu,
 	case EIOINTC_BOUNCE_START ... EIOINTC_BOUNCE_END:
 		/* do not emulate hw bounced irq routing */
 		index = (offset - EIOINTC_BOUNCE_START) >> 3;
-		old = s->bounce.reg_u64[index];
-		s->bounce.reg_u64[index] = (old & ~mask) | data;
+		old = s->bounce[index];
+		s->bounce[index] = (old & ~mask) | data;
 		break;
 	case EIOINTC_COREISR_START ... EIOINTC_COREISR_END:
 		index = (offset - EIOINTC_COREISR_START) >> 3;
 		/* use attrs to get current cpu index */
 		cpu = vcpu->vcpu_id;
-		old = s->coreisr.reg_u64[cpu][index];
+		old = s->coreisr[cpu][index];
 		/* write 1 to clear interrupt */
-		s->coreisr.reg_u64[cpu][index] = old & ~data;
+		s->coreisr[cpu][index] = old & ~data;
 		data &= old;
 		while (data) {
 			irq = __ffs(data);
@@ -281,9 +281,9 @@ static int loongarch_eiointc_write(struct kvm_vcpu *vcpu,
 		break;
 	case EIOINTC_COREMAP_START ... EIOINTC_COREMAP_END:
 		index = (offset - EIOINTC_COREMAP_START) >> 3;
-		old = s->coremap.reg_u64[index];
-		s->coremap.reg_u64[index] = (old & ~mask) | data;
-		data = s->coremap.reg_u64[index];
+		old = s->coremap[index];
+		s->coremap[index] = (old & ~mask) | data;
+		data = s->coremap[index];
 		eiointc_update_sw_coremap(s, index * 8, data, sizeof(data), true);
 		break;
 	default:
@@ -451,10 +451,10 @@ static int kvm_eiointc_ctrl_access(struct kvm_device *dev,
 		break;
 	case KVM_DEV_LOONGARCH_EXTIOI_CTRL_LOAD_FINISHED:
 		eiointc_set_sw_coreisr(s);
-		for (i = 0; i < (EIOINTC_IRQS / 4); i++) {
-			start_irq = i * 4;
+		for (i = 0; i < (EIOINTC_IRQS / 8); i++) {
+			start_irq = i * 8;
 			eiointc_update_sw_coremap(s, start_irq,
-					s->coremap.reg_u32[i], sizeof(u32), false);
+					s->coremap[i], sizeof(u64), false);
 		}
 		break;
 	default:
@@ -481,34 +481,34 @@ static int kvm_eiointc_regs_access(struct kvm_device *dev,
 	switch (addr) {
 	case EIOINTC_NODETYPE_START ... EIOINTC_NODETYPE_END:
 		offset = (addr - EIOINTC_NODETYPE_START) / 4;
-		p = &s->nodetype.reg_u32[offset];
+		p = s->nodetype + offset * 4;
 		break;
 	case EIOINTC_IPMAP_START ... EIOINTC_IPMAP_END:
 		offset = (addr - EIOINTC_IPMAP_START) / 4;
-		p = &s->ipmap.reg_u32[offset];
+		p = &s->ipmap + offset * 4;
 		break;
 	case EIOINTC_ENABLE_START ... EIOINTC_ENABLE_END:
 		offset = (addr - EIOINTC_ENABLE_START) / 4;
-		p = &s->enable.reg_u32[offset];
+		p = s->enable + offset * 4;
 		break;
 	case EIOINTC_BOUNCE_START ... EIOINTC_BOUNCE_END:
 		offset = (addr - EIOINTC_BOUNCE_START) / 4;
-		p = &s->bounce.reg_u32[offset];
+		p = s->bounce + offset * 4;
 		break;
 	case EIOINTC_ISR_START ... EIOINTC_ISR_END:
 		offset = (addr - EIOINTC_ISR_START) / 4;
-		p = &s->isr.reg_u32[offset];
+		p = s->isr + offset * 4;
 		break;
 	case EIOINTC_COREISR_START ... EIOINTC_COREISR_END:
 		if (cpu >= s->num_cpu)
 			return -EINVAL;
 
 		offset = (addr - EIOINTC_COREISR_START) / 4;
-		p = &s->coreisr.reg_u32[cpu][offset];
+		p = s->coreisr[cpu] + offset * 4;
 		break;
 	case EIOINTC_COREMAP_START ... EIOINTC_COREMAP_END:
 		offset = (addr - EIOINTC_COREMAP_START) / 4;
-		p = &s->coremap.reg_u32[offset];
+		p = s->coremap + offset * 4;
 		break;
 	default:
 		kvm_err("%s: unknown eiointc register, addr = %d\n", __func__, addr);

arch/loongarch/kvm/interrupt.c

@@ -21,6 +21,7 @@ static unsigned int priority_to_irq[EXCCODE_INT_NUM] = {
 	[INT_HWI5]	= CPU_IP5,
 	[INT_HWI6]	= CPU_IP6,
 	[INT_HWI7]	= CPU_IP7,
+	[INT_AVEC]	= CPU_AVEC,
 };
 
 static int kvm_irq_deliver(struct kvm_vcpu *vcpu, unsigned int priority)
@@ -31,6 +32,11 @@ static int kvm_irq_deliver(struct kvm_vcpu *vcpu, unsigned int priority)
 	if (priority < EXCCODE_INT_NUM)
 		irq = priority_to_irq[priority];
 
+	if (cpu_has_msgint && (priority == INT_AVEC)) {
+		set_gcsr_estat(irq);
+		return 1;
+	}
+
 	switch (priority) {
 	case INT_TI:
 	case INT_IPI:
@@ -58,6 +64,11 @@ static int kvm_irq_clear(struct kvm_vcpu *vcpu, unsigned int priority)
 	if (priority < EXCCODE_INT_NUM)
 		irq = priority_to_irq[priority];
 
+	if (cpu_has_msgint && (priority == INT_AVEC)) {
+		clear_gcsr_estat(irq);
+		return 1;
+	}
+
 	switch (priority) {
 	case INT_TI:
 	case INT_IPI:
@@ -83,10 +94,10 @@ void kvm_deliver_intr(struct kvm_vcpu *vcpu)
 	unsigned long *pending = &vcpu->arch.irq_pending;
 	unsigned long *pending_clr = &vcpu->arch.irq_clear;
 
-	for_each_set_bit(priority, pending_clr, INT_IPI + 1)
+	for_each_set_bit(priority, pending_clr, EXCCODE_INT_NUM)
 		kvm_irq_clear(vcpu, priority);
 
-	for_each_set_bit(priority, pending, INT_IPI + 1)
+	for_each_set_bit(priority, pending, EXCCODE_INT_NUM)
 		kvm_irq_deliver(vcpu, priority);
 }