Merge tag 'loongarch-kvm-6.10' of git://git.kernel.org/pub/scm/linux/kernel/git/chenhuacai/linux-loongson into HEAD

LoongArch KVM changes for v6.10

1. Add ParaVirt IPI support.
2. Add software breakpoint support.
3. Add mmio trace events support.

4 files changed, 263 insertions, 24 deletions

diff --git a/arch/loongarch/kvm/exit.c b/arch/loongarch/kvm/exit.c
index ed1d89d53e2e..c86e099af5ca 100644
--- a/arch/loongarch/kvm/exit.c
+++ b/arch/loongarch/kvm/exit.c
@@ -9,6 +9,7 @@
 #include <linux/module.h>
 #include <linux/preempt.h>
 #include <linux/vmalloc.h>
+#include <trace/events/kvm.h>
 #include <asm/fpu.h>
 #include <asm/inst.h>
 #include <asm/loongarch.h>
@@ -20,6 +21,46 @@
 #include <asm/kvm_vcpu.h>
 #include "trace.h"
 
+static int kvm_emu_cpucfg(struct kvm_vcpu *vcpu, larch_inst inst)
+{
+	int rd, rj;
+	unsigned int index;
+
+	if (inst.reg2_format.opcode != cpucfg_op)
+		return EMULATE_FAIL;
+
+	rd = inst.reg2_format.rd;
+	rj = inst.reg2_format.rj;
+	++vcpu->stat.cpucfg_exits;
+	index = vcpu->arch.gprs[rj];
+
+	/*
+	 * By LoongArch Reference Manual 2.2.10.5
+	 * Return value is 0 for undefined CPUCFG index
+	 *
+	 * Disable preemption since hw gcsr is accessed
+	 */
+	preempt_disable();
+	switch (index) {
+	case 0 ... (KVM_MAX_CPUCFG_REGS - 1):
+		vcpu->arch.gprs[rd] = vcpu->arch.cpucfg[index];
+		break;
+	case CPUCFG_KVM_SIG:
+		/* CPUCFG emulation between 0x40000000 -- 0x400000ff */
+		vcpu->arch.gprs[rd] = *(unsigned int *)KVM_SIGNATURE;
+		break;
+	case CPUCFG_KVM_FEATURE:
+		vcpu->arch.gprs[rd] = KVM_FEATURE_IPI;
+		break;
+	default:
+		vcpu->arch.gprs[rd] = 0;
+		break;
+	}
+	preempt_enable();
+
+	return EMULATE_DONE;
+}
+
 static unsigned long kvm_emu_read_csr(struct kvm_vcpu *vcpu, int csrid)
 {
 	unsigned long val = 0;
@@ -208,8 +249,6 @@ int kvm_emu_idle(struct kvm_vcpu *vcpu)
 
 static int kvm_trap_handle_gspr(struct kvm_vcpu *vcpu)
 {
-	int rd, rj;
-	unsigned int index;
 	unsigned long curr_pc;
 	larch_inst inst;
 	enum emulation_result er = EMULATE_DONE;
@@ -224,21 +263,7 @@ static int kvm_trap_handle_gspr(struct kvm_vcpu *vcpu)
 	er = EMULATE_FAIL;
 	switch (((inst.word >> 24) & 0xff)) {
 	case 0x0: /* CPUCFG GSPR */
-		if (inst.reg2_format.opcode == 0x1B) {
-			rd = inst.reg2_format.rd;
-			rj = inst.reg2_format.rj;
-			++vcpu->stat.cpucfg_exits;
-			index = vcpu->arch.gprs[rj];
-			er = EMULATE_DONE;
-			/*
-			 * By LoongArch Reference Manual 2.2.10.5
-			 * return value is 0 for undefined cpucfg index
-			 */
-			if (index < KVM_MAX_CPUCFG_REGS)
-				vcpu->arch.gprs[rd] = vcpu->arch.cpucfg[index];
-			else
-				vcpu->arch.gprs[rd] = 0;
-		}
+		er = kvm_emu_cpucfg(vcpu, inst);
 		break;
 	case 0x4: /* CSR{RD,WR,XCHG} GSPR */
 		er = kvm_handle_csr(vcpu, inst);
@@ -417,6 +442,8 @@ int kvm_emu_mmio_read(struct kvm_vcpu *vcpu, larch_inst inst)
 		vcpu->arch.io_gpr = rd;
 		run->mmio.is_write = 0;
 		vcpu->mmio_is_write = 0;
+		trace_kvm_mmio(KVM_TRACE_MMIO_READ_UNSATISFIED, run->mmio.len,
+				run->mmio.phys_addr, NULL);
 	} else {
 		kvm_err("Read not supported Inst=0x%08x @%lx BadVaddr:%#lx\n",
 			inst.word, vcpu->arch.pc, vcpu->arch.badv);
@@ -463,6 +490,9 @@ int kvm_complete_mmio_read(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		break;
 	}
 
