* [RFC PATCH v2 1/4] KVM: x86: Split the APIC from the rest of IRQCHIP.
@ 2015-05-15 2:52 Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 2/4] KVM: x86: Add KVM exit for IOAPIC EOIs Steve Rutherford
` (2 more replies)
0 siblings, 3 replies; 4+ messages in thread
From: Steve Rutherford @ 2015-05-15 2:52 UTC (permalink / raw)
To: kvm
First patch in a series which enables the relocation of the
PIC/IOAPIC/PIT to userspace.
Adds capability KVM_CAP_SPLIT_IRQCHIP;
KVM_CAP_SPLIT_IRQCHIP enables the construction of LAPICs without the
rest of the irqchip.
Compile tested for x86.
Signed-off-by: Steve Rutherford <srutherford@google.com>
Suggested-by: Andrew Honig <ahonig@google.com>
---
Documentation/virtual/kvm/api.txt | 15 ++++++++++++
arch/powerpc/kvm/irq.h | 5 ++++
arch/s390/kvm/irq.h | 4 ++++
arch/x86/include/asm/kvm_host.h | 2 ++
arch/x86/kvm/assigned-dev.c | 4 ++--
arch/x86/kvm/irq.c | 6 ++---
arch/x86/kvm/irq.h | 11 +++++++++
arch/x86/kvm/irq_comm.c | 7 ++++++
arch/x86/kvm/lapic.c | 13 +++++++----
arch/x86/kvm/mmu.c | 2 +-
arch/x86/kvm/svm.c | 4 ++--
arch/x86/kvm/vmx.c | 12 +++++-----
arch/x86/kvm/x86.c | 49 +++++++++++++++++++++++++++------------
include/kvm/arm_vgic.h | 1 +
include/linux/kvm_host.h | 1 +
include/uapi/linux/kvm.h | 1 +
virt/kvm/irqchip.c | 2 +-
17 files changed, 104 insertions(+), 35 deletions(-)
diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index 6955444..9a43d42 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -2979,6 +2979,7 @@ len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0
and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq),
which is the maximum number of possibly pending cpu-local interrupts.
+
5. The kvm_run structure
------------------------
@@ -3575,6 +3576,20 @@ struct {
KVM handlers should exit to userspace with rc = -EREMOTE.
+7.5 KVM_SPLIT_IRQCHIP
+
+Capability: KVM_CAP_SPLIT_IRQCHIP
+Architectures: x86
+Type: VM ioctl
+Parameters: None
+Returns: 0 on success, -1 on error
+
+Create a local apic for each processor in the kernel. This differs from
+KVM_CREATE_IRQCHIP in that it only creates the local apic; it creates neither
+the ioapic nor the pic in the kernel. Also, enables in kernel routing of
+interrupt requests. Fails if VCPU has already been created, or if the irqchip is
+already in the kernel.
+
8. Other capabilities.
----------------------
diff --git a/arch/powerpc/kvm/irq.h b/arch/powerpc/kvm/irq.h
index 5a9a10b..5e6fa06 100644
--- a/arch/powerpc/kvm/irq.h
+++ b/arch/powerpc/kvm/irq.h
@@ -17,4 +17,9 @@ static inline int irqchip_in_kernel(struct kvm *kvm)
return ret;
}
+static inline int lapic_in_kernel(struct kvm *kvm)
+{
+ return irqchip_in_kernel(kvm);
+}
+
#endif
diff --git a/arch/s390/kvm/irq.h b/arch/s390/kvm/irq.h
index d98e415..db876c3 100644
--- a/arch/s390/kvm/irq.h
+++ b/arch/s390/kvm/irq.h
@@ -19,4 +19,8 @@ static inline int irqchip_in_kernel(struct kvm *kvm)
return 1;
}
+static inline int lapic_in_kernel(struct kvm *kvm)
+{
+ return irqchip_in_kernel(kvm);
+}
#endif
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index bbb8f4e..3ddc134 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -638,6 +638,8 @@ struct kvm_arch {
bool boot_vcpu_runs_old_kvmclock;
u64 disabled_quirks;
+
+ bool irqchip_split;
};
struct kvm_vm_stat {
diff --git a/arch/x86/kvm/assigned-dev.c b/arch/x86/kvm/assigned-dev.c
index d090ecf..1237e92 100644
--- a/arch/x86/kvm/assigned-dev.c
+++ b/arch/x86/kvm/assigned-dev.c
@@ -291,7 +291,7 @@ static int kvm_deassign_irq(struct kvm *kvm,
{
unsigned long guest_irq_type, host_irq_type;
- if (!irqchip_in_kernel(kvm))
+ if (!lapic_in_kernel(kvm))
return -EINVAL;
/* no irq assignment to deassign */
if (!assigned_dev->irq_requested_type)
@@ -568,7 +568,7 @@ static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
struct kvm_assigned_dev_kernel *match;
unsigned long host_irq_type, guest_irq_type;
- if (!irqchip_in_kernel(kvm))
+ if (!lapic_in_kernel(kvm))
return r;
mutex_lock(&kvm->lock);
diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
index a1ec6a50..706e47a 100644
--- a/arch/x86/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -57,7 +57,7 @@ static int kvm_cpu_has_extint(struct kvm_vcpu *v)
*/
int kvm_cpu_has_injectable_intr(struct kvm_vcpu *v)
{
- if (!irqchip_in_kernel(v->kvm))
+ if (!lapic_in_kernel(v->kvm))
return v->arch.interrupt.pending;
if (kvm_cpu_has_extint(v))
@@ -75,7 +75,7 @@ int kvm_cpu_has_injectable_intr(struct kvm_vcpu *v)
*/
int kvm_cpu_has_interrupt(struct kvm_vcpu *v)
{
- if (!irqchip_in_kernel(v->kvm))
+ if (!lapic_in_kernel(v->kvm))
return v->arch.interrupt.pending;
if (kvm_cpu_has_extint(v))
@@ -103,7 +103,7 @@ int kvm_cpu_get_interrupt(struct kvm_vcpu *v)
{
int vector;
- if (!irqchip_in_kernel(v->kvm))
+ if (!lapic_in_kernel(v->kvm))
return v->arch.interrupt.nr;
vector = kvm_cpu_get_extint(v);
diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
index ad68c73..e46abf3 100644
--- a/arch/x86/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -92,6 +92,17 @@ static inline int irqchip_in_kernel(struct kvm *kvm)
return ret;
}
+static inline int irqchip_split(struct kvm *kvm)
+{
+ return kvm->arch.irqchip_split;
+}
+
+static inline int lapic_in_kernel(struct kvm *kvm)
+{
+ return irqchip_split(kvm) || irqchip_in_kernel(kvm);
+}
+
+
void kvm_pic_reset(struct kvm_kpic_state *s);
void kvm_inject_pending_timer_irqs(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/kvm/irq_comm.c b/arch/x86/kvm/irq_comm.c
index 9efff9e..f43c59a 100644
--- a/arch/x86/kvm/irq_comm.c
+++ b/arch/x86/kvm/irq_comm.c
@@ -328,3 +328,10 @@ int kvm_setup_default_irq_routing(struct kvm *kvm)
return kvm_set_irq_routing(kvm, default_routing,
ARRAY_SIZE(default_routing), 0);
}
+
+static const struct kvm_irq_routing_entry empty_routing[] = {};
+
+int kvm_setup_empty_irq_routing(struct kvm *kvm)
+{
+ return kvm_set_irq_routing(kvm, empty_routing, 0, 0);
+}
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index dc5b57b..bc392a6 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -209,7 +209,8 @@ out:
if (old)
kfree_rcu(old, rcu);
- kvm_vcpu_request_scan_ioapic(kvm);
+ if (!irqchip_split(kvm))
+ kvm_vcpu_request_scan_ioapic(kvm);
}
static inline void apic_set_spiv(struct kvm_lapic *apic, u32 val)
@@ -1819,7 +1820,8 @@ void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu,
kvm_x86_ops->hwapic_isr_update(vcpu->kvm,
apic_find_highest_isr(apic));
kvm_make_request(KVM_REQ_EVENT, vcpu);
- kvm_rtc_eoi_tracking_restore_one(vcpu);
+ if (!irqchip_split(vcpu->kvm))
+ kvm_rtc_eoi_tracking_restore_one(vcpu);
}
void __kvm_migrate_apic_timer(struct kvm_vcpu *vcpu)
@@ -1902,7 +1904,8 @@ static void apic_sync_pv_eoi_to_guest(struct kvm_vcpu *vcpu,
/* Cache not set: could be safe but we don't bother. */
apic->highest_isr_cache == -1 ||
/* Need EOI to update ioapic. */
- kvm_ioapic_handles_vector(vcpu->kvm, apic->highest_isr_cache)) {
+ kvm_ioapic_handles_vector(vcpu->kvm, apic->highest_isr_cache) ||
+ irqchip_split(vcpu->kvm)) {
/*
* PV EOI was disabled by apic_sync_pv_eoi_from_guest
* so we need not do anything here.
@@ -1958,7 +1961,7 @@ int kvm_x2apic_msr_write(struct kvm_vcpu *vcpu, u32 msr, u64 data)
struct kvm_lapic *apic = vcpu->arch.apic;
u32 reg = (msr - APIC_BASE_MSR) << 4;
- if (!irqchip_in_kernel(vcpu->kvm) || !apic_x2apic_mode(apic))
+ if (!lapic_in_kernel(vcpu->kvm) || !apic_x2apic_mode(apic))
return 1;
if (reg == APIC_ICR2)
@@ -1975,7 +1978,7 @@ int kvm_x2apic_msr_read(struct kvm_vcpu *vcpu, u32 msr, u64 *data)
struct kvm_lapic *apic = vcpu->arch.apic;
u32 reg = (msr - APIC_BASE_MSR) << 4, low, high = 0;
- if (!irqchip_in_kernel(vcpu->kvm) || !apic_x2apic_mode(apic))
+ if (!lapic_in_kernel(vcpu->kvm) || !apic_x2apic_mode(apic))
return 1;
if (reg == APIC_DFR || reg == APIC_ICR2) {
diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c
index b78e83f..e5bf6db 100644
--- a/arch/x86/kvm/mmu.c
+++ b/arch/x86/kvm/mmu.c
@@ -3465,7 +3465,7 @@ static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn)
static bool can_do_async_pf(struct kvm_vcpu *vcpu)
{
- if (unlikely(!irqchip_in_kernel(vcpu->kvm) ||
+ if (unlikely(!lapic_in_kernel(vcpu->kvm) ||
kvm_event_needs_reinjection(vcpu)))
return false;
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 8954d7a..7ac9ec2 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -3054,7 +3054,7 @@ static int cr8_write_interception(struct vcpu_svm *svm)
u8 cr8_prev = kvm_get_cr8(&svm->vcpu);
/* instruction emulation calls kvm_set_cr8() */
r = cr_interception(svm);
- if (irqchip_in_kernel(svm->vcpu.kvm))
+ if (lapic_in_kernel(svm->vcpu.kvm))
return r;
if (cr8_prev <= kvm_get_cr8(&svm->vcpu))
return r;
@@ -3295,7 +3295,7 @@ static int interrupt_window_interception(struct vcpu_svm *svm)
* If the user space waits to inject interrupts, exit as soon as
* possible
*/
- if (!irqchip_in_kernel(svm->vcpu.kvm) &&
+ if (!lapic_in_kernel(svm->vcpu.kvm) &&
kvm_run->request_interrupt_window &&
!kvm_cpu_has_interrupt(&svm->vcpu)) {
kvm_run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index bcb61b0..a53747f 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -948,7 +948,7 @@ static inline bool cpu_has_vmx_tpr_shadow(void)
static inline bool vm_need_tpr_shadow(struct kvm *kvm)
{
- return (cpu_has_vmx_tpr_shadow()) && (irqchip_in_kernel(kvm));
+ return (cpu_has_vmx_tpr_shadow()) && lapic_in_kernel(kvm);
}
static inline bool cpu_has_secondary_exec_ctrls(void)
@@ -1064,7 +1064,7 @@ static inline bool cpu_has_vmx_ple(void)
static inline bool vm_need_virtualize_apic_accesses(struct kvm *kvm)
{
- return flexpriority_enabled && irqchip_in_kernel(kvm);
+ return flexpriority_enabled && lapic_in_kernel(kvm);
}
static inline bool cpu_has_vmx_vpid(void)
@@ -4341,7 +4341,7 @@ static void vmx_disable_intercept_msr_write_x2apic(u32 msr)
static int vmx_vm_has_apicv(struct kvm *kvm)
{
- return enable_apicv && irqchip_in_kernel(kvm);
+ return enable_apicv && lapic_in_kernel(kvm);
}
static int vmx_complete_nested_posted_interrupt(struct kvm_vcpu *vcpu)
@@ -5317,7 +5317,7 @@ static int handle_cr(struct kvm_vcpu *vcpu)
u8 cr8 = (u8)val;
err = kvm_set_cr8(vcpu, cr8);
kvm_complete_insn_gp(vcpu, err);
- if (irqchip_in_kernel(vcpu->kvm))
+ if (lapic_in_kernel(vcpu->kvm))
return 1;
if (cr8_prev <= cr8)
return 1;
@@ -5534,7 +5534,7 @@ static int handle_interrupt_window(struct kvm_vcpu *vcpu)
* If the user space waits to inject interrupts, exit as soon as
* possible
*/
- if (!irqchip_in_kernel(vcpu->kvm) &&
+ if (!lapic_in_kernel(vcpu->kvm) &&
vcpu->run->request_interrupt_window &&
!kvm_cpu_has_interrupt(vcpu)) {
vcpu->run->exit_reason = KVM_EXIT_IRQ_WINDOW_OPEN;
@@ -9419,7 +9419,7 @@ static void prepare_vmcs02(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12)
/* vmcs12's VM_ENTRY_LOAD_IA32_EFER and VM_ENTRY_IA32E_MODE are
* emulated by vmx_set_efer(), below.
*/
- vm_entry_controls_init(vmx,
+ vm_entry_controls_init(vmx,
(vmcs12->vm_entry_controls & ~VM_ENTRY_LOAD_IA32_EFER &
~VM_ENTRY_IA32E_MODE) |
(vmcs_config.vmentry_ctrl & ~VM_ENTRY_IA32E_MODE));
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index cde5d61..fc347fa 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -782,7 +782,7 @@ int kvm_set_cr8(struct kvm_vcpu *vcpu, unsigned long cr8)
{
if (cr8 & CR8_RESERVED_BITS)
return 1;
- if (irqchip_in_kernel(vcpu->kvm))
+ if (lapic_in_kernel(vcpu->kvm))
kvm_lapic_set_tpr(vcpu, cr8);
else
vcpu->arch.cr8 = cr8;
@@ -792,7 +792,7 @@ EXPORT_SYMBOL_GPL(kvm_set_cr8);
unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu)
{
- if (irqchip_in_kernel(vcpu->kvm))
+ if (lapic_in_kernel(vcpu->kvm))
return kvm_lapic_get_cr8(vcpu);
else
return vcpu->arch.cr8;
@@ -2800,6 +2800,7 @@ int kvm_vm_ioctl_check_extension(struct kvm *kvm, long ext)
case KVM_CAP_TSC_DEADLINE_TIMER:
case KVM_CAP_ENABLE_CAP_VM:
case KVM_CAP_DISABLE_QUIRKS:
+ case KVM_CAP_SPLIT_IRQCHIP:
#ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
case KVM_CAP_ASSIGN_DEV_IRQ:
case KVM_CAP_PCI_2_3:
@@ -3002,7 +3003,7 @@ static int kvm_vcpu_ioctl_interrupt(struct kvm_vcpu *vcpu,
{
if (irq->irq >= KVM_NR_INTERRUPTS)
return -EINVAL;
- if (irqchip_in_kernel(vcpu->kvm))
+ if (lapic_in_kernel(vcpu->kvm))
return -ENXIO;
kvm_queue_interrupt(vcpu, irq->irq, false);
@@ -3480,7 +3481,7 @@ long kvm_arch_vcpu_ioctl(struct file *filp,
struct kvm_vapic_addr va;
r = -EINVAL;
- if (!irqchip_in_kernel(vcpu->kvm))
+ if (!lapic_in_kernel(vcpu->kvm))
goto out;
r = -EFAULT;
if (copy_from_user(&va, argp, sizeof va))
@@ -3838,7 +3839,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_dirty_log *log)
int kvm_vm_ioctl_irq_line(struct kvm *kvm, struct kvm_irq_level *irq_event,
bool line_status)
{
- if (!irqchip_in_kernel(kvm))
+ if (!lapic_in_kernel(kvm))
return -ENXIO;
irq_event->status = kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
@@ -3860,6 +3861,23 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
kvm->arch.disabled_quirks = cap->args[0];
r = 0;
break;
+ case KVM_CAP_SPLIT_IRQCHIP: {
+ mutex_lock(&kvm->lock);
+ r = -EEXIST;
+ if (lapic_in_kernel(kvm))
+ goto split_irqchip_unlock;
+ r = -EINVAL;
+ if (atomic_read(&kvm->online_vcpus))
+ goto split_irqchip_unlock;
+ r = kvm_setup_empty_irq_routing(kvm);
+ if (r)
+ goto split_irqchip_unlock;
+ kvm->arch.irqchip_split = true;
+ r = 0;
+split_irqchip_unlock:
+ mutex_unlock(&kvm->lock);
+ break;
+ }
default:
r = -EINVAL;
break;
@@ -4128,6 +4146,7 @@ long kvm_arch_vm_ioctl(struct file *filp,
r = kvm_vm_ioctl_enable_cap(kvm, &cap);
break;
}
+
default:
r = kvm_vm_ioctl_assigned_device(kvm, ioctl, arg);
}
@@ -5893,7 +5912,7 @@ void kvm_arch_exit(void)
int kvm_vcpu_halt(struct kvm_vcpu *vcpu)
{
++vcpu->stat.halt_exits;
- if (irqchip_in_kernel(vcpu->kvm)) {
+ if (lapic_in_kernel(vcpu->kvm)) {
vcpu->arch.mp_state = KVM_MP_STATE_HALTED;
return 1;
} else {
@@ -6060,7 +6079,7 @@ static int emulator_fix_hypercall(struct x86_emulate_ctxt *ctxt)
*/
static int dm_request_for_irq_injection(struct kvm_vcpu *vcpu)
{
- return (!irqchip_in_kernel(vcpu->kvm) && !kvm_cpu_has_interrupt(vcpu) &&
+ return (!lapic_in_kernel(vcpu->kvm) && !kvm_cpu_has_interrupt(vcpu) &&
vcpu->run->request_interrupt_window &&
kvm_arch_interrupt_allowed(vcpu));
}
@@ -6072,7 +6091,7 @@ static void post_kvm_run_save(struct kvm_vcpu *vcpu)
kvm_run->if_flag = (kvm_get_rflags(vcpu) & X86_EFLAGS_IF) != 0;
kvm_run->cr8 = kvm_get_cr8(vcpu);
kvm_run->apic_base = kvm_get_apic_base(vcpu);
- if (irqchip_in_kernel(vcpu->kvm))
+ if (lapic_in_kernel(vcpu->kvm))
kvm_run->ready_for_interrupt_injection = 1;
else
kvm_run->ready_for_interrupt_injection =
@@ -6219,7 +6238,7 @@ void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu)
{
struct page *page = NULL;
- if (!irqchip_in_kernel(vcpu->kvm))
+ if (!lapic_in_kernel(vcpu->kvm))
return;
if (!kvm_x86_ops->set_apic_access_page_addr)
@@ -6255,7 +6274,7 @@ void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
{
int r;
- bool req_int_win = !irqchip_in_kernel(vcpu->kvm) &&
+ bool req_int_win = !lapic_in_kernel(vcpu->kvm) &&
vcpu->run->request_interrupt_window;
bool req_immediate_exit = false;
@@ -6644,7 +6663,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
}
/* re-sync apic's tpr */
- if (!irqchip_in_kernel(vcpu->kvm)) {
+ if (!lapic_in_kernel(vcpu->kvm)) {
if (kvm_set_cr8(vcpu, kvm_run->cr8) != 0) {
r = -EINVAL;
goto out;
@@ -7340,7 +7359,7 @@ void kvm_arch_check_processor_compat(void *rtn)
bool kvm_vcpu_compatible(struct kvm_vcpu *vcpu)
{
- return irqchip_in_kernel(vcpu->kvm) == (vcpu->arch.apic != NULL);
+ return lapic_in_kernel(vcpu->kvm) == (vcpu->arch.apic != NULL);
}
struct static_key kvm_no_apic_vcpu __read_mostly;
@@ -7356,7 +7375,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
vcpu->arch.pv.pv_unhalted = false;
vcpu->arch.emulate_ctxt.ops = &emulate_ops;
- if (!irqchip_in_kernel(kvm) || kvm_vcpu_is_reset_bsp(vcpu))
+ if (!lapic_in_kernel(kvm) || kvm_vcpu_is_reset_bsp(vcpu))
vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE;
else
vcpu->arch.mp_state = KVM_MP_STATE_UNINITIALIZED;
@@ -7374,7 +7393,7 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
if (r < 0)
goto fail_free_pio_data;
- if (irqchip_in_kernel(kvm)) {
+ if (lapic_in_kernel(kvm)) {
r = kvm_create_lapic(vcpu);
if (r < 0)
goto fail_mmu_destroy;
@@ -7437,7 +7456,7 @@ void kvm_arch_vcpu_uninit(struct kvm_vcpu *vcpu)
kvm_mmu_destroy(vcpu);
srcu_read_unlock(&vcpu->kvm->srcu, idx);
free_page((unsigned long)vcpu->arch.pio_data);
- if (!irqchip_in_kernel(vcpu->kvm))
+ if (!lapic_in_kernel(vcpu->kvm))
static_key_slow_dec(&kvm_no_apic_vcpu);
}
diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
index 133ea00..ffe1f4e 100644
--- a/include/kvm/arm_vgic.h
+++ b/include/kvm/arm_vgic.h
@@ -329,6 +329,7 @@ int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu);
int kvm_vgic_vcpu_active_irq(struct kvm_vcpu *vcpu);
#define irqchip_in_kernel(k) (!!((k)->arch.vgic.in_kernel))
+#define lapic_in_kernel(k) (irqchip_in_kernel(k))
#define vgic_initialized(k) (!!((k)->arch.vgic.nr_cpus))
#define vgic_ready(k) ((k)->arch.vgic.ready)
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 87fd74a..277b7a1 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -934,6 +934,7 @@ static inline int mmu_notifier_retry(struct kvm *kvm, unsigned long mmu_seq)
#endif
int kvm_setup_default_irq_routing(struct kvm *kvm);
+int kvm_setup_empty_irq_routing(struct kvm *kvm);
int kvm_set_irq_routing(struct kvm *kvm,
const struct kvm_irq_routing_entry *entries,
unsigned nr,
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 75bd9f7..1e6f6c3 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -815,6 +815,7 @@ struct kvm_ppc_smmu_info {
#define KVM_CAP_S390_IRQ_STATE 114
#define KVM_CAP_PPC_HWRNG 115
#define KVM_CAP_DISABLE_QUIRKS 116
+#define KVM_CAP_SPLIT_IRQCHIP 117
#ifdef KVM_CAP_IRQ_ROUTING
diff --git a/virt/kvm/irqchip.c b/virt/kvm/irqchip.c
index 1d56a90..8aaceed 100644
--- a/virt/kvm/irqchip.c
+++ b/virt/kvm/irqchip.c
@@ -73,7 +73,7 @@ int kvm_send_userspace_msi(struct kvm *kvm, struct kvm_msi *msi)
{
struct kvm_kernel_irq_routing_entry route;
- if (!irqchip_in_kernel(kvm) || msi->flags != 0)
+ if (!lapic_in_kernel(kvm) || msi->flags != 0)
return -EINVAL;
route.msi.address_lo = msi->address_lo;
--
2.2.0.rc0.207.ga3a616c
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [RFC PATCH v2 2/4] KVM: x86: Add KVM exit for IOAPIC EOIs
2015-05-15 2:52 [RFC PATCH v2 1/4] KVM: x86: Split the APIC from the rest of IRQCHIP Steve Rutherford
@ 2015-05-15 2:52 ` Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 3/4] KVM: x86: Add EOI exit bitmap inference Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 4/4] KVM: x86: Add support for local interrupt requests from userspace Steve Rutherford
2 siblings, 0 replies; 4+ messages in thread
From: Steve Rutherford @ 2015-05-15 2:52 UTC (permalink / raw)
To: kvm
Adds KVM_EXIT_IOAPIC_EOI which allows the kernel to EOI
level-triggered IOAPIC interrupts.
Uses a per VCPU exit bitmap to decide whether or not the IOAPIC needs
to be informed (which is identical to the EOI_EXIT_BITMAP field used
by modern x86 processors, but can also be used to elide kvm IOAPIC EOI
exits on older processors).
[Note: A prototype using ResampleFDs found that decoupling the EOI
from the VCPU's thread made it possible for the VCPU to not see a
recent EOI after reentering the guest. This does not match real
hardware.]
Compile tested for Intel x86.
Signed-off-by: Steve Rutherford <srutherford@google.com>
---
Documentation/virtual/kvm/api.txt | 10 ++++++++++
arch/x86/include/asm/kvm_host.h | 3 +++
arch/x86/kvm/lapic.c | 9 +++++++++
arch/x86/kvm/x86.c | 11 +++++++++++
include/linux/kvm_host.h | 1 +
include/uapi/linux/kvm.h | 5 +++++
6 files changed, 39 insertions(+)
diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index 9a43d42..6ab2a3f7 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -3271,6 +3271,16 @@ Valid values for 'type' are:
*/
__u64 kvm_valid_regs;
__u64 kvm_dirty_regs;
+
+ /* KVM_EXIT_IOAPIC_EOI */
+ struct {
+ __u8 vector;
+ } eoi;
+
+Indicates that an eoi of a level triggered IOAPIC interrupt on vector has
+occurred, which should be handled by the userspace IOAPIC. Triggers when
+the Irqchip has been split between userspace and the kernel.
+
union {
struct kvm_sync_regs regs;
char padding[1024];
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 3ddc134..b1978f1 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -539,6 +539,9 @@ struct kvm_vcpu_arch {
struct {
bool pv_unhalted;
} pv;
+
+ u64 eoi_exit_bitmaps[4];
+ int pending_ioapic_eoi;
};
struct kvm_lpage_info {
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index bc392a6..42fada6f 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -860,6 +860,15 @@ int kvm_apic_compare_prio(struct kvm_vcpu *vcpu1, struct kvm_vcpu *vcpu2)
static void kvm_ioapic_send_eoi(struct kvm_lapic *apic, int vector)
{
+ if (irqchip_split(apic->vcpu->kvm)) {
+ if (test_bit(vector,
+ (void *) apic->vcpu->arch.eoi_exit_bitmaps)) {
+ apic->vcpu->arch.pending_ioapic_eoi = vector;
+ kvm_make_request(KVM_REQ_IOAPIC_EOI_EXIT, apic->vcpu);
+ }
+ return;
+ }
+
if (kvm_ioapic_handles_vector(apic->vcpu->kvm, vector)) {
int trigger_mode;
if (apic_test_vector(vector, apic->regs + APIC_TMR))
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index fc347fa..e7377df 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -6324,6 +6324,17 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
kvm_handle_pmu_event(vcpu);
if (kvm_check_request(KVM_REQ_PMI, vcpu))
kvm_deliver_pmi(vcpu);
+ if (kvm_check_request(KVM_REQ_IOAPIC_EOI_EXIT, vcpu)) {
+ BUG_ON(vcpu->arch.pending_ioapic_eoi > 255);
+ if (test_bit(vcpu->arch.pending_ioapic_eoi,
+ (void *) vcpu->arch.eoi_exit_bitmaps)) {
+ vcpu->run->exit_reason = KVM_EXIT_IOAPIC_EOI;
+ vcpu->run->eoi.vector =
+ vcpu->arch.pending_ioapic_eoi;
+ r = 0;
+ goto out;
+ }
+ }
if (kvm_check_request(KVM_REQ_SCAN_IOAPIC, vcpu))
vcpu_scan_ioapic(vcpu);
if (kvm_check_request(KVM_REQ_APIC_PAGE_RELOAD, vcpu))
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 277b7a1..cef20ad 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -134,6 +134,7 @@ static inline bool is_error_page(struct page *page)
#define KVM_REQ_ENABLE_IBS 23
#define KVM_REQ_DISABLE_IBS 24
#define KVM_REQ_APIC_PAGE_RELOAD 25
+#define KVM_REQ_IOAPIC_EOI_EXIT 26
#define KVM_USERSPACE_IRQ_SOURCE_ID 0
#define KVM_IRQFD_RESAMPLE_IRQ_SOURCE_ID 1
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 1e6f6c3..826a08d 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -183,6 +183,7 @@ struct kvm_s390_skeys {
#define KVM_EXIT_EPR 23
#define KVM_EXIT_SYSTEM_EVENT 24
#define KVM_EXIT_S390_STSI 25
+#define KVM_EXIT_IOAPIC_EOI 26
/* For KVM_EXIT_INTERNAL_ERROR */
/* Emulate instruction failed. */
@@ -329,6 +330,10 @@ struct kvm_run {
__u8 sel1;
__u16 sel2;
} s390_stsi;
+ /* KVM_EXIT_IOAPIC_EOI */
+ struct {
+ __u8 vector;
+ } eoi;
/* Fix the size of the union. */
char padding[256];
};
--
2.2.0.rc0.207.ga3a616c
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [RFC PATCH v2 3/4] KVM: x86: Add EOI exit bitmap inference
2015-05-15 2:52 [RFC PATCH v2 1/4] KVM: x86: Split the APIC from the rest of IRQCHIP Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 2/4] KVM: x86: Add KVM exit for IOAPIC EOIs Steve Rutherford
@ 2015-05-15 2:52 ` Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 4/4] KVM: x86: Add support for local interrupt requests from userspace Steve Rutherford
2 siblings, 0 replies; 4+ messages in thread
From: Steve Rutherford @ 2015-05-15 2:52 UTC (permalink / raw)
To: kvm
In order to support a userspace IOAPIC interacting with an in kernel
APIC, the EOI exit bitmaps need to be configurable.
If the IOAPIC is in userspace (i.e. the irqchip has been split), the
EOI exit bitmaps will be set whenever the GSI Routes are configured.
In particular, for the low MSI routes are reservable for userspace
IOAPICs. For these MSI routes, the EOI Exit bit corresponding to the
destination vector of the route will be set for the destination VCPU.
The intention is for the userspace IOAPICs to use the reservable MSI
routes to inject interrupts into the guest.
This is a slight abuse of the notion of an MSI Route, given that MSIs
classically bypass the IOAPIC. It might be worthwhile to add an
additional route type to improve clarity.
Compile tested for Intel x86.
Signed-off-by: Steve Rutherford <srutherford@google.com>
---
arch/x86/include/asm/kvm_host.h | 1 +
arch/x86/kvm/ioapic.c | 16 ++++++++++++++++
arch/x86/kvm/ioapic.h | 2 ++
arch/x86/kvm/lapic.c | 3 +--
arch/x86/kvm/x86.c | 30 ++++++++++++++++++++++--------
include/linux/kvm_host.h | 9 +++++++++
virt/kvm/irqchip.c | 37 +++++++++++++++++++++++++++++++++++++
7 files changed, 88 insertions(+), 10 deletions(-)
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index b1978f1..16a6187 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -643,6 +643,7 @@ struct kvm_arch {
u64 disabled_quirks;
bool irqchip_split;
+ u8 nr_reserved_ioapic_pins;
};
struct kvm_vm_stat {
diff --git a/arch/x86/kvm/ioapic.c b/arch/x86/kvm/ioapic.c
index 856f791..fb5281b 100644
--- a/arch/x86/kvm/ioapic.c
+++ b/arch/x86/kvm/ioapic.c
@@ -672,3 +672,19 @@ int kvm_set_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state)
spin_unlock(&ioapic->lock);
return 0;
}
+
+void kvm_arch_irq_routing_update(struct kvm *kvm)
+{
+ struct kvm_ioapic *ioapic = kvm->arch.vioapic;
+
+ if (ioapic)
+ return;
+ if (!lapic_in_kernel(kvm))
+ return;
+ kvm_make_scan_ioapic_request(kvm);
+}
+
+u8 kvm_arch_nr_userspace_ioapic_pins(struct kvm *kvm)
+{
+ return kvm->arch.nr_reserved_ioapic_pins;
+}
diff --git a/arch/x86/kvm/ioapic.h b/arch/x86/kvm/ioapic.h
index ca0b0b4..3af349c 100644
--- a/arch/x86/kvm/ioapic.h
+++ b/arch/x86/kvm/ioapic.h
@@ -9,6 +9,7 @@ struct kvm;
struct kvm_vcpu;
#define IOAPIC_NUM_PINS KVM_IOAPIC_NUM_PINS
+#define MAX_NR_RESERVED_IOAPIC_PINS 48
#define IOAPIC_VERSION_ID 0x11 /* IOAPIC version */
#define IOAPIC_EDGE_TRIG 0
#define IOAPIC_LEVEL_TRIG 1
@@ -123,4 +124,5 @@ int kvm_set_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state);
void kvm_ioapic_scan_entry(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap,
u32 *tmr);
+void kvm_scan_ioapic_routes(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap);
#endif
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 42fada6f..befd54a 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -209,8 +209,7 @@ out:
if (old)
kfree_rcu(old, rcu);
- if (!irqchip_split(kvm))
- kvm_vcpu_request_scan_ioapic(kvm);
+ kvm_make_scan_ioapic_request(kvm);
}
static inline void apic_set_spiv(struct kvm_lapic *apic, u32 val)
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index e7377df..405d0d3 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -3864,15 +3864,20 @@ static int kvm_vm_ioctl_enable_cap(struct kvm *kvm,
case KVM_CAP_SPLIT_IRQCHIP: {
mutex_lock(&kvm->lock);
r = -EEXIST;
- if (lapic_in_kernel(kvm))
+ if (irqchip_in_kernel(kvm))
goto split_irqchip_unlock;
r = -EINVAL;
- if (atomic_read(&kvm->online_vcpus))
- goto split_irqchip_unlock;
- r = kvm_setup_empty_irq_routing(kvm);
- if (r)
+ if (cap->args[0] > MAX_NR_RESERVED_IOAPIC_PINS)
goto split_irqchip_unlock;
- kvm->arch.irqchip_split = true;
+ if (!irqchip_split(kvm)) {
+ if (atomic_read(&kvm->online_vcpus))
+ goto split_irqchip_unlock;
+ r = kvm_setup_empty_irq_routing(kvm);
+ if (r)
+ goto split_irqchip_unlock;
+ kvm->arch.irqchip_split = true;
+ }
+ kvm->arch.nr_reserved_ioapic_pins = cap->args[0];
r = 0;
split_irqchip_unlock:
mutex_unlock(&kvm->lock);
@@ -6335,8 +6340,17 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
goto out;
}
}
- if (kvm_check_request(KVM_REQ_SCAN_IOAPIC, vcpu))
- vcpu_scan_ioapic(vcpu);
+ if (kvm_check_request(KVM_REQ_SCAN_IOAPIC, vcpu)) {
+ if (irqchip_split(vcpu->kvm)) {
+ memset(vcpu->arch.eoi_exit_bitmaps, 0, 32);
+ kvm_scan_ioapic_routes(
+ vcpu, vcpu->arch.eoi_exit_bitmaps);
+ kvm_x86_ops->load_eoi_exitmap(
+ vcpu, vcpu->arch.eoi_exit_bitmaps);
+
+ } else
+ vcpu_scan_ioapic(vcpu);
+ }
if (kvm_check_request(KVM_REQ_APIC_PAGE_RELOAD, vcpu))
kvm_vcpu_reload_apic_access_page(vcpu);
}
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index cef20ad..93bd490 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -438,10 +438,19 @@ void vcpu_put(struct kvm_vcpu *vcpu);
#ifdef __KVM_HAVE_IOAPIC
void kvm_vcpu_request_scan_ioapic(struct kvm *kvm);
+void kvm_arch_irq_routing_update(struct kvm *kvm);
+u8 kvm_arch_nr_userspace_ioapic_pins(struct kvm *kvm);
#else
static inline void kvm_vcpu_request_scan_ioapic(struct kvm *kvm)
{
}
+static inline void kvm_arch_irq_routing_update(struct kvm *kvm)
+{
+}
+static inline u8 kvm_arch_nr_userspace_ioapic_pins(struct kvm *kvm)
+{
+ return 0;
+}
#endif
#ifdef CONFIG_HAVE_KVM_IRQFD
diff --git a/virt/kvm/irqchip.c b/virt/kvm/irqchip.c
index 8aaceed..208fdd3 100644
--- a/virt/kvm/irqchip.c
+++ b/virt/kvm/irqchip.c
@@ -203,6 +203,8 @@ int kvm_set_irq_routing(struct kvm *kvm,
kvm_irq_routing_update(kvm);
mutex_unlock(&kvm->irq_lock);
+ kvm_arch_irq_routing_update(kvm);
+
synchronize_srcu_expedited(&kvm->irq_srcu);
new = old;
@@ -212,3 +214,38 @@ out:
kfree(new);
return r;
}
+
+void kvm_scan_ioapic_routes(struct kvm_vcpu *vcpu, u64 *eoi_exit_bitmap)
+{
+ struct kvm *kvm = vcpu->kvm;
+ struct kvm_kernel_irq_routing_entry *entry;
+ struct kvm_irq_routing_table *table;
+ u32 i, nr_ioapic_pins;
+ int idx;
+
+ /* kvm->irq_routing must be read after clearing
+ * KVM_SCAN_IOAPIC. */
+ smp_mb();
+ idx = srcu_read_lock(&kvm->irq_srcu);
+ table = kvm->irq_routing;
+ nr_ioapic_pins = min_t(u32, table->nr_rt_entries,
+ kvm_arch_nr_userspace_ioapic_pins(kvm));
+ for (i = 0; i < nr_ioapic_pins; ++i) {
+ hlist_for_each_entry(entry, &table->map[i], link) {
+ u32 dest_id, dest_mode;
+
+ if (entry->type != KVM_IRQ_ROUTING_MSI)
+ continue;
+ dest_id = (entry->msi.address_lo >> 12) & 0xff;
+ dest_mode = (entry->msi.address_lo >> 2) & 0x1;
+ if (kvm_apic_match_dest(vcpu, NULL, 0, dest_id,
+ dest_mode)) {
+ u32 vector = entry->msi.data & 0xff;
+
+ __set_bit(vector,
+ (unsigned long *) eoi_exit_bitmap);
+ }
+ }
+ }
+ srcu_read_unlock(&kvm->irq_srcu, idx);
+}
--
2.2.0.rc0.207.ga3a616c
^ permalink raw reply related [flat|nested] 4+ messages in thread
* [RFC PATCH v2 4/4] KVM: x86: Add support for local interrupt requests from userspace
2015-05-15 2:52 [RFC PATCH v2 1/4] KVM: x86: Split the APIC from the rest of IRQCHIP Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 2/4] KVM: x86: Add KVM exit for IOAPIC EOIs Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 3/4] KVM: x86: Add EOI exit bitmap inference Steve Rutherford
@ 2015-05-15 2:52 ` Steve Rutherford
2 siblings, 0 replies; 4+ messages in thread
From: Steve Rutherford @ 2015-05-15 2:52 UTC (permalink / raw)
To: kvm
In order to enable userspace PIC support, the userspace PIC needs to
be able to inject local interrupt requests.
This adds the ioctl KVM_REQUEST_PIC_INJECTION and kvm exit
KVM_EXIT_GET_EXTINT.
The vm ioctl KVM_REQUEST_PIC_INJECTION makes a KVM_REQ_EVENT request
on the BSP, which causes the BSP to exit to userspace to fetch the
vector of the underlying external interrupt, which the BSP then
injects into the guest. This matches the PIC spec, and is necessary to
boot Windows.
Compiles for x86.
Has not undergone incremental testing since previous version.
Signed-off-by: Steve Rutherford <srutherford@google.com>
---
Documentation/virtual/kvm/api.txt | 9 ++++++
arch/x86/include/asm/kvm_host.h | 2 ++
arch/x86/kvm/irq.c | 22 +++++++++++++--
arch/x86/kvm/lapic.c | 7 +++++
arch/x86/kvm/lapic.h | 2 ++
arch/x86/kvm/x86.c | 59 +++++++++++++++++++++++++++++++++++++--
include/uapi/linux/kvm.h | 7 +++++
7 files changed, 103 insertions(+), 5 deletions(-)
diff --git a/Documentation/virtual/kvm/api.txt b/Documentation/virtual/kvm/api.txt
index 6ab2a3f7..b5d90cb 100644
--- a/Documentation/virtual/kvm/api.txt
+++ b/Documentation/virtual/kvm/api.txt
@@ -2979,6 +2979,15 @@ len must be a multiple of sizeof(struct kvm_s390_irq). It must be > 0
and it must not exceed (max_vcpus + 32) * sizeof(struct kvm_s390_irq),
which is the maximum number of possibly pending cpu-local interrupts.
+4.96 KVM_REQUEST_PIC_INJECTION
+
+Capability: KVM_CAP_SPLIT_IRQCHIP
+Type: VM ioctl
+Parameters: none
+Returns: 0 on success, -1 on error.
+
+Informs the kernel that userspace has a pending external interrupt.
+
5. The kvm_run structure
------------------------
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 16a6187..2d5390e 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -542,6 +542,8 @@ struct kvm_vcpu_arch {
u64 eoi_exit_bitmaps[4];
int pending_ioapic_eoi;
+ bool userspace_extint_available;
+ int pending_external_vector;
};
struct kvm_lpage_info {
diff --git a/arch/x86/kvm/irq.c b/arch/x86/kvm/irq.c
index 706e47a..1270b2a 100644
--- a/arch/x86/kvm/irq.c
+++ b/arch/x86/kvm/irq.c
@@ -38,12 +38,25 @@ int kvm_cpu_has_pending_timer(struct kvm_vcpu *vcpu)
EXPORT_SYMBOL(kvm_cpu_has_pending_timer);
/*
+ * check if there is a pending userspace external interrupt
+ */
+static int pending_userspace_extint(struct kvm_vcpu *v)
+{
+ return v->arch.userspace_extint_available ||
+ v->arch.pending_external_vector != -1;
+}
+
+/*
* check if there is pending interrupt from
* non-APIC source without intack.
*/
static int kvm_cpu_has_extint(struct kvm_vcpu *v)
{
- if (kvm_apic_accept_pic_intr(v))
+ u8 accept = kvm_apic_accept_pic_intr(v);
+
+ if (accept && irqchip_split(v->kvm))
+ return pending_userspace_extint(v);
+ else if (accept)
return pic_irqchip(v->kvm)->output; /* PIC */
else
return 0;
@@ -91,7 +104,12 @@ EXPORT_SYMBOL_GPL(kvm_cpu_has_interrupt);
*/
static int kvm_cpu_get_extint(struct kvm_vcpu *v)
{
- if (kvm_cpu_has_extint(v))
+ if (irqchip_split(v->kvm) && kvm_cpu_has_extint(v)) {
+ int vector = v->arch.pending_external_vector;
+
+ v->arch.pending_external_vector = -1;
+ return vector;
+ } else if (kvm_cpu_has_extint(v))
return kvm_pic_read_irq(v->kvm); /* PIC */
return -1;
}
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index befd54a..95e0ded 100644
--- a/arch/x86/kvm/lapic.c
+++ b/arch/x86/kvm/lapic.c
@@ -2086,3 +2086,10 @@ void kvm_lapic_init(void)
jump_label_rate_limit(&apic_hw_disabled, HZ);
jump_label_rate_limit(&apic_sw_disabled, HZ);
}
+
+void kvm_request_pic_injection(struct kvm_vcpu *vcpu)
+{
+ vcpu->arch.userspace_extint_available = true;
+ kvm_make_request(KVM_REQ_EVENT, vcpu);
+ kvm_vcpu_kick(vcpu);
+}
diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h
index 71b150c..7831e4d 100644
--- a/arch/x86/kvm/lapic.h
+++ b/arch/x86/kvm/lapic.h
@@ -63,6 +63,8 @@ int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq,
unsigned long *dest_map);
int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type);
+void kvm_request_pic_injection(struct kvm_vcpu *vcpu);
+
bool kvm_irq_delivery_to_apic_fast(struct kvm *kvm, struct kvm_lapic *src,
struct kvm_lapic_irq *irq, int *r, unsigned long *dest_map);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 405d0d3..1dc2877 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -65,6 +65,8 @@
#include <asm/pvclock.h>
#include <asm/div64.h>
+#define GET_VECTOR_FROM_USERSPACE 1
+
#define MAX_IO_MSRS 256
#define KVM_MAX_MCE_BANKS 32
#define KVM_MCE_CAP_SUPPORTED (MCG_CTL_P | MCG_SER_P)
@@ -4151,6 +4153,30 @@ long kvm_arch_vm_ioctl(struct file *filp,
r = kvm_vm_ioctl_enable_cap(kvm, &cap);
break;
}
+ case KVM_REQUEST_PIC_INJECTION: {
+ int i;
+ struct kvm_vcpu *vcpu;
+ struct kvm_vcpu *bsp_vcpu = NULL;
+
+ mutex_lock(&kvm->lock);
+ r = -EEXIST;
+ if (!irqchip_split(kvm))
+ goto out;
+ kvm_for_each_vcpu(i, vcpu, kvm) {
+ if (kvm_vcpu_is_reset_bsp(vcpu)) {
+ bsp_vcpu = vcpu;
+ continue;
+ }
+ }
+ r = -EINVAL;
+ if (bsp_vcpu == NULL)
+ goto interrupt_unlock;
+ kvm_request_pic_injection(bsp_vcpu);
+ r = 0;
+interrupt_unlock:
+ mutex_unlock(&kvm->lock);
+ break;
+ }
default:
r = kvm_vm_ioctl_assigned_device(kvm, ioctl, arg);
@@ -6128,6 +6154,17 @@ static void update_cr8_intercept(struct kvm_vcpu *vcpu)
kvm_x86_ops->update_cr8_intercept(vcpu, tpr, max_irr);
}
+static int must_fetch_userspace_extint(struct kvm_vcpu *vcpu)
+{
+ if (vcpu->arch.userspace_extint_available &&
+ kvm_apic_accept_pic_intr(vcpu)) {
+ vcpu->arch.userspace_extint_available = false;
+ return true;
+ } else
+ return false;
+
+}
+
static int inject_pending_event(struct kvm_vcpu *vcpu, bool req_int_win)
{
int r;
@@ -6192,7 +6229,12 @@ static int inject_pending_event(struct kvm_vcpu *vcpu, bool req_int_win)
return r;
}
if (kvm_x86_ops->interrupt_allowed(vcpu)) {
- kvm_queue_interrupt(vcpu, kvm_cpu_get_interrupt(vcpu),
+ if (irqchip_split(vcpu->kvm) &&
+ must_fetch_userspace_extint(vcpu)) {
+ return GET_VECTOR_FROM_USERSPACE;
+ }
+ kvm_queue_interrupt(vcpu,
+ kvm_cpu_get_interrupt(vcpu),
false);
kvm_x86_ops->set_irq(vcpu);
}
@@ -6356,13 +6398,19 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu)
}
if (kvm_check_request(KVM_REQ_EVENT, vcpu) || req_int_win) {
+ int event;
kvm_apic_accept_events(vcpu);
if (vcpu->arch.mp_state == KVM_MP_STATE_INIT_RECEIVED) {
r = 1;
goto out;
}
-
- if (inject_pending_event(vcpu, req_int_win) != 0)
+ event = inject_pending_event(vcpu, req_int_win);
+ if (event == GET_VECTOR_FROM_USERSPACE) {
+ vcpu->run->exit_reason = KVM_EXIT_GET_EXTINT;
+ kvm_make_request(KVM_REQ_EVENT, vcpu);
+ r = 0;
+ goto out;
+ } else if (event != 0)
req_immediate_exit = true;
/* enable NMI/IRQ window open exits if needed */
else if (vcpu->arch.nmi_pending)
@@ -6679,6 +6727,9 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run)
if (vcpu->sigset_active)
sigprocmask(SIG_SETMASK, &vcpu->sigset, &sigsaved);
+ if (vcpu->run->exit_reason == KVM_EXIT_GET_EXTINT)
+ vcpu->arch.pending_external_vector = vcpu->run->extint.vector;
+
if (unlikely(vcpu->arch.mp_state == KVM_MP_STATE_UNINITIALIZED)) {
kvm_vcpu_block(vcpu);
kvm_apic_accept_events(vcpu);
@@ -7455,6 +7506,8 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu)
kvm_async_pf_hash_reset(vcpu);
kvm_pmu_init(vcpu);
+ vcpu->arch.pending_external_vector = -1;
+
return 0;
fail_free_wbinvd_dirty_mask:
free_cpumask_var(vcpu->arch.wbinvd_dirty_mask);
diff --git a/include/uapi/linux/kvm.h b/include/uapi/linux/kvm.h
index 826a08d..0cf7ed6 100644
--- a/include/uapi/linux/kvm.h
+++ b/include/uapi/linux/kvm.h
@@ -184,6 +184,7 @@ struct kvm_s390_skeys {
#define KVM_EXIT_SYSTEM_EVENT 24
#define KVM_EXIT_S390_STSI 25
#define KVM_EXIT_IOAPIC_EOI 26
+#define KVM_EXIT_GET_EXTINT 27
/* For KVM_EXIT_INTERNAL_ERROR */
/* Emulate instruction failed. */
@@ -334,6 +335,10 @@ struct kvm_run {
struct {
__u8 vector;
} eoi;
+ /* KVM_EXIT_GET_EXTINT */
+ struct {
+ __u8 vector;
+ } extint;
/* Fix the size of the union. */
char padding[256];
};
@@ -1206,6 +1211,8 @@ struct kvm_s390_ucas_mapping {
/* Available with KVM_CAP_S390_IRQ_STATE */
#define KVM_S390_SET_IRQ_STATE _IOW(KVMIO, 0xb5, struct kvm_s390_irq_state)
#define KVM_S390_GET_IRQ_STATE _IOW(KVMIO, 0xb6, struct kvm_s390_irq_state)
+/* Available with KVM_CAP_SPLIT_IRQCHIP */
+#define KVM_REQUEST_PIC_INJECTION _IO(KVMIO, 0xb7)
#define KVM_DEV_ASSIGN_ENABLE_IOMMU (1 << 0)
#define KVM_DEV_ASSIGN_PCI_2_3 (1 << 1)
--
2.2.0.rc0.207.ga3a616c
^ permalink raw reply related [flat|nested] 4+ messages in thread
end of thread, other threads:[~2015-05-15 2:52 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2015-05-15 2:52 [RFC PATCH v2 1/4] KVM: x86: Split the APIC from the rest of IRQCHIP Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 2/4] KVM: x86: Add KVM exit for IOAPIC EOIs Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 3/4] KVM: x86: Add EOI exit bitmap inference Steve Rutherford
2015-05-15 2:52 ` [RFC PATCH v2 4/4] KVM: x86: Add support for local interrupt requests from userspace Steve Rutherford
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox