Merge tag 'kvm-arm-for-4-7-take2' of git://git.kernel.org/pub/scm/linux/kernel/git/kvmarm/kvmarm into kvm-next

KVM/ARM Changes for v4.7 take 2

"The GIC is dead; Long live the GIC"

This set of changes include the new vgic, which is a reimplementation of
our horribly broken legacy vgic implementation.  The two implementations
will live side-by-side (with the new being the configured default) for
one kernel release and then we'll remove it.

Also fixes a non-critical issue with virtual abort injection to guests.

arch/arm/include/asm/kvm_host.h

@@ -41,6 +41,8 @@
 
 #define KVM_MAX_VCPUS VGIC_V2_MAX_CPUS
 
+#define KVM_REQ_VCPU_EXIT	8
+
 u32 *kvm_vcpu_reg(struct kvm_vcpu *vcpu, u8 reg_num, u32 mode);
 int __attribute_const__ kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
@@ -226,6 +228,10 @@ static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
 
 struct kvm_vcpu *kvm_arm_get_running_vcpu(void);
 struct kvm_vcpu __percpu **kvm_get_running_vcpus(void);
+void kvm_arm_halt_guest(struct kvm *kvm);
+void kvm_arm_resume_guest(struct kvm *kvm);
+void kvm_arm_halt_vcpu(struct kvm_vcpu *vcpu);
+void kvm_arm_resume_vcpu(struct kvm_vcpu *vcpu);
 
 int kvm_arm_copy_coproc_indices(struct kvm_vcpu *vcpu, u64 __user *uindices);
 unsigned long kvm_arm_num_coproc_regs(struct kvm_vcpu *vcpu);

arch/arm/include/asm/kvm_mmio.h

@@ -28,6 +28,9 @@ struct kvm_decode {
 	bool sign_extend;
 };
 
+void kvm_mmio_write_buf(void *buf, unsigned int len, unsigned long data);
+unsigned long kvm_mmio_read_buf(const void *buf, unsigned int len);
+
 int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run);
 int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		 phys_addr_t fault_ipa);

arch/arm/kvm/Kconfig

@@ -46,6 +46,13 @@ config KVM_ARM_HOST
 	---help---
 	  Provides host support for ARM processors.
 
+config KVM_NEW_VGIC
+	bool "New VGIC implementation"
+	depends on KVM
+	default y
+	---help---
+	  uses the new VGIC implementation
+
 source drivers/vhost/Kconfig
 
 endif # VIRTUALIZATION

arch/arm/kvm/Makefile

@@ -21,7 +21,18 @@ obj-$(CONFIG_KVM_ARM_HOST) += hyp/
 obj-y += kvm-arm.o init.o interrupts.o
 obj-y += arm.o handle_exit.o guest.o mmu.o emulate.o reset.o
 obj-y += coproc.o coproc_a15.o coproc_a7.o mmio.o psci.o perf.o
+
+ifeq ($(CONFIG_KVM_NEW_VGIC),y)
+obj-y += $(KVM)/arm/vgic/vgic.o
+obj-y += $(KVM)/arm/vgic/vgic-init.o
+obj-y += $(KVM)/arm/vgic/vgic-irqfd.o
+obj-y += $(KVM)/arm/vgic/vgic-v2.o
+obj-y += $(KVM)/arm/vgic/vgic-mmio.o
+obj-y += $(KVM)/arm/vgic/vgic-mmio-v2.o
+obj-y += $(KVM)/arm/vgic/vgic-kvm-device.o
+else
 obj-y += $(KVM)/arm/vgic.o
 obj-y += $(KVM)/arm/vgic-v2.o
 obj-y += $(KVM)/arm/vgic-v2-emul.o
+endif
 obj-y += $(KVM)/arm/arch_timer.o

arch/arm/kvm/arm.c

@@ -459,7 +459,7 @@ static void update_vttbr(struct kvm *kvm)
 static int kvm_vcpu_first_run_init(struct kvm_vcpu *vcpu)
 {
 	struct kvm *kvm = vcpu->kvm;
-	int ret;
+	int ret = 0;
 
 	if (likely(vcpu->arch.has_run_once))
 		return 0;
@@ -482,9 +482,9 @@ static int kvm_vcpu_first_run_init(struct kvm_vcpu *vcpu)
 	 * interrupts from the virtual timer with a userspace gic.
 	 */
 	if (irqchip_in_kernel(kvm) && vgic_initialized(kvm))
-		kvm_timer_enable(kvm);
+		ret = kvm_timer_enable(vcpu);
 
-	return 0;
+	return ret;
 }
 
 bool kvm_arch_intc_initialized(struct kvm *kvm)
@@ -492,30 +492,37 @@ bool kvm_arch_intc_initialized(struct kvm *kvm)
 	return vgic_initialized(kvm);
 }
 