+	trace_kvm_mmio(KVM_TRACE_MMIO_READ, run->mmio.len,
+			run->mmio.phys_addr, run->mmio.data);
+
 	return er;
 }
 
@@ -564,6 +594,8 @@ int kvm_emu_mmio_write(struct kvm_vcpu *vcpu, larch_inst inst)
 		run->mmio.is_write = 1;
 		vcpu->mmio_needed = 1;
 		vcpu->mmio_is_write = 1;
+		trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, run->mmio.len,
+				run->mmio.phys_addr, data);
 	} else {
 		vcpu->arch.pc = curr_pc;
 		kvm_err("Write not supported Inst=0x%08x @%lx BadVaddr:%#lx\n",
@@ -685,6 +717,90 @@ static int kvm_handle_lasx_disabled(struct kvm_vcpu *vcpu)
 	return RESUME_GUEST;
 }
 
+static int kvm_send_pv_ipi(struct kvm_vcpu *vcpu)
+{
+	unsigned int min, cpu, i;
+	unsigned long ipi_bitmap;
+	struct kvm_vcpu *dest;
+
+	min = kvm_read_reg(vcpu, LOONGARCH_GPR_A3);
+	for (i = 0; i < 2; i++, min += BITS_PER_LONG) {
+		ipi_bitmap = kvm_read_reg(vcpu, LOONGARCH_GPR_A1 + i);
+		if (!ipi_bitmap)
+			continue;
+
+		cpu = find_first_bit((void *)&ipi_bitmap, BITS_PER_LONG);
+		while (cpu < BITS_PER_LONG) {
+			dest = kvm_get_vcpu_by_cpuid(vcpu->kvm, cpu + min);
+			cpu = find_next_bit((void *)&ipi_bitmap, BITS_PER_LONG, cpu + 1);
+			if (!dest)
+				continue;
+
+			/* Send SWI0 to dest vcpu to emulate IPI interrupt */
+			kvm_queue_irq(dest, INT_SWI0);
+			kvm_vcpu_kick(dest);
+		}
+	}
+
+	return 0;
+}
+
+/*
+ * Hypercall emulation always return to guest, Caller should check retval.
+ */
+static void kvm_handle_service(struct kvm_vcpu *vcpu)
+{
+	unsigned long func = kvm_read_reg(vcpu, LOONGARCH_GPR_A0);
+	long ret;
+
+	switch (func) {
+	case KVM_HCALL_FUNC_IPI:
+		kvm_send_pv_ipi(vcpu);
+		ret = KVM_HCALL_SUCCESS;
+		break;
+	default:
+		ret = KVM_HCALL_INVALID_CODE;
+		break;
+	};
+
+	kvm_write_reg(vcpu, LOONGARCH_GPR_A0, ret);
+}
+
+static int kvm_handle_hypercall(struct kvm_vcpu *vcpu)
+{
+	int ret;
+	larch_inst inst;
+	unsigned int code;
+
+	inst.word = vcpu->arch.badi;
+	code = inst.reg0i15_format.immediate;
+	ret = RESUME_GUEST;
+
+	switch (code) {
+	case KVM_HCALL_SERVICE:
+		vcpu->stat.hypercall_exits++;
+		kvm_handle_service(vcpu);
+		break;
+	case KVM_HCALL_SWDBG:
+		/* KVM_HCALL_SWDBG only in effective when SW_BP is enabled */
+		if (vcpu->guest_debug & KVM_GUESTDBG_SW_BP_MASK) {
+			vcpu->run->exit_reason = KVM_EXIT_DEBUG;
+			ret = RESUME_HOST;
+			break;
+		}
+		fallthrough;
+	default:
+		/* Treat it as noop intruction, only set return value */
+		kvm_write_reg(vcpu, LOONGARCH_GPR_A0, KVM_HCALL_INVALID_CODE);
+		break;
+	}
+
+	if (ret == RESUME_GUEST)
+		update_pc(&vcpu->arch);
+
+	return ret;
+}
+
 /*
  * LoongArch KVM callback handling for unimplemented guest exiting
  */
@@ -716,6 +832,7 @@ static exit_handle_fn kvm_fault_tables[EXCCODE_INT_START] = {
 	[EXCCODE_LSXDIS]		= kvm_handle_lsx_disabled,
 	[EXCCODE_LASXDIS]		= kvm_handle_lasx_disabled,
 	[EXCCODE_GSPR]			= kvm_handle_gspr,
+	[EXCCODE_HVC]			= kvm_handle_hypercall,
 };
 
