KVM: X86: Implement "send IPI" hypercall

Using hypercall to send IPIs by one vmexit instead of one by one for
xAPIC/x2APIC physical mode and one vmexit per-cluster for x2APIC cluster
mode. Intel guest can enter x2apic cluster mode when interrupt remmaping
is enabled in qemu, however, latest AMD EPYC still just supports xapic
mode which can get great improvement by Exit-less IPIs. This patchset
lets a guest send multicast IPIs, with at most 128 destinations per
hypercall in 64-bit mode and 64 vCPUs per hypercall in 32-bit mode.

Hardware: Xeon Skylake 2.5GHz, 2 sockets, 40 cores, 80 threads, the VM
is 80 vCPUs, IPI microbenchmark(https://lkml.org/lkml/2017/12/19/141):

x2apic cluster mode, vanilla

 Dry-run:                         0,            2392199 ns
 Self-IPI:                  6907514,           15027589 ns
 Normal IPI:              223910476,          251301666 ns
 Broadcast IPI:                   0,         9282161150 ns
 Broadcast lock:                  0,         8812934104 ns

x2apic cluster mode, pv-ipi

 Dry-run:                         0,            2449341 ns
 Self-IPI:                  6720360,           15028732 ns
 Normal IPI:              228643307,          255708477 ns
 Broadcast IPI:                   0,         7572293590 ns  => 22% performance boost
 Broadcast lock:                  0,         8316124651 ns

x2apic physical mode, vanilla

 Dry-run:                         0,            3135933 ns
 Self-IPI:                  8572670,           17901757 ns
 Normal IPI:              226444334,          255421709 ns
 Broadcast IPI:                   0,        19845070887 ns
 Broadcast lock:                  0,        19827383656 ns

x2apic physical mode, pv-ipi

 Dry-run:                         0,            2446381 ns
 Self-IPI:                  6788217,           15021056 ns
 Normal IPI:              219454441,          249583458 ns
 Broadcast IPI:                   0,         7806540019 ns  => 154% performance boost
 Broadcast lock:                  0,         9143618799 ns

Cc: Paolo Bonzini <[email protected]>
Cc: Radim Krčmář <[email protected]>
Cc: Vitaly Kuznetsov <[email protected]>
Signed-off-by: Wanpeng Li <[email protected]>
Signed-off-by: Paolo Bonzini <[email protected]>

Author: Wanpeng Li

Documentation/virtual/kvm/cpuid.txt

@@ -62,6 +62,10 @@ KVM_FEATURE_ASYNC_PF_VMEXIT        ||    10 || paravirtualized async PF VM exit
                                    ||       || can be enabled by setting bit 2
                                    ||       || when writing to msr 0x4b564d02
 ------------------------------------------------------------------------------
+KVM_FEATURE_PV_SEND_IPI            ||    11 || guest checks this feature bit
+                                   ||       || before using paravirtualized
+                                   ||       || send IPIs.
+------------------------------------------------------------------------------
 KVM_FEATURE_CLOCKSOURCE_STABLE_BIT ||    24 || host will warn if no guest-side
                                    ||       || per-cpu warps are expected in
                                    ||       || kvmclock.

Documentation/virtual/kvm/hypercalls.txt

@@ -121,3 +121,23 @@ compute the CLOCK_REALTIME for its clock, at the same instant.
 
 Returns KVM_EOPNOTSUPP if the host does not use TSC clocksource,
 or if clock type is different than KVM_CLOCK_PAIRING_WALLCLOCK.
+
+6. KVM_HC_SEND_IPI
+------------------------
+Architecture: x86
+Status: active
+Purpose: Send IPIs to multiple vCPUs.
+
+a0: lower part of the bitmap of destination APIC IDs
+a1: higher part of the bitmap of destination APIC IDs
+a2: the lowest APIC ID in bitmap
+a3: APIC ICR
+
+The hypercall lets a guest send multicast IPIs, with at most 128
+128 destinations per hypercall in 64-bit mode and 64 vCPUs per
+hypercall in 32-bit mode.  The destinations are represented by a
+bitmap contained in the first two arguments (a0 and a1). Bit 0 of
+a0 corresponds to the APIC ID in the third argument (a2), bit 1
+corresponds to the APIC ID a2+1, and so on.
+
+Returns the number of CPUs to which the IPIs were delivered successfully.

arch/x86/include/asm/kvm_host.h

@@ -1457,6 +1457,10 @@ int kvm_cpu_get_interrupt(struct kvm_vcpu *v);
 void kvm_vcpu_reset(struct kvm_vcpu *vcpu, bool init_event);
 void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu);
 
+int kvm_pv_send_ipi(struct kvm *kvm, unsigned long ipi_bitmap_low,
+    		    unsigned long ipi_bitmap_high, int min,
+		    unsigned long icr, int op_64_bit);
+
 void kvm_define_shared_msr(unsigned index, u32 msr);
 int kvm_set_shared_msr(unsigned index, u64 val, u64 mask);
 

arch/x86/kvm/cpuid.c

@@ -621,7 +621,8 @@ static inline int __do_cpuid_ent(struct kvm_cpuid_entry2 *entry, u32 function,
 			     (1 << KVM_FEATURE_CLOCKSOURCE_STABLE_BIT) |
 			     (1 << KVM_FEATURE_PV_UNHALT) |
 			     (1 << KVM_FEATURE_PV_TLB_FLUSH) |
-			     (1 << KVM_FEATURE_ASYNC_PF_VMEXIT);
+			     (1 << KVM_FEATURE_ASYNC_PF_VMEXIT) |
+			     (1 << KVM_FEATURE_PV_SEND_IPI);
 
 		if (sched_info_on())
 			entry->eax |= (1 << KVM_FEATURE_STEAL_TIME);

arch/x86/kvm/lapic.c

@@ -547,6 +547,46 @@ int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq,
 			irq->level, irq->trig_mode, dest_map);
 }
 
+int kvm_pv_send_ipi(struct kvm *kvm, unsigned long ipi_bitmap_low,
+    		    unsigned long ipi_bitmap_high, int min,
+		    unsigned long icr, int op_64_bit)
+{
+	int i;
+	struct kvm_apic_map *map;
+	struct kvm_vcpu *vcpu;
+	struct kvm_lapic_irq irq = {0};
+	int cluster_size = op_64_bit ? 64 : 32;
+	int count = 0;
+
+	irq.vector = icr & APIC_VECTOR_MASK;
+	irq.delivery_mode = icr & APIC_MODE_MASK;
+	irq.level = (icr & APIC_INT_ASSERT) != 0;
+	irq.trig_mode = icr & APIC_INT_LEVELTRIG;
+
+	if (icr & APIC_DEST_MASK)
+		return -KVM_EINVAL;
+	if (icr & APIC_SHORT_MASK)
+		return -KVM_EINVAL;
+
+	rcu_read_lock();
+	map = rcu_dereference(kvm->arch.apic_map);
+
+	/* Bits above cluster_size are masked in the caller.  */
+	for_each_set_bit(i, &ipi_bitmap_low, BITS_PER_LONG) {
+		vcpu = map->phys_map[min + i]->vcpu;
+		count += kvm_apic_set_irq(vcpu, &irq, NULL);
+	}
+
+	min += cluster_size;
+	for_each_set_bit(i, &ipi_bitmap_high, BITS_PER_LONG) {
+		vcpu = map->phys_map[min + i]->vcpu;
+		count += kvm_apic_set_irq(vcpu, &irq, NULL);
+	}
+
+	rcu_read_unlock();
+	return count;
+}
+
 static int pv_eoi_put_user(struct kvm_vcpu *vcpu, u8 val)
 {
 

arch/x86/kvm/x86.c

@@ -6802,6 +6802,9 @@ int kvm_emulate_hypercall(struct kvm_vcpu *vcpu)
 	case KVM_HC_CLOCK_PAIRING:
 		ret = kvm_pv_clock_pairing(vcpu, a0, a1);
 		break;
+	case KVM_HC_SEND_IPI:
+		ret = kvm_pv_send_ipi(vcpu->kvm, a0, a1, a2, a3, op_64_bit);
+		break;
 #endif
 	default:
 		ret = -KVM_ENOSYS;

include/uapi/linux/kvm_para.h

@@ -13,6 +13,7 @@
 /* Return values for hypercalls */
 #define KVM_ENOSYS		1000
 #define KVM_EFAULT		EFAULT
+#define KVM_EINVAL		EINVAL
 #define KVM_E2BIG		E2BIG
 #define KVM_EPERM		EPERM
 #define KVM_EOPNOTSUPP		95