-static void kvm_arm_halt_guest(struct kvm *kvm) __maybe_unused;
-static void kvm_arm_resume_guest(struct kvm *kvm) __maybe_unused;
-
-static void kvm_arm_halt_guest(struct kvm *kvm)
+void kvm_arm_halt_guest(struct kvm *kvm)
 {
 	int i;
 	struct kvm_vcpu *vcpu;
 
 	kvm_for_each_vcpu(i, vcpu, kvm)
 		vcpu->arch.pause = true;
-	force_vm_exit(cpu_all_mask);
+	kvm_make_all_cpus_request(kvm, KVM_REQ_VCPU_EXIT);
 }
 
-static void kvm_arm_resume_guest(struct kvm *kvm)
+void kvm_arm_halt_vcpu(struct kvm_vcpu *vcpu)
+{
+	vcpu->arch.pause = true;
+	kvm_vcpu_kick(vcpu);
+}
+
+void kvm_arm_resume_vcpu(struct kvm_vcpu *vcpu)
+{
+	struct swait_queue_head *wq = kvm_arch_vcpu_wq(vcpu);
+
+	vcpu->arch.pause = false;
+	swake_up(wq);
+}
+
+void kvm_arm_resume_guest(struct kvm *kvm)
 {
 	int i;
 	struct kvm_vcpu *vcpu;
 
-	kvm_for_each_vcpu(i, vcpu, kvm) {
-		struct swait_queue_head *wq = kvm_arch_vcpu_wq(vcpu);
-
-		vcpu->arch.pause = false;
-		swake_up(wq);
-	}
+	kvm_for_each_vcpu(i, vcpu, kvm)
+		kvm_arm_resume_vcpu(vcpu);
 }
 
 static void vcpu_sleep(struct kvm_vcpu *vcpu)

arch/arm/kvm/mmio.c

@@ -23,7 +23,7 @@
 
 #include "trace.h"
 
-static void mmio_write_buf(char *buf, unsigned int len, unsigned long data)
+void kvm_mmio_write_buf(void *buf, unsigned int len, unsigned long data)
 {
 	void *datap = NULL;
 	union {
@@ -55,7 +55,7 @@ static void mmio_write_buf(char *buf, unsigned int len, unsigned long data)
 	memcpy(buf, datap, len);
 }
 
-static unsigned long mmio_read_buf(char *buf, unsigned int len)
+unsigned long kvm_mmio_read_buf(const void *buf, unsigned int len)
 {
 	unsigned long data = 0;
 	union {
@@ -66,7 +66,7 @@ static unsigned long mmio_read_buf(char *buf, unsigned int len)
 
 	switch (len) {
 	case 1:
-		data = buf[0];
+		data = *(u8 *)buf;
 		break;
 	case 2:
 		memcpy(&tmp.hword, buf, len);
@@ -87,11 +87,10 @@ static unsigned long mmio_read_buf(char *buf, unsigned int len)
 
 /**
  * kvm_handle_mmio_return -- Handle MMIO loads after user space emulation
+ *			     or in-kernel IO emulation
+ *
  * @vcpu: The VCPU pointer
  * @run:  The VCPU run struct containing the mmio data
- *
- * This should only be called after returning from userspace for MMIO load
- * emulation.
  */
 int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run)
 {
@@ -104,7 +103,7 @@ int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run)
 		if (len > sizeof(unsigned long))
 			return -EINVAL;
 
-		data = mmio_read_buf(run->mmio.data, len);
+		data = kvm_mmio_read_buf(run->mmio.data, len);
 
 		if (vcpu->arch.mmio_decode.sign_extend &&
 		    len < sizeof(unsigned long)) {
@@ -190,7 +189,7 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 					       len);
 
 		trace_kvm_mmio(KVM_TRACE_MMIO_WRITE, len, fault_ipa, data);
-		mmio_write_buf(data_buf, len, data);
+		kvm_mmio_write_buf(data_buf, len, data);
 
 		ret = kvm_io_bus_write(vcpu, KVM_MMIO_BUS, fault_ipa, len,
 				       data_buf);
@@ -206,18 +205,19 @@ int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 	run->mmio.is_write	= is_write;
 	run->mmio.phys_addr	= fault_ipa;
 	run->mmio.len		= len;
-	if (is_write)
-		memcpy(run->mmio.data, data_buf, len);
 
 	if (!ret) {
 		/* We handled the access successfully in the kernel. */
+		if (!is_write)
+			memcpy(run->mmio.data, data_buf, len);
 		vcpu->stat.mmio_exit_kernel++;
 		kvm_handle_mmio_return(vcpu, run);
 		return 1;
-	} else {
-		vcpu->stat.mmio_exit_user++;
 	}
 
+	if (is_write)
+		memcpy(run->mmio.data, data_buf, len);
+	vcpu->stat.mmio_exit_user++;
 	run->exit_reason	= KVM_EXIT_MMIO;
 	return 0;
 }

arch/arm64/include/asm/kvm_host.h

@@ -43,6 +43,8 @@
 
 #define KVM_VCPU_MAX_FEATURES 4
 
+#define KVM_REQ_VCPU_EXIT	8
+
 int __attribute_const__ kvm_target_cpu(void);
 int kvm_reset_vcpu(struct kvm_vcpu *vcpu);
 int kvm_arch_dev_ioctl_check_extension(long ext);
@@ -325,6 +327,10 @@ static inline void kvm_arch_mmu_notifier_invalidate_page(struct kvm *kvm,
 
 struct kvm_vcpu *kvm_arm_get_running_vcpu(void);
 struct kvm_vcpu * __percpu *kvm_get_running_vcpus(void);
+void kvm_arm_halt_guest(struct kvm *kvm);
+void kvm_arm_resume_guest(struct kvm *kvm);
+void kvm_arm_halt_vcpu(struct kvm_vcpu *vcpu);
+void kvm_arm_resume_vcpu(struct kvm_vcpu *vcpu);
 
 u64 __kvm_call_hyp(void *hypfn, ...);
 #define kvm_call_hyp(f, ...) __kvm_call_hyp(kvm_ksym_ref(f), ##__VA_ARGS__)

arch/arm64/include/asm/kvm_mmio.h

@@ -30,6 +30,9 @@ struct kvm_decode {
 	bool sign_extend;
 };
 
+void kvm_mmio_write_buf(void *buf, unsigned int len, unsigned long data);
+unsigned long kvm_mmio_read_buf(const void *buf, unsigned int len);
+
 int kvm_handle_mmio_return(struct kvm_vcpu *vcpu, struct kvm_run *run);
 int io_mem_abort(struct kvm_vcpu *vcpu, struct kvm_run *run,
 		 phys_addr_t fault_ipa);

arch/arm64/kvm/Kconfig

@@ -54,6 +54,13 @@ config KVM_ARM_PMU
 	  Adds support for a virtual Performance Monitoring Unit (PMU) in
 	  virtual machines.
 
+config KVM_NEW_VGIC
+	bool "New VGIC implementation"
+	depends on KVM
+	default y
+        ---help---
+          uses the new VGIC implementation
+
 source drivers/vhost/Kconfig
 
 endif # VIRTUALIZATION

arch/arm64/kvm/Makefile

@@ -20,10 +20,22 @@ kvm-$(CONFIG_KVM_ARM_HOST) += emulate.o inject_fault.o regmap.o
 kvm-$(CONFIG_KVM_ARM_HOST) += hyp.o hyp-init.o handle_exit.o
 kvm-$(CONFIG_KVM_ARM_HOST) += guest.o debug.o reset.o sys_regs.o sys_regs_generic_v8.o
 
+ifeq ($(CONFIG_KVM_NEW_VGIC),y)
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-init.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-irqfd.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-v2.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-v3.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio-v2.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-mmio-v3.o
+kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic/vgic-kvm-device.o
+else
 kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic.o
 kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v2.o
 kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v2-emul.o
 kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v3.o
 kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/vgic-v3-emul.o
+endif
 kvm-$(CONFIG_KVM_ARM_HOST) += $(KVM)/arm/arch_timer.o
 kvm-$(CONFIG_KVM_ARM_PMU) += $(KVM)/arm/pmu.o

arch/arm64/kvm/inject_fault.c

@@ -162,7 +162,7 @@ static void inject_abt64(struct kvm_vcpu *vcpu, bool is_iabt, unsigned long addr
 		esr |= (ESR_ELx_EC_IABT_CUR << ESR_ELx_EC_SHIFT);
 
 	if (!is_iabt)
-		esr |= ESR_ELx_EC_DABT_LOW;
+		esr |= ESR_ELx_EC_DABT_LOW << ESR_ELx_EC_SHIFT;
 
 	vcpu_sys_reg(vcpu, ESR_EL1) = esr | ESR_ELx_FSC_EXTABT;
 }

include/kvm/arm_arch_timer.h

@@ -24,9 +24,6 @@
 #include <linux/workqueue.h>
 
 struct arch_timer_kvm {
-	/* Is the timer enabled */
-	bool			enabled;
-
 	/* Virtual offset */
 	cycle_t			cntvoff;
 };
@@ -53,15 +50,15 @@ struct arch_timer_cpu {
 	/* Timer IRQ */
 	struct kvm_irq_level		irq;
 
-	/* VGIC mapping */
-	struct irq_phys_map		*map;
-
 	/* Active IRQ state caching */
 	bool				active_cleared_last;
+
+	/* Is the timer enabled */
+	bool			enabled;
 };
 
 int kvm_timer_hyp_init(void);
-void kvm_timer_enable(struct kvm *kvm);
+int kvm_timer_enable(struct kvm_vcpu *vcpu);
 void kvm_timer_init(struct kvm *kvm);
 int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu,
 			 const struct kvm_irq_level *irq);

include/kvm/arm_vgic.h

@@ -19,6 +19,10 @@
 #ifndef __ASM_ARM_KVM_VGIC_H
 #define __ASM_ARM_KVM_VGIC_H
 
+#ifdef CONFIG_KVM_NEW_VGIC
+#include <kvm/vgic/vgic.h>
+#else
+
 #include <linux/kernel.h>
 #include <linux/kvm.h>
 #include <linux/irqreturn.h>
@@ -158,7 +162,6 @@ struct vgic_io_device {
 struct irq_phys_map {
 	u32			virt_irq;
 	u32			phys_irq;
-	u32			irq;
 };
 
 struct irq_phys_map_entry {
@@ -305,9 +308,6 @@ struct vgic_cpu {
 	unsigned long   *active_shared;
 	unsigned long   *pend_act_shared;
 
-	/* Number of list registers on this CPU */
-	int		nr_lr;
-
 	/* CPU vif control registers for world switch */
 	union {
 		struct vgic_v2_cpu_if	vgic_v2;
@@ -342,17 +342,18 @@ void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu);
 int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int irq_num,
 			bool level);
 int kvm_vgic_inject_mapped_irq(struct kvm *kvm, int cpuid,
-			       struct irq_phys_map *map, bool level);
+			       unsigned int virt_irq, bool level);
 void vgic_v3_dispatch_sgi(struct kvm_vcpu *vcpu, u64 reg);
 int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu);
-struct irq_phys_map *kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu,
-					   int virt_irq, int irq);
-int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, struct irq_phys_map *map);
-bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, struct irq_phys_map *map);
+int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, int virt_irq, int phys_irq);
+int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int virt_irq);
+bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int virt_irq);
 
 #define irqchip_in_kernel(k)	(!!((k)->arch.vgic.in_kernel))
 #define vgic_initialized(k)	(!!((k)->arch.vgic.nr_cpus))
 #define vgic_ready(k)		((k)->arch.vgic.ready)
+#define vgic_valid_spi(k, i)	(((i) >= VGIC_NR_PRIVATE_IRQS) && \
+				 ((i) < (k)->arch.vgic.nr_irqs))
 
 int vgic_v2_probe(const struct gic_kvm_info *gic_kvm_info,
 		  const struct vgic_ops **ops,
@@ -370,4 +371,5 @@ static inline int vgic_v3_probe(const struct gic_kvm_info *gic_kvm_info,
 }
 #endif
 
+#endif	/* old VGIC include */
 #endif

include/kvm/vgic/vgic.h

@@ -0,0 +1,246 @@
+/*
+ * Copyright (C) 2015, 2016 ARM Ltd.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program.  If not, see <http://www.gnu.org/licenses/>.
+ */
+#ifndef __ASM_ARM_KVM_VGIC_VGIC_H
+#define __ASM_ARM_KVM_VGIC_VGIC_H
+
+#include <linux/kernel.h>
+#include <linux/kvm.h>
+#include <linux/irqreturn.h>
+#include <linux/spinlock.h>
+#include <linux/types.h>
+#include <kvm/iodev.h>
+
+#define VGIC_V3_MAX_CPUS	255
+#define VGIC_V2_MAX_CPUS	8
+#define VGIC_NR_IRQS_LEGACY     256
+#define VGIC_NR_SGIS		16
+#define VGIC_NR_PPIS		16
+#define VGIC_NR_PRIVATE_IRQS	(VGIC_NR_SGIS + VGIC_NR_PPIS)
+#define VGIC_MAX_PRIVATE	(VGIC_NR_PRIVATE_IRQS - 1)
+#define VGIC_MAX_SPI		1019
+#define VGIC_MAX_RESERVED	1023
+#define VGIC_MIN_LPI		8192
+
+enum vgic_type {
+	VGIC_V2,		/* Good ol' GICv2 */
+	VGIC_V3,		/* New fancy GICv3 */
+};
+
+/* same for all guests, as depending only on the _host's_ GIC model */
+struct vgic_global {
+	/* type of the host GIC */
+	enum vgic_type		type;
+
+	/* Physical address of vgic virtual cpu interface */
+	phys_addr_t		vcpu_base;
+
+	/* virtual control interface mapping */
+	void __iomem		*vctrl_base;
+
+	/* Number of implemented list registers */
+	int			nr_lr;
+
+	/* Maintenance IRQ number */
+	unsigned int		maint_irq;
+
+	/* maximum number of VCPUs allowed (GICv2 limits us to 8) */
+	int			max_gic_vcpus;
+
+	/* Only needed for the legacy KVM_CREATE_IRQCHIP */
+	bool			can_emulate_gicv2;
+};
+
+extern struct vgic_global kvm_vgic_global_state;
+
+#define VGIC_V2_MAX_LRS		(1 << 6)
+#define VGIC_V3_MAX_LRS		16
+#define VGIC_V3_LR_INDEX(lr)	(VGIC_V3_MAX_LRS - 1 - lr)
+
+enum vgic_irq_config {
+	VGIC_CONFIG_EDGE = 0,
+	VGIC_CONFIG_LEVEL
+};
+
+struct vgic_irq {
+	spinlock_t irq_lock;		/* Protects the content of the struct */
+	struct list_head ap_list;
+
+	struct kvm_vcpu *vcpu;		/* SGIs and PPIs: The VCPU
+					 * SPIs and LPIs: The VCPU whose ap_list
+					 * this is queued on.
+					 */
+
+	struct kvm_vcpu *target_vcpu;	/* The VCPU that this interrupt should
+					 * be sent to, as a result of the
+					 * targets reg (v2) or the
+					 * affinity reg (v3).
+					 */
+
+	u32 intid;			/* Guest visible INTID */
+	bool pending;
+	bool line_level;		/* Level only */
+	bool soft_pending;		/* Level only */
+	bool active;			/* not used for LPIs */
+	bool enabled;
+	bool hw;			/* Tied to HW IRQ */
+	u32 hwintid;			/* HW INTID number */
+	union {
+		u8 targets;			/* GICv2 target VCPUs mask */
+		u32 mpidr;			/* GICv3 target VCPU */
+	};
+	u8 source;			/* GICv2 SGIs only */
+	u8 priority;
+	enum vgic_irq_config config;	/* Level or edge */
+};
+
+struct vgic_register_region;
+
+struct vgic_io_device {
+	gpa_t base_addr;
+	struct kvm_vcpu *redist_vcpu;
+	const struct vgic_register_region *regions;
+	int nr_regions;
+	struct kvm_io_device dev;
+};
+
+struct vgic_dist {
+	bool			in_kernel;
+	bool			ready;
+	bool			initialized;
+
+	/* vGIC model the kernel emulates for the guest (GICv2 or GICv3) */
+	u32			vgic_model;
+
+	int			nr_spis;
+
+	/* TODO: Consider moving to global state */
+	/* Virtual control interface mapping */
+	void __iomem		*vctrl_base;
+
+	/* base addresses in guest physical address space: */
+	gpa_t			vgic_dist_base;		/* distributor */
+	union {
+		/* either a GICv2 CPU interface */
+		gpa_t			vgic_cpu_base;
+		/* or a number of GICv3 redistributor regions */
+		gpa_t			vgic_redist_base;
+	};
+
+	/* distributor enabled */
+	bool			enabled;
+
+	struct vgic_irq		*spis;
+
+	struct vgic_io_device	dist_iodev;
+	struct vgic_io_device	*redist_iodevs;
+};
+
+struct vgic_v2_cpu_if {
+	u32		vgic_hcr;
+	u32		vgic_vmcr;
+	u32		vgic_misr;	/* Saved only */
+	u64		vgic_eisr;	/* Saved only */
+	u64		vgic_elrsr;	/* Saved only */
+	u32		vgic_apr;
+	u32		vgic_lr[VGIC_V2_MAX_LRS];
+};
+
+struct vgic_v3_cpu_if {
+#ifdef CONFIG_KVM_ARM_VGIC_V3
+	u32		vgic_hcr;
+	u32		vgic_vmcr;
+	u32		vgic_sre;	/* Restored only, change ignored */
+	u32		vgic_misr;	/* Saved only */
+	u32		vgic_eisr;	/* Saved only */
+	u32		vgic_elrsr;	/* Saved only */
+	u32		vgic_ap0r[4];
+	u32		vgic_ap1r[4];
+	u64		vgic_lr[VGIC_V3_MAX_LRS];
+#endif
+};
+
+struct vgic_cpu {
+	/* CPU vif control registers for world switch */
+	union {
+		struct vgic_v2_cpu_if	vgic_v2;
+		struct vgic_v3_cpu_if	vgic_v3;
+	};
+
+	unsigned int used_lrs;
+	struct vgic_irq private_irqs[VGIC_NR_PRIVATE_IRQS];
+
+	spinlock_t ap_list_lock;	/* Protects the ap_list */
+
+	/*
+	 * List of IRQs that this VCPU should consider because they are either
+	 * Active or Pending (hence the name; AP list), or because they recently
+	 * were one of the two and need to be migrated off this list to another
+	 * VCPU.
+	 */
+	struct list_head ap_list_head;
+
+	u64 live_lrs;
+};
+
+int kvm_vgic_addr(struct kvm *kvm, unsigned long type, u64 *addr, bool write);
+void kvm_vgic_early_init(struct kvm *kvm);
+int kvm_vgic_create(struct kvm *kvm, u32 type);
+void kvm_vgic_destroy(struct kvm *kvm);
+void kvm_vgic_vcpu_early_init(struct kvm_vcpu *vcpu);
+void kvm_vgic_vcpu_destroy(struct kvm_vcpu *vcpu);
+int kvm_vgic_map_resources(struct kvm *kvm);
+int kvm_vgic_hyp_init(void);
+
+int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid,
+			bool level);
+int kvm_vgic_inject_mapped_irq(struct kvm *kvm, int cpuid, unsigned int intid,
+			       bool level);
+int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, u32 virt_irq, u32 phys_irq);
+int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int virt_irq);
+bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int virt_irq);
+
+int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu);
+
+#define irqchip_in_kernel(k)	(!!((k)->arch.vgic.in_kernel))
+#define vgic_initialized(k)	((k)->arch.vgic.initialized)
+#define vgic_ready(k)		((k)->arch.vgic.ready)
+#define vgic_valid_spi(k, i)	(((i) >= VGIC_NR_PRIVATE_IRQS) && \
+			((i) < (k)->arch.vgic.nr_spis + VGIC_NR_PRIVATE_IRQS))
+
+bool kvm_vcpu_has_pending_irqs(struct kvm_vcpu *vcpu);
+void kvm_vgic_sync_hwstate(struct kvm_vcpu *vcpu);
+void kvm_vgic_flush_hwstate(struct kvm_vcpu *vcpu);
+
+#ifdef CONFIG_KVM_ARM_VGIC_V3
+void vgic_v3_dispatch_sgi(struct kvm_vcpu *vcpu, u64 reg);
+#else
+static inline void vgic_v3_dispatch_sgi(struct kvm_vcpu *vcpu, u64 reg)
+{
+}
+#endif
+
+/**
+ * kvm_vgic_get_max_vcpus - Get the maximum number of VCPUs allowed by HW
+ *
+ * The host's GIC naturally limits the maximum amount of VCPUs a guest
+ * can use.
+ */
+static inline int kvm_vgic_get_max_vcpus(void)
+{
+	return kvm_vgic_global_state.max_gic_vcpus;
+}
+
+#endif /* __ASM_ARM_KVM_VGIC_VGIC_H */

include/linux/irqchip/arm-gic-v3.h

@@ -275,6 +275,12 @@
 #define ICH_LR_ACTIVE_BIT		(1ULL << 63)
 #define ICH_LR_PHYS_ID_SHIFT		32
 #define ICH_LR_PHYS_ID_MASK		(0x3ffULL << ICH_LR_PHYS_ID_SHIFT)
+#define ICH_LR_PRIORITY_SHIFT		48
+
+/* These are for GICv2 emulation only */
+#define GICH_LR_VIRTUALID		(0x3ffUL << 0)
+#define GICH_LR_PHYSID_CPUID_SHIFT	(10)
+#define GICH_LR_PHYSID_CPUID		(7UL << GICH_LR_PHYSID_CPUID_SHIFT)
 
 #define ICH_MISR_EOI			(1 << 0)
 #define ICH_MISR_U			(1 << 1)

include/linux/irqchip/arm-gic.h

@@ -33,6 +33,7 @@
 
 #define GIC_DIST_CTRL			0x000
 #define GIC_DIST_CTR			0x004
+#define GIC_DIST_IIDR			0x008
 #define GIC_DIST_IGROUP			0x080
 #define GIC_DIST_ENABLE_SET		0x100
 #define GIC_DIST_ENABLE_CLEAR		0x180
@@ -76,6 +77,7 @@
 #define GICH_LR_VIRTUALID		(0x3ff << 0)
 #define GICH_LR_PHYSID_CPUID_SHIFT	(10)
 #define GICH_LR_PHYSID_CPUID		(0x3ff << GICH_LR_PHYSID_CPUID_SHIFT)
+#define GICH_LR_PRIORITY_SHIFT		23
 #define GICH_LR_STATE			(3 << 28)
 #define GICH_LR_PENDING_BIT		(1 << 28)
 #define GICH_LR_ACTIVE_BIT		(1 << 29)

virt/kvm/arm/arch_timer.c

@@ -20,6 +20,7 @@
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
 #include <linux/interrupt.h>
+#include <linux/irq.h>
 
 #include <clocksource/arm_arch_timer.h>
 #include <asm/arch_timer.h>
@@ -174,10 +175,10 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level)
 
 	timer->active_cleared_last = false;
 	timer->irq.level = new_level;
-	trace_kvm_timer_update_irq(vcpu->vcpu_id, timer->map->virt_irq,
+	trace_kvm_timer_update_irq(vcpu->vcpu_id, timer->irq.irq,
 				   timer->irq.level);
 	ret = kvm_vgic_inject_mapped_irq(vcpu->kvm, vcpu->vcpu_id,
-					 timer->map,
+					 timer->irq.irq,
 					 timer->irq.level);
 	WARN_ON(ret);
 }
@@ -196,7 +197,7 @@ static int kvm_timer_update_state(struct kvm_vcpu *vcpu)
 	 * because the guest would never see the interrupt.  Instead wait
 	 * until we call this function from kvm_timer_flush_hwstate.
 	 */
-	if (!vgic_initialized(vcpu->kvm))
+	if (!vgic_initialized(vcpu->kvm) || !timer->enabled)
 		return -ENODEV;
 
 	if (kvm_timer_should_fire(vcpu) != timer->irq.level)
@@ -274,10 +275,8 @@ void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu)
 	* to ensure that hardware interrupts from the timer triggers a guest
 	* exit.
 	*/
-	if (timer->irq.level || kvm_vgic_map_is_active(vcpu, timer->map))
-		phys_active = true;
-	else
-		phys_active = false;
+	phys_active = timer->irq.level ||
+			kvm_vgic_map_is_active(vcpu, timer->irq.irq);
 
 	/*
 	 * We want to avoid hitting the (re)distributor as much as
@@ -302,7 +301,7 @@ void kvm_timer_flush_hwstate(struct kvm_vcpu *vcpu)
 	if (timer->active_cleared_last && !phys_active)
 		return;
 
-	ret = irq_set_irqchip_state(timer->map->irq,
+	ret = irq_set_irqchip_state(host_vtimer_irq,
 				    IRQCHIP_STATE_ACTIVE,
 				    phys_active);
 	WARN_ON(ret);
@@ -334,7 +333,6 @@ int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu,
 			 const struct kvm_irq_level *irq)
 {
 	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
-	struct irq_phys_map *map;
 
 	/*
 	 * The vcpu timer irq number cannot be determined in
@@ -353,15 +351,6 @@ int kvm_timer_vcpu_reset(struct kvm_vcpu *vcpu,
 	timer->cntv_ctl = 0;
 	kvm_timer_update_state(vcpu);
 
-	/*
-	 * Tell the VGIC that the virtual interrupt is tied to a
-	 * physical interrupt. We do that once per VCPU.
-	 */
-	map = kvm_vgic_map_phys_irq(vcpu, irq->irq, host_vtimer_irq);
-	if (WARN_ON(IS_ERR(map)))
-		return PTR_ERR(map);
-
-	timer->map = map;
 	return 0;
 }
 
@@ -487,14 +476,43 @@ void kvm_timer_vcpu_terminate(struct kvm_vcpu *vcpu)
 	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
 
 	timer_disarm(timer);
-	if (timer->map)
-		kvm_vgic_unmap_phys_irq(vcpu, timer->map);
+	kvm_vgic_unmap_phys_irq(vcpu, timer->irq.irq);
 }
 
-void kvm_timer_enable(struct kvm *kvm)
+int kvm_timer_enable(struct kvm_vcpu *vcpu)
 {
-	if (kvm->arch.timer.enabled)
-		return;
+	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
+	struct irq_desc *desc;
+	struct irq_data *data;
+	int phys_irq;
+	int ret;
+
+	if (timer->enabled)
+		return 0;
+
+	/*
+	 * Find the physical IRQ number corresponding to the host_vtimer_irq
+	 */
+	desc = irq_to_desc(host_vtimer_irq);
+	if (!desc) {
+		kvm_err("%s: no interrupt descriptor\n", __func__);
+		return -EINVAL;
+	}
+
+	data = irq_desc_get_irq_data(desc);
+	while (data->parent_data)
+		data = data->parent_data;
+
+	phys_irq = data->hwirq;
+
+	/*
+	 * Tell the VGIC that the virtual interrupt is tied to a
+	 * physical interrupt. We do that once per VCPU.
+	 */
+	ret = kvm_vgic_map_phys_irq(vcpu, timer->irq.irq, phys_irq);
+	if (ret)
+		return ret;
+
 
 	/*
 	 * There is a potential race here between VCPUs starting for the first
@@ -505,7 +523,9 @@ void kvm_timer_enable(struct kvm *kvm)
 	 * the arch timers are enabled.
 	 */
 	if (timecounter && wqueue)
-		kvm->arch.timer.enabled = 1;
+		timer->enabled = 1;
+
+	return 0;
 }
 
 void kvm_timer_init(struct kvm *kvm)

virt/kvm/arm/hyp/timer-sr.c

@@ -24,11 +24,10 @@
 /* vcpu is already in the HYP VA space */
 void __hyp_text __timer_save_state(struct kvm_vcpu *vcpu)
 {
-	struct kvm *kvm = kern_hyp_va(vcpu->kvm);
 	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
 	u64 val;
 
-	if (kvm->arch.timer.enabled) {
+	if (timer->enabled) {
 		timer->cntv_ctl = read_sysreg_el0(cntv_ctl);
 		timer->cntv_cval = read_sysreg_el0(cntv_cval);
 	}
@@ -60,7 +59,7 @@ void __hyp_text __timer_restore_state(struct kvm_vcpu *vcpu)
 	val |= CNTHCTL_EL1PCTEN;
 	write_sysreg(val, cnthctl_el2);
 
-	if (kvm->arch.timer.enabled) {
+	if (timer->enabled) {
 		write_sysreg(kvm->arch.timer.cntvoff, cntvoff_el2);
 		write_sysreg_el0(timer->cntv_cval, cntv_cval);
 		isb();

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

@@ -21,11 +21,18 @@
 
 #include <asm/kvm_hyp.h>
 
+#ifdef CONFIG_KVM_NEW_VGIC
+extern struct vgic_global kvm_vgic_global_state;
+#define vgic_v2_params kvm_vgic_global_state
+#else
+extern struct vgic_params vgic_v2_params;
+#endif
+
 static void __hyp_text save_maint_int_state(struct kvm_vcpu *vcpu,
 					    void __iomem *base)
 {
 	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
-	int nr_lr = vcpu->arch.vgic_cpu.nr_lr;
+	int nr_lr = (kern_hyp_va(&vgic_v2_params))->nr_lr;
 	u32 eisr0, eisr1;
 	int i;
 	bool expect_mi;
@@ -67,7 +74,7 @@ static void __hyp_text save_maint_int_state(struct kvm_vcpu *vcpu,
 static void __hyp_text save_elrsr(struct kvm_vcpu *vcpu, void __iomem *base)
 {
 	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
-	int nr_lr = vcpu->arch.vgic_cpu.nr_lr;
+	int nr_lr = (kern_hyp_va(&vgic_v2_params))->nr_lr;
 	u32 elrsr0, elrsr1;
 
 	elrsr0 = readl_relaxed(base + GICH_ELRSR0);
@@ -86,7 +93,7 @@ static void __hyp_text save_elrsr(struct kvm_vcpu *vcpu, void __iomem *base)
 static void __hyp_text save_lrs(struct kvm_vcpu *vcpu, void __iomem *base)
 {
 	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
-	int nr_lr = vcpu->arch.vgic_cpu.nr_lr;
+	int nr_lr = (kern_hyp_va(&vgic_v2_params))->nr_lr;
 	int i;
 
 	for (i = 0; i < nr_lr; i++) {
@@ -141,13 +148,13 @@ void __hyp_text __vgic_v2_restore_state(struct kvm_vcpu *vcpu)
 	struct vgic_v2_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v2;
 	struct vgic_dist *vgic = &kvm->arch.vgic;
 	void __iomem *base = kern_hyp_va(vgic->vctrl_base);
-	int i, nr_lr;
+	int nr_lr = (kern_hyp_va(&vgic_v2_params))->nr_lr;
+	int i;
 	u64 live_lrs = 0;
 
 	if (!base)
 		return;
 
-	nr_lr = vcpu->arch.vgic_cpu.nr_lr;
 
 	for (i = 0; i < nr_lr; i++)
 		if (cpu_if->vgic_lr[i] & GICH_LR_STATE)

virt/kvm/arm/pmu.c

@@ -436,7 +436,14 @@ static int kvm_arm_pmu_v3_init(struct kvm_vcpu *vcpu)
 	return 0;
 }
 
-static bool irq_is_valid(struct kvm *kvm, int irq, bool is_ppi)
+#define irq_is_ppi(irq) ((irq) >= VGIC_NR_SGIS && (irq) < VGIC_NR_PRIVATE_IRQS)
+
+/*
+ * For one VM the interrupt type must be same for each vcpu.
+ * As a PPI, the interrupt number is the same for all vcpus,
+ * while as an SPI it must be a separate number per vcpu.
+ */
+static bool pmu_irq_is_valid(struct kvm *kvm, int irq)
 {
 	int i;
 	struct kvm_vcpu *vcpu;
@@ -445,7 +452,7 @@ static bool irq_is_valid(struct kvm *kvm, int irq, bool is_ppi)
 		if (!kvm_arm_pmu_irq_initialized(vcpu))
 			continue;
 
-		if (is_ppi) {
+		if (irq_is_ppi(irq)) {
 			if (vcpu->arch.pmu.irq_num != irq)
 				return false;
 		} else {
@@ -457,7 +464,6 @@ static bool irq_is_valid(struct kvm *kvm, int irq, bool is_ppi)
 	return true;
 }
 
-
 int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 {
 	switch (attr->attr) {
@@ -471,14 +477,11 @@ int kvm_arm_pmu_v3_set_attr(struct kvm_vcpu *vcpu, struct kvm_device_attr *attr)
 		if (get_user(irq, uaddr))
 			return -EFAULT;
 
-		/*
-		 * The PMU overflow interrupt could be a PPI or SPI, but for one
-		 * VM the interrupt type must be same for each vcpu. As a PPI,
-		 * the interrupt number is the same for all vcpus, while as an
-		 * SPI it must be a separate number per vcpu.
-		 */
-		if (irq < VGIC_NR_SGIS || irq >= vcpu->kvm->arch.vgic.nr_irqs ||
-		    !irq_is_valid(vcpu->kvm, irq, irq < VGIC_NR_PRIVATE_IRQS))
+		/* The PMU overflow interrupt can be a PPI or a valid SPI. */
+		if (!(irq_is_ppi(irq) || vgic_valid_spi(vcpu->kvm, irq)))
+			return -EINVAL;
+
+		if (!pmu_irq_is_valid(vcpu->kvm, irq))
 			return -EINVAL;
 
 		if (kvm_arm_pmu_irq_initialized(vcpu))