 int kvm_handle_fault(struct kvm_vcpu *vcpu, int fault)
diff --git a/arch/loongarch/kvm/trace.h b/arch/loongarch/kvm/trace.h
index c2484ad4cffa..1783397b1bc8 100644
--- a/arch/loongarch/kvm/trace.h
+++ b/arch/loongarch/kvm/trace.h
@@ -19,14 +19,16 @@ DECLARE_EVENT_CLASS(kvm_transition,
 	TP_PROTO(struct kvm_vcpu *vcpu),
 	TP_ARGS(vcpu),
 	TP_STRUCT__entry(
+		__field(unsigned int, vcpu_id)
 		__field(unsigned long, pc)
 	),
 
 	TP_fast_assign(
+		__entry->vcpu_id = vcpu->vcpu_id;
 		__entry->pc = vcpu->arch.pc;
 	),
 
-	TP_printk("PC: 0x%08lx", __entry->pc)
+	TP_printk("vcpu %u PC: 0x%08lx", __entry->vcpu_id, __entry->pc)
 );
 
 DEFINE_EVENT(kvm_transition, kvm_enter,
@@ -54,19 +56,22 @@ DECLARE_EVENT_CLASS(kvm_exit,
 	    TP_PROTO(struct kvm_vcpu *vcpu, unsigned int reason),
 	    TP_ARGS(vcpu, reason),
 	    TP_STRUCT__entry(
+			__field(unsigned int, vcpu_id)
 			__field(unsigned long, pc)
 			__field(unsigned int, reason)
 	    ),
 
 	    TP_fast_assign(
+			__entry->vcpu_id = vcpu->vcpu_id;
 			__entry->pc = vcpu->arch.pc;
 			__entry->reason = reason;
 	    ),
 
-	    TP_printk("[%s]PC: 0x%08lx",
-		      __print_symbolic(__entry->reason,
-				       kvm_trace_symbol_exit_types),
-		      __entry->pc)
+	    TP_printk("vcpu %u [%s] PC: 0x%08lx",
+			__entry->vcpu_id,
+			__print_symbolic(__entry->reason,
+				kvm_trace_symbol_exit_types),
+			__entry->pc)
 );
 
 DEFINE_EVENT(kvm_exit, kvm_exit_idle,
@@ -85,14 +90,17 @@ TRACE_EVENT(kvm_exit_gspr,
 	    TP_PROTO(struct kvm_vcpu *vcpu, unsigned int inst_word),
 	    TP_ARGS(vcpu, inst_word),
 	    TP_STRUCT__entry(
+			__field(unsigned int, vcpu_id)
 			__field(unsigned int, inst_word)
 	    ),
 
 	    TP_fast_assign(
+			__entry->vcpu_id = vcpu->vcpu_id;
 			__entry->inst_word = inst_word;
 	    ),
 
-	    TP_printk("Inst word: 0x%08x", __entry->inst_word)
+	    TP_printk("vcpu %u Inst word: 0x%08x", __entry->vcpu_id,
+			__entry->inst_word)
 );
 
 #define KVM_TRACE_AUX_SAVE		0
diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c
index 3a8779065f73..9e8030d45129 100644
--- a/arch/loongarch/kvm/vcpu.c
+++ b/arch/loongarch/kvm/vcpu.c
@@ -19,6 +19,7 @@ const struct _kvm_stats_desc kvm_vcpu_stats_desc[] = {
 	STATS_DESC_COUNTER(VCPU, idle_exits),
 	STATS_DESC_COUNTER(VCPU, cpucfg_exits),
 	STATS_DESC_COUNTER(VCPU, signal_exits),
+	STATS_DESC_COUNTER(VCPU, hypercall_exits)
 };
 
 const struct kvm_stats_header kvm_vcpu_stats_header = {
@@ -247,7 +248,101 @@ int kvm_arch_vcpu_ioctl_set_mpstate(struct kvm_vcpu *vcpu,
 int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu,
 					struct kvm_guest_debug *dbg)
 {
-	return -EINVAL;
+	if (dbg->control & ~KVM_GUESTDBG_VALID_MASK)
+		return -EINVAL;
+
+	if (dbg->control & KVM_GUESTDBG_ENABLE)
+		vcpu->guest_debug = dbg->control;
+	else
+		vcpu->guest_debug = 0;
+
+	return 0;
+}
+
+static inline int kvm_set_cpuid(struct kvm_vcpu *vcpu, u64 val)
+{
+	int cpuid;
+	struct kvm_phyid_map *map;
+	struct loongarch_csrs *csr = vcpu->arch.csr;
+
+	if (val >= KVM_MAX_PHYID)
+		return -EINVAL;
+
+	map = vcpu->kvm->arch.phyid_map;
+	cpuid = kvm_read_sw_gcsr(csr, LOONGARCH_CSR_CPUID);
+
+	spin_lock(&vcpu->kvm->arch.phyid_map_lock);
+	if ((cpuid < KVM_MAX_PHYID) && map->phys_map[cpuid].enabled) {
+		/* Discard duplicated CPUID set operation */
+		if (cpuid == val) {
+			spin_unlock(&vcpu->kvm->arch.phyid_map_lock);
+			return 0;
+		}
+
+		/*
+		 * CPUID is already set before
+		 * Forbid changing to a different CPUID at runtime
+		 */
+		spin_unlock(&vcpu->kvm->arch.phyid_map_lock);
+		return -EINVAL;
+	}
+
+	if (map->phys_map[val].enabled) {
+		/* Discard duplicated CPUID set operation */
+		if (vcpu == map->phys_map[val].vcpu) {
+			spin_unlock(&vcpu->kvm->arch.phyid_map_lock);
+			return 0;
+		}
+
+		/*
+		 * New CPUID is already set with other vcpu
+		 * Forbid sharing the same CPUID between different vcpus
+		 */
+		spin_unlock(&vcpu->kvm->arch.phyid_map_lock);
+		return -EINVAL;
+	}
+
+	kvm_write_sw_gcsr(csr, LOONGARCH_CSR_CPUID, val);
+	map->phys_map[val].enabled	= true;
+	map->phys_map[val].vcpu		= vcpu;
+	spin_unlock(&vcpu->kvm->arch.phyid_map_lock);
+
+	return 0;
+}
+
+static inline void kvm_drop_cpuid(struct kvm_vcpu *vcpu)
+{
+	int cpuid;
+	struct kvm_phyid_map *map;
+	struct loongarch_csrs *csr = vcpu->arch.csr;
+
+	map = vcpu->kvm->arch.phyid_map;
+	cpuid = kvm_read_sw_gcsr(csr, LOONGARCH_CSR_CPUID);
+
+	if (cpuid >= KVM_MAX_PHYID)
+		return;
+
+	spin_lock(&vcpu->kvm->arch.phyid_map_lock);
+	if (map->phys_map[cpuid].enabled) {
+		map->phys_map[cpuid].vcpu = NULL;
+		map->phys_map[cpuid].enabled = false;
+		kvm_write_sw_gcsr(csr, LOONGARCH_CSR_CPUID, KVM_MAX_PHYID);
+	}
+	spin_unlock(&vcpu->kvm->arch.phyid_map_lock);
+}
+
+struct kvm_vcpu *kvm_get_vcpu_by_cpuid(struct kvm *kvm, int cpuid)
+{
+	struct kvm_phyid_map *map;
+
+	if (cpuid >= KVM_MAX_PHYID)
+		return NULL;
+
+	map = kvm->arch.phyid_map;
+	if (!map->phys_map[cpuid].enabled)
+		return NULL;
+
+	return map->phys_map[cpuid].vcpu;
 }
 
 static int _kvm_getcsr(struct kvm_vcpu *vcpu, unsigned int id, u64 *val)
@@ -282,6 +377,9 @@ static int _kvm_setcsr(struct kvm_vcpu *vcpu, unsigned int id, u64 val)
 	if (get_gcsr_flag(id) & INVALID_GCSR)
 		return -EINVAL;
 
+	if (id == LOONGARCH_CSR_CPUID)
+		return kvm_set_cpuid(vcpu, val);
+
 	if (id == LOONGARCH_CSR_ESTAT) {
 		/* ESTAT IP0~IP7 inject through GINTC */
 		gintc = (val >> 2) & 0xff;
@@ -409,6 +507,9 @@ static int kvm_get_one_reg(struct kvm_vcpu *vcpu,
 		case KVM_REG_LOONGARCH_COUNTER:
 			*v = drdtime() + vcpu->kvm->arch.time_offset;
 			break;
+		case KVM_REG_LOONGARCH_DEBUG_INST:
+			*v = INSN_HVCL | KVM_HCALL_SWDBG;
+			break;
 		default:
 			ret = -EINVAL;
 			break;
@@ -924,6 +1025,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu)
 
 	/* Set cpuid */
 	kvm_write_sw_gcsr(csr, LOONGARCH_CSR_TMID, vcpu->vcpu_id);
+	kvm_write_sw_gcsr(csr, LOONGARCH_CSR_CPUID, KVM_MAX_PHYID);
 
 	/* Start with no pending virtual guest interrupts */
 	csr->csrs[LOONGARCH_CSR_GINTC] = 0;
@@ -942,6 +1044,7 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu)
 
 	hrtimer_cancel(&vcpu->arch.swtimer);
 	kvm_mmu_free_memory_cache(&vcpu->arch.mmu_page_cache);
+	kvm_drop_cpuid(vcpu);
 	kfree(vcpu->arch.csr);
 
 	/*
diff --git a/arch/loongarch/kvm/vm.c b/arch/loongarch/kvm/vm.c
index 0a37f6fa8f2d..6b2e4f66ad26 100644
--- a/arch/loongarch/kvm/vm.c
+++ b/arch/loongarch/kvm/vm.c
@@ -30,6 +30,14 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	if (!kvm->arch.pgd)
 		return -ENOMEM;
 
+	kvm->arch.phyid_map = kvzalloc(sizeof(struct kvm_phyid_map), GFP_KERNEL_ACCOUNT);
+	if (!kvm->arch.phyid_map) {
+		free_page((unsigned long)kvm->arch.pgd);
+		kvm->arch.pgd = NULL;
+		return -ENOMEM;
+	}
+	spin_lock_init(&kvm->arch.phyid_map_lock);
+
 	kvm_init_vmcs(kvm);
 	kvm->arch.gpa_size = BIT(cpu_vabits - 1);
 	kvm->arch.root_level = CONFIG_PGTABLE_LEVELS - 1;
@@ -52,6 +60,8 @@ void kvm_arch_destroy_vm(struct kvm *kvm)
 	kvm_destroy_vcpus(kvm);
 	free_page((unsigned long)kvm->arch.pgd);
 	kvm->arch.pgd = NULL;
+	kvfree(kvm->arch.phyid_map);
+	kvm->arch.phyid_map = NULL;
 }
 
 int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
@@ -66,6 +76,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
 	case KVM_CAP_IMMEDIATE_EXIT:
 	case KVM_CAP_IOEVENTFD:
 	case KVM_CAP_MP_STATE:
+	case KVM_CAP_SET_GUEST_DEBUG:
 		r = 1;
 		break;
 	case KVM_CAP_NR_VCPUS: