[PATCH v2 00/18] KVM: arm64: vgic-v5 fixes for 7.2

[PATCH v2 00/18] KVM: arm64: vgic-v5 fixes for 7.2

[Marc Zyngier]

Having completely missed the 7.1 window, this is a repost of this
series cleaning-up the vgic-v5 PPI support now targeting 7.2.

Not a lot has changed since v1 [1], only the documentation typo
spotted by our eagle-eyed Joey.

[1] https://lore.kernel.org/r/20260415115559.2227718-1-maz@kernel.org

Marc Zyngier (9):
  KVM: arm64: vgic-v5: Add for_each_visible_v5_ppi() iterator
  KVM: arm64: vgic-v5: Move PPI caps into kvm_vgic_global_state
  KVM: arm64: vgic-v5: Remove use of __assign_bit() with a constant
  KVM: arm64: vgic-v5: Drop pointless ARM64_HAS_GICV5_CPUIF check
  KVM: arm64: vgic: Constify struct irq_ops usage
  KVM: arm64: vgic: Consolidate vgic_allocate_private_irqs_locked()
  KVM: arm64: vgic-v5: Drop defensive checks from
    vgic_v5_ppi_queue_irq_unlock()
  KVM: arm64: vgic: Rationalise per-CPU irq accessor
  KVM: arm64: vgic-v5: Limit support to 64 PPIs

Sascha Bischoff (9):
  KVM: arm64: vgic-v5: Add missing trap handing for NV triage
  KVM: arm64: vgic-v5: Atomically assign bits to PPI DVI bitmap
  KVM: arm64: selftests: Add missing GIC CDEN to no-vgic-v5 selftest
  KVM: arm64: selftests: Cleanup unused vars in GICv5 PPI selftest
  KVM: arm64: selftests: Improve error handling for GICv5 PPI selftest
  Documentation: KVM: Fix typos in VGICv5 documentation
  Documentation: KVM: Clarify that PMU_V3_IRQ IntID requirements for
    GICv5
  irqchip/gic-v5: Immediately exec priority drop following activate
  KVM: arm64: Fix arch timer interrupts for GICv3-on-GICv5 guests

 .../virt/kvm/devices/arm-vgic-v5.rst          |  6 +-
 Documentation/virt/kvm/devices/vcpu.rst       |  7 +-
 arch/arm64/kvm/arch_timer.c                   | 31 +++----
 arch/arm64/kvm/emulate-nested.c               |  8 ++
 arch/arm64/kvm/hyp/vgic-v5-sr.c               | 82 ++++---------------
 arch/arm64/kvm/sys_regs.c                     | 19 ++---
 arch/arm64/kvm/vgic/vgic-init.c               | 45 ++++------
 arch/arm64/kvm/vgic/vgic-kvm-device.c         |  9 +-
 arch/arm64/kvm/vgic/vgic-v5.c                 | 51 ++++--------
 arch/arm64/kvm/vgic/vgic.c                    | 27 +++---
 arch/arm64/kvm/vgic/vgic.h                    |  3 +
 drivers/irqchip/irq-gic-v5.c                  | 13 +--
 include/kvm/arm_vgic.h                        | 19 +++--
 tools/testing/selftests/kvm/arm64/no-vgic.c   |  1 +
 tools/testing/selftests/kvm/arm64/vgic_v5.c   | 10 +--
 15 files changed, 132 insertions(+), 199 deletions(-)

-- 
2.47.3

[PATCH v2 01/18] KVM: arm64: vgic-v5: Add for_each_visible_v5_ppi() iterator

[Marc Zyngier]

We have multiple instances of iterators walking the vgic_ppi_mask
mask, and the way it is written has a tendency to make one's eyes
bleed.

Factor it as a helper and use that across the code base.

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/sys_regs.c     |  2 +-
 arch/arm64/kvm/vgic/vgic-v5.c | 10 ++++------
 arch/arm64/kvm/vgic/vgic.h    |  3 +++
 3 files changed, 8 insertions(+), 7 deletions(-)

diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 148fc3400ea81..513f5f1429b5f 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -751,7 +751,7 @@ static bool access_gicv5_ppi_enabler(struct kvm_vcpu *vcpu,
 	 * Sync the change in enable states to the vgic_irqs. We consider all
 	 * PPIs as we don't expose many to the guest.
 	 */
-	for_each_set_bit(i, mask, VGIC_V5_NR_PRIVATE_IRQS) {
+	for_each_visible_v5_ppi(i, vcpu->kvm) {
 		u32 intid = vgic_v5_make_ppi(i);
 		struct vgic_irq *irq;
 
diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index fdd39ea7f83ec..c0d36658ffe74 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -316,7 +316,7 @@ static void vgic_v5_sync_ppi_priorities(struct kvm_vcpu *vcpu)
 	 * those actually exposed to the guest by first iterating over the mask
 	 * of exposed PPIs.
 	 */
-	for_each_set_bit(i, vcpu->kvm->arch.vgic.gicv5_vm.vgic_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS) {
+	for_each_visible_v5_ppi(i, vcpu->kvm) {
 		u32 intid = vgic_v5_make_ppi(i);
 		struct vgic_irq *irq;
 		int pri_idx, pri_reg, pri_bit;
@@ -358,7 +358,7 @@ bool vgic_v5_has_pending_ppi(struct kvm_vcpu *vcpu)
 	if (!priority_mask)
 		return false;
 
-	for_each_set_bit(i, vcpu->kvm->arch.vgic.gicv5_vm.vgic_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS) {
+	for_each_visible_v5_ppi(i, vcpu->kvm) {
 		u32 intid = vgic_v5_make_ppi(i);
 		bool has_pending = false;
 		struct vgic_irq *irq;
@@ -391,8 +391,7 @@ void vgic_v5_fold_ppi_state(struct kvm_vcpu *vcpu)
 	activer = host_data_ptr(vgic_v5_ppi_state)->activer_exit;
 	pendr = host_data_ptr(vgic_v5_ppi_state)->pendr;
 
-	for_each_set_bit(i, vcpu->kvm->arch.vgic.gicv5_vm.vgic_ppi_mask,
-			 VGIC_V5_NR_PRIVATE_IRQS) {
+	for_each_visible_v5_ppi(i, vcpu->kvm) {
 		u32 intid = vgic_v5_make_ppi(i);
 		struct vgic_irq *irq;
 
@@ -429,8 +428,7 @@ void vgic_v5_flush_ppi_state(struct kvm_vcpu *vcpu)
 	 * ICC_PPI_PENDRx_EL1, however.
 	 */
 	bitmap_zero(pendr, VGIC_V5_NR_PRIVATE_IRQS);
-	for_each_set_bit(i, vcpu->kvm->arch.vgic.gicv5_vm.vgic_ppi_mask,
-			 VGIC_V5_NR_PRIVATE_IRQS) {
+	for_each_visible_v5_ppi(i, vcpu->kvm) {
 		u32 intid = vgic_v5_make_ppi(i);
 		struct vgic_irq *irq;
 
diff --git a/arch/arm64/kvm/vgic/vgic.h b/arch/arm64/kvm/vgic/vgic.h
index 9d941241c8a2b..f45f7e3ec4d6e 100644
--- a/arch/arm64/kvm/vgic/vgic.h
+++ b/arch/arm64/kvm/vgic/vgic.h
@@ -378,6 +378,9 @@ void vgic_v5_get_vmcr(struct kvm_vcpu *vcpu, struct vgic_vmcr *vmcr);
 void vgic_v5_restore_state(struct kvm_vcpu *vcpu);
 void vgic_v5_save_state(struct kvm_vcpu *vcpu);
 
+#define for_each_visible_v5_ppi(__i, __k)		\
+	for_each_set_bit(__i, (__k)->arch.vgic.gicv5_vm.vgic_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS)
+
 static inline int vgic_v3_max_apr_idx(struct kvm_vcpu *vcpu)
 {
 	struct vgic_cpu *cpu_if = &vcpu->arch.vgic_cpu;
-- 
2.47.3

[PATCH v2 02/18] KVM: arm64: vgic-v5: Move PPI caps into kvm_vgic_global_state

[Marc Zyngier]

Constant vgic properties are usually kept in kvm_vgic_global_state,
but the vgic-v5 code does its own thing.

Move the ppi_caps data into the global structure, which has the
modest additional advantage of making it ro_after_init.

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/vgic/vgic-v5.c |  2 +-
 include/kvm/arm_vgic.h        | 10 +++++-----
 2 files changed, 6 insertions(+), 6 deletions(-)

diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index c0d36658ffe74..7c146fccc9689 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -10,7 +10,7 @@
 
 #include "vgic.h"
 
-static struct vgic_v5_ppi_caps ppi_caps;
+#define ppi_caps	kvm_vgic_global_state.vgic_v5_ppi_caps
 
 /*
  * Not all PPIs are guaranteed to be implemented for GICv5. Deterermine which
diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
index 1388dc6028a9a..ea793479ab254 100644
--- a/include/kvm/arm_vgic.h
+++ b/include/kvm/arm_vgic.h
@@ -177,6 +177,11 @@ struct vgic_global {
 	bool			has_gcie_v3_compat;
 
 	u32			ich_vtr_el2;
+
+	/* GICv5 PPI capabilities */
+	struct {
+		DECLARE_BITMAP(impl_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS);
+	} vgic_v5_ppi_caps;
 };
 
 extern struct vgic_global kvm_vgic_global_state;
@@ -492,11 +497,6 @@ struct vgic_v5_cpu_if {
 	struct gicv5_vpe gicv5_vpe;
 };
 
-/* What PPI capabilities does a GICv5 host have */
-struct vgic_v5_ppi_caps {
-	DECLARE_BITMAP(impl_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS);
-};
-
 struct vgic_cpu {
 	/* CPU vif control registers for world switch */
 	union {
-- 
2.47.3

[PATCH v2 03/18] KVM: arm64: vgic-v5: Remove use of __assign_bit() with a constant

[Marc Zyngier]

Using __assign_bit() is very useful when the value of the bit is
not known at compile time. In all other cases, __set_bit() and
__clear_bit() are the correct tool for the job.

This also fixes an odd case of using VGIC_V5_NR_PRIVATE_IRQS as
the bit value...

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/vgic/vgic-v5.c | 16 +++++++---------
 1 file changed, 7 insertions(+), 9 deletions(-)

diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index 7c146fccc9689..4d62b1c31fe8b 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -25,13 +25,13 @@ static void vgic_v5_get_implemented_ppis(void)
 	 * If we have KVM, we have EL2, which means that we have support for the
 	 * EL1 and EL2 Physical & Virtual timers.
 	 */
-	__assign_bit(GICV5_ARCH_PPI_CNTHP, ppi_caps.impl_ppi_mask, 1);
-	__assign_bit(GICV5_ARCH_PPI_CNTV, ppi_caps.impl_ppi_mask, 1);
-	__assign_bit(GICV5_ARCH_PPI_CNTHV, ppi_caps.impl_ppi_mask, 1);
-	__assign_bit(GICV5_ARCH_PPI_CNTP, ppi_caps.impl_ppi_mask, 1);
+	__set_bit(GICV5_ARCH_PPI_CNTHP, ppi_caps.impl_ppi_mask);
+	__set_bit(GICV5_ARCH_PPI_CNTV, ppi_caps.impl_ppi_mask);
+	__set_bit(GICV5_ARCH_PPI_CNTHV, ppi_caps.impl_ppi_mask);
+	__set_bit(GICV5_ARCH_PPI_CNTP, ppi_caps.impl_ppi_mask);
 
 	/* The SW_PPI should be available */
-	__assign_bit(GICV5_ARCH_PPI_SW_PPI, ppi_caps.impl_ppi_mask, 1);
+	__set_bit(GICV5_ARCH_PPI_SW_PPI, ppi_caps.impl_ppi_mask);
 
 	/* The PMUIRQ is available if we have the PMU */
 	__assign_bit(GICV5_ARCH_PPI_PMUIRQ, ppi_caps.impl_ppi_mask, system_supports_pmuv3());
@@ -146,9 +146,7 @@ int vgic_v5_init(struct kvm *kvm)
 	/* We only allow userspace to drive the SW_PPI, if it is implemented. */
 	bitmap_zero(kvm->arch.vgic.gicv5_vm.userspace_ppis,
 		    VGIC_V5_NR_PRIVATE_IRQS);
-	__assign_bit(GICV5_ARCH_PPI_SW_PPI,
-		     kvm->arch.vgic.gicv5_vm.userspace_ppis,
-		     VGIC_V5_NR_PRIVATE_IRQS);
+	__set_bit(GICV5_ARCH_PPI_SW_PPI, kvm->arch.vgic.gicv5_vm.userspace_ppis);
 	bitmap_and(kvm->arch.vgic.gicv5_vm.userspace_ppis,
 		   kvm->arch.vgic.gicv5_vm.userspace_ppis,
 		   ppi_caps.impl_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS);
@@ -197,7 +195,7 @@ int vgic_v5_finalize_ppi_state(struct kvm *kvm)
 		/* Expose PPIs with an owner or the SW_PPI, only */
 		scoped_guard(raw_spinlock_irqsave, &irq->irq_lock) {
 			if (irq->owner || i == GICV5_ARCH_PPI_SW_PPI) {
-				__assign_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask, 1);
+				__set_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask);
 				__assign_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_hmr,
 					     irq->config == VGIC_CONFIG_LEVEL);
 			}
-- 
2.47.3

Re: [PATCH v2 03/18] KVM: arm64: vgic-v5: Remove use of __assign_bit() with a constant

[Joey Gouly]

On Wed, May 20, 2026 at 10:19:34AM +0100, Marc Zyngier wrote:
> Using __assign_bit() is very useful when the value of the bit is
> not known at compile time. In all other cases, __set_bit() and
> __clear_bit() are the correct tool for the job.
> 
> This also fixes an odd case of using VGIC_V5_NR_PRIVATE_IRQS as
> the bit value...
> 
> Signed-off-by: Marc Zyngier <maz@kernel.org>

Reviewed-by: Joey Gouly <joey.gouly@arm.com>

> ---
>  arch/arm64/kvm/vgic/vgic-v5.c | 16 +++++++---------
>  1 file changed, 7 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
> index 7c146fccc9689..4d62b1c31fe8b 100644
> --- a/arch/arm64/kvm/vgic/vgic-v5.c
> +++ b/arch/arm64/kvm/vgic/vgic-v5.c
> @@ -25,13 +25,13 @@ static void vgic_v5_get_implemented_ppis(void)
>  	 * If we have KVM, we have EL2, which means that we have support for the
>  	 * EL1 and EL2 Physical & Virtual timers.
>  	 */
> -	__assign_bit(GICV5_ARCH_PPI_CNTHP, ppi_caps.impl_ppi_mask, 1);
> -	__assign_bit(GICV5_ARCH_PPI_CNTV, ppi_caps.impl_ppi_mask, 1);
> -	__assign_bit(GICV5_ARCH_PPI_CNTHV, ppi_caps.impl_ppi_mask, 1);
> -	__assign_bit(GICV5_ARCH_PPI_CNTP, ppi_caps.impl_ppi_mask, 1);
> +	__set_bit(GICV5_ARCH_PPI_CNTHP, ppi_caps.impl_ppi_mask);
> +	__set_bit(GICV5_ARCH_PPI_CNTV, ppi_caps.impl_ppi_mask);
> +	__set_bit(GICV5_ARCH_PPI_CNTHV, ppi_caps.impl_ppi_mask);
> +	__set_bit(GICV5_ARCH_PPI_CNTP, ppi_caps.impl_ppi_mask);
>  
>  	/* The SW_PPI should be available */
> -	__assign_bit(GICV5_ARCH_PPI_SW_PPI, ppi_caps.impl_ppi_mask, 1);
> +	__set_bit(GICV5_ARCH_PPI_SW_PPI, ppi_caps.impl_ppi_mask);
>  
>  	/* The PMUIRQ is available if we have the PMU */
>  	__assign_bit(GICV5_ARCH_PPI_PMUIRQ, ppi_caps.impl_ppi_mask, system_supports_pmuv3());
> @@ -146,9 +146,7 @@ int vgic_v5_init(struct kvm *kvm)
>  	/* We only allow userspace to drive the SW_PPI, if it is implemented. */
>  	bitmap_zero(kvm->arch.vgic.gicv5_vm.userspace_ppis,
>  		    VGIC_V5_NR_PRIVATE_IRQS);
> -	__assign_bit(GICV5_ARCH_PPI_SW_PPI,
> -		     kvm->arch.vgic.gicv5_vm.userspace_ppis,
> -		     VGIC_V5_NR_PRIVATE_IRQS);
> +	__set_bit(GICV5_ARCH_PPI_SW_PPI, kvm->arch.vgic.gicv5_vm.userspace_ppis);
>  	bitmap_and(kvm->arch.vgic.gicv5_vm.userspace_ppis,
>  		   kvm->arch.vgic.gicv5_vm.userspace_ppis,
>  		   ppi_caps.impl_ppi_mask, VGIC_V5_NR_PRIVATE_IRQS);
> @@ -197,7 +195,7 @@ int vgic_v5_finalize_ppi_state(struct kvm *kvm)
>  		/* Expose PPIs with an owner or the SW_PPI, only */
>  		scoped_guard(raw_spinlock_irqsave, &irq->irq_lock) {
>  			if (irq->owner || i == GICV5_ARCH_PPI_SW_PPI) {
> -				__assign_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask, 1);
> +				__set_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_mask);
>  				__assign_bit(i, kvm->arch.vgic.gicv5_vm.vgic_ppi_hmr,
>  					     irq->config == VGIC_CONFIG_LEVEL);
>  			}
> -- 
> 2.47.3
> 

[PATCH v2 04/18] KVM: arm64: vgic-v5: Drop pointless ARM64_HAS_GICV5_CPUIF check

[Marc Zyngier]

vgic_v5_get_implemented_ppis() can only be called when we have
a GICv5, by construction.

Remove the pointless check against ARM64_HAS_GICV5_CPUIF.

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/vgic/vgic-v5.c | 3 ---
 1 file changed, 3 deletions(-)

diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index 4d62b1c31fe8b..0101ec3f55283 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -18,9 +18,6 @@
  */
 static void vgic_v5_get_implemented_ppis(void)
 {
-	if (!cpus_have_final_cap(ARM64_HAS_GICV5_CPUIF))
-		return;
-
 	/*
 	 * If we have KVM, we have EL2, which means that we have support for the
 	 * EL1 and EL2 Physical & Virtual timers.
-- 
2.47.3

[PATCH v2 05/18] KVM: arm64: vgic: Constify struct irq_ops usage

[Marc Zyngier]

vgic-v5 has introduced much more prevalent usage of the struct
irq_ops mechanism.

In the process, it becomes evident that suffers from two related
problems:

- it contains flags, rather than only callbacks
- it is mutable, because we need to update the above flags

Swap the flags for a helper retrieving the flags, and make all
irq_ops const, something that is slightly satisfying.

Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/arch_timer.c   | 14 +++++++++-----
 arch/arm64/kvm/vgic/vgic-v5.c |  2 +-
 arch/arm64/kvm/vgic/vgic.c    |  2 +-
 include/kvm/arm_vgic.h        |  9 +++++----
 4 files changed, 16 insertions(+), 11 deletions(-)

diff --git a/arch/arm64/kvm/arch_timer.c b/arch/arm64/kvm/arch_timer.c
index cbea4d9ee9552..f003df76fdda7 100644
--- a/arch/arm64/kvm/arch_timer.c
+++ b/arch/arm64/kvm/arch_timer.c
@@ -52,11 +52,17 @@ static u64 kvm_arm_timer_read(struct kvm_vcpu *vcpu,
 			      enum kvm_arch_timer_regs treg);
 static bool kvm_arch_timer_get_input_level(int vintid);
 
-static struct irq_ops arch_timer_irq_ops = {
+static unsigned long kvm_arch_timer_get_irq_flags(void)
+{
+	return kvm_vgic_global_state.no_hw_deactivation ? VGIC_IRQ_SW_RESAMPLE : 0;
+}
+
+static const struct irq_ops arch_timer_irq_ops = {
+	.get_flags	 = kvm_arch_timer_get_irq_flags,
 	.get_input_level = kvm_arch_timer_get_input_level,
 };
 
-static struct irq_ops arch_timer_irq_ops_vgic_v5 = {
+static const struct irq_ops arch_timer_irq_ops_vgic_v5 = {
 	.get_input_level = kvm_arch_timer_get_input_level,
 	.queue_irq_unlock = vgic_v5_ppi_queue_irq_unlock,
 	.set_direct_injection = vgic_v5_set_ppi_dvi,
@@ -1392,8 +1398,6 @@ static int kvm_irq_init(struct arch_timer_kvm_info *info)
 			return -ENOMEM;
 		}
 
-		if (kvm_vgic_global_state.no_hw_deactivation)
-			arch_timer_irq_ops.flags |= VGIC_IRQ_SW_RESAMPLE;
 		WARN_ON(irq_domain_push_irq(domain, host_vtimer_irq,
 					    (void *)TIMER_VTIMER));
 	}
@@ -1591,8 +1595,8 @@ static bool kvm_arch_timer_get_input_level(int vintid)
 int kvm_timer_enable(struct kvm_vcpu *vcpu)
 {
 	struct arch_timer_cpu *timer = vcpu_timer(vcpu);
+	const struct irq_ops *ops;
 	struct timer_map map;
-	struct irq_ops *ops;
 	int ret;
 
 	if (timer->enabled)
diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index 0101ec3f55283..757484d2493b2 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -285,7 +285,7 @@ void vgic_v5_set_ppi_dvi(struct kvm_vcpu *vcpu, struct vgic_irq *irq, bool dvi)
 	__assign_bit(ppi, cpu_if->vgic_ppi_dvir, dvi);
 }
 
-static struct irq_ops vgic_v5_ppi_irq_ops = {
+static const struct irq_ops vgic_v5_ppi_irq_ops = {
 	.queue_irq_unlock = vgic_v5_ppi_queue_irq_unlock,
 	.set_direct_injection = vgic_v5_set_ppi_dvi,
 };
diff --git a/arch/arm64/kvm/vgic/vgic.c b/arch/arm64/kvm/vgic/vgic.c
index 1e9fe8764584d..3ac6d49bc4876 100644
--- a/arch/arm64/kvm/vgic/vgic.c
+++ b/arch/arm64/kvm/vgic/vgic.c
@@ -573,7 +573,7 @@ int kvm_vgic_inject_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 }
 
 void kvm_vgic_set_irq_ops(struct kvm_vcpu *vcpu, u32 vintid,
-			  struct irq_ops *ops)
+			  const struct irq_ops *ops)
 {
 	struct vgic_irq *irq = vgic_get_vcpu_irq(vcpu, vintid);
 
diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
index ea793479ab254..fe49fb56dc3c9 100644
--- a/include/kvm/arm_vgic.h
+++ b/include/kvm/arm_vgic.h
@@ -205,7 +205,7 @@ struct vgic_irq;
  */
 struct irq_ops {
 	/* Per interrupt flags for special-cased interrupts */
-	unsigned long flags;
+	unsigned long (*get_flags)(void);
 
 #define VGIC_IRQ_SW_RESAMPLE	BIT(0)	/* Clear the active state for resampling */
 
@@ -271,7 +271,7 @@ struct vgic_irq {
 	u8 priority;
 	u8 group;			/* 0 == group 0, 1 == group 1 */
 
-	struct irq_ops *ops;
+	const struct irq_ops *ops;
 
 	void *owner;			/* Opaque pointer to reserve an interrupt
 					   for in-kernel devices. */
@@ -279,7 +279,8 @@ struct vgic_irq {
 
 static inline bool vgic_irq_needs_resampling(struct vgic_irq *irq)
 {
-	return irq->ops && (irq->ops->flags & VGIC_IRQ_SW_RESAMPLE);
+	return irq->ops && irq->ops->get_flags &&
+	       (irq->ops->get_flags() & VGIC_IRQ_SW_RESAMPLE);
 }
 
 struct vgic_register_region;
@@ -557,7 +558,7 @@ void kvm_vgic_init_cpu_hardware(void);
 int kvm_vgic_inject_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 			unsigned int intid, bool level, void *owner);
 void kvm_vgic_set_irq_ops(struct kvm_vcpu *vcpu, u32 vintid,
-			  struct irq_ops *ops);
+			  const struct irq_ops *ops);
 void kvm_vgic_clear_irq_ops(struct kvm_vcpu *vcpu, u32 vintid);
 int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
 			  u32 vintid);
-- 
2.47.3

[PATCH v2 06/18] KVM: arm64: vgic: Consolidate vgic_allocate_private_irqs_locked()

[Marc Zyngier]

vgic_allocate_private_irqs_locked() calls two helpers, oddly named
vgic_{,v5_}allocate_private_irq().

Not only these helpers don't allocate anything, but they also
contain duplicate init code that would be better placed in the
caller.

Consolidate the common init code in the caller, rename the helpers
to vgic_{,v5_}setup_private_irq(), and pass the irq pointer around
instead of the index of the interrupt.

Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/vgic/vgic-init.c | 45 +++++++++++++--------------------
 1 file changed, 18 insertions(+), 27 deletions(-)

diff --git a/arch/arm64/kvm/vgic/vgic-init.c b/arch/arm64/kvm/vgic/vgic-init.c
index 933983bb20052..907057881b26a 100644
--- a/arch/arm64/kvm/vgic/vgic-init.c
+++ b/arch/arm64/kvm/vgic/vgic-init.c
@@ -271,18 +271,12 @@ int kvm_vgic_vcpu_nv_init(struct kvm_vcpu *vcpu)
 	return ret;
 }
 
-static void vgic_allocate_private_irq(struct kvm_vcpu *vcpu, int i, u32 type)
+static void vgic_setup_private_irq(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
+				   u32 type)
 {
-	struct vgic_irq *irq = &vcpu->arch.vgic_cpu.private_irqs[i];
+	irq->intid = irq - &vcpu->arch.vgic_cpu.private_irqs[0];
 
-	INIT_LIST_HEAD(&irq->ap_list);
-	raw_spin_lock_init(&irq->irq_lock);
-	irq->vcpu = NULL;
-	irq->target_vcpu = vcpu;
-	refcount_set(&irq->refcount, 0);
-
-	irq->intid = i;
-	if (vgic_irq_is_sgi(i)) {
+	if (vgic_irq_is_sgi(irq->intid)) {
 		/* SGIs */
 		irq->enabled = 1;
 		irq->config = VGIC_CONFIG_EDGE;
@@ -303,18 +297,11 @@ static void vgic_allocate_private_irq(struct kvm_vcpu *vcpu, int i, u32 type)
 	}
 }
 
-static void vgic_v5_allocate_private_irq(struct kvm_vcpu *vcpu, int i, u32 type)
+static void vgic_v5_setup_private_irq(struct kvm_vcpu *vcpu, struct vgic_irq *irq)
 {
-	struct vgic_irq *irq = &vcpu->arch.vgic_cpu.private_irqs[i];
-	u32 intid = vgic_v5_make_ppi(i);
-
-	INIT_LIST_HEAD(&irq->ap_list);
-	raw_spin_lock_init(&irq->irq_lock);
-	irq->vcpu = NULL;
-	irq->target_vcpu = vcpu;
-	refcount_set(&irq->refcount, 0);
+	int i = irq - &vcpu->arch.vgic_cpu.private_irqs[0];
 
-	irq->intid = intid;
+	irq->intid = vgic_v5_make_ppi(i);
 
 	/* The only Edge architected PPI is the SW_PPI */
 	if (i == GICV5_ARCH_PPI_SW_PPI)
@@ -323,7 +310,7 @@ static void vgic_v5_allocate_private_irq(struct kvm_vcpu *vcpu, int i, u32 type)
 		irq->config = VGIC_CONFIG_LEVEL;
 
 	/* Register the GICv5-specific PPI ops */
-	vgic_v5_set_ppi_ops(vcpu, intid);
+	vgic_v5_set_ppi_ops(vcpu, irq->intid);
 }
 
 static int vgic_allocate_private_irqs_locked(struct kvm_vcpu *vcpu, u32 type)
@@ -349,15 +336,19 @@ static int vgic_allocate_private_irqs_locked(struct kvm_vcpu *vcpu, u32 type)
 	if (!vgic_cpu->private_irqs)
 		return -ENOMEM;
 
-	/*
-	 * Enable and configure all SGIs to be edge-triggered and
-	 * configure all PPIs as level-triggered.
-	 */
 	for (i = 0; i < num_private_irqs; i++) {
+		struct vgic_irq *irq = &vcpu->arch.vgic_cpu.private_irqs[i];
+
+		INIT_LIST_HEAD(&irq->ap_list);
+		raw_spin_lock_init(&irq->irq_lock);
+		irq->vcpu = NULL;
+		irq->target_vcpu = vcpu;
+		refcount_set(&irq->refcount, 0);
+
 		if (vgic_is_v5(vcpu->kvm))
-			vgic_v5_allocate_private_irq(vcpu, i, type);
+			vgic_v5_setup_private_irq(vcpu, irq);
 		else
-			vgic_allocate_private_irq(vcpu, i, type);
+			vgic_setup_private_irq(vcpu, irq, type);
 	}
 
 	return 0;
-- 
2.47.3

[PATCH v2 07/18] KVM: arm64: vgic-v5: Drop defensive checks from vgic_v5_ppi_queue_irq_unlock()

[Marc Zyngier]

vgic_v5_ppi_queue_irq_unlock() performs a bunch of sanity checks
that are pretty pointless as there is no code path that can
result in these invariants to be violated. And if they are, a nice
crash is just as instructive than a warning.

Drop what is evidently debug code and simplify the whole thing.

Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/vgic/vgic-v5.c | 16 +++-------------
 1 file changed, 3 insertions(+), 13 deletions(-)

diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index 757484d2493b2..7916bd8d564ef 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -238,9 +238,9 @@ static u32 vgic_v5_get_effective_priority_mask(struct kvm_vcpu *vcpu)
 
 /*
  * For GICv5, the PPIs are mostly directly managed by the hardware. We (the
- * hypervisor) handle the pending, active, enable state save/restore, but don't
- * need the PPIs to be queued on a per-VCPU AP list. Therefore, sanity check the
- * state, unlock, and return.
+ * hypervisor) handle the pending, active, enable state save/restore, but
+ * don't need the PPIs to be queued on a per-VCPU AP list. Therefore,
+ * unlock, kick the vcpu and return.
  */
 bool vgic_v5_ppi_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq,
 				  unsigned long flags)
@@ -250,12 +250,7 @@ bool vgic_v5_ppi_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq,
 
 	lockdep_assert_held(&irq->irq_lock);
 
-	if (WARN_ON_ONCE(!__irq_is_ppi(KVM_DEV_TYPE_ARM_VGIC_V5, irq->intid)))
-		goto out_unlock_fail;
-
 	vcpu = irq->target_vcpu;
-	if (WARN_ON_ONCE(!vcpu))
-		goto out_unlock_fail;
 
 	raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
 
@@ -264,11 +259,6 @@ bool vgic_v5_ppi_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq,
 	kvm_vcpu_kick(vcpu);
 
 	return true;
-
-out_unlock_fail:
-	raw_spin_unlock_irqrestore(&irq->irq_lock, flags);
-
-	return false;
 }
 
 /*
-- 
2.47.3

[PATCH v2 08/18] KVM: arm64: vgic: Rationalise per-CPU irq accessor

[Marc Zyngier]

Despite adding the necessary infrastructure to identify irq types,
vgic_get_vcpu_irq() treats GICv5 PPIs in a special way, which
impairs the readability of the code.

Use the existing irq classifiers to handle per-CPU irqs for all
vgic types, and let the normal control flow reach global interrupt
handling without any v5-specific path.

Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/vgic/vgic.c | 25 ++++++++++++-------------
 1 file changed, 12 insertions(+), 13 deletions(-)

diff --git a/arch/arm64/kvm/vgic/vgic.c b/arch/arm64/kvm/vgic/vgic.c
index 3ac6d49bc4876..b697678d68b01 100644
--- a/arch/arm64/kvm/vgic/vgic.c
+++ b/arch/arm64/kvm/vgic/vgic.c
@@ -106,24 +106,23 @@ struct vgic_irq *vgic_get_irq(struct kvm *kvm, u32 intid)
 
 struct vgic_irq *vgic_get_vcpu_irq(struct kvm_vcpu *vcpu, u32 intid)
 {
+	enum kvm_device_type type;
+
 	if (WARN_ON(!vcpu))
 		return NULL;
 
-	if (vgic_is_v5(vcpu->kvm)) {
-		u32 int_num, hwirq_id;
-
-		if (!__irq_is_ppi(KVM_DEV_TYPE_ARM_VGIC_V5, intid))
-			return NULL;
-
-		hwirq_id = FIELD_GET(GICV5_HWIRQ_ID, intid);
-		int_num = array_index_nospec(hwirq_id, VGIC_V5_NR_PRIVATE_IRQS);
+	type = vcpu->kvm->arch.vgic.vgic_model;
 
-		return &vcpu->arch.vgic_cpu.private_irqs[int_num];
-	}
+	if (__irq_is_sgi(type, intid) || __irq_is_ppi(type, intid)) {
+		switch (type) {
+		case KVM_DEV_TYPE_ARM_VGIC_V5:
+			intid = vgic_v5_get_hwirq_id(intid);
+			intid = array_index_nospec(intid, VGIC_V5_NR_PRIVATE_IRQS);
+			break;
+		default:
+			intid = array_index_nospec(intid, VGIC_NR_PRIVATE_IRQS);
+		}
 
-	/* SGIs and PPIs */
-	if (intid < VGIC_NR_PRIVATE_IRQS) {
-		intid = array_index_nospec(intid, VGIC_NR_PRIVATE_IRQS);
 		return &vcpu->arch.vgic_cpu.private_irqs[intid];
 	}
 
-- 
2.47.3

[PATCH v2 09/18] KVM: arm64: vgic-v5: Limit support to 64 PPIs

[Marc Zyngier]

Although we have some code supporting 128 PPIs, the only supported
configuration is 64 PPIs. There is no way to test the 128 PPI code,
so it is bound to bitrot very quickly.

Given that KVM/arm64's goal has always been to stick to non-IMPDEF
behaviours, drop the 128 PPI support. Someone motivated enough and
with very strong arguments can always bring it back -- it's all in
the git history.

Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/hyp/vgic-v5-sr.c       | 82 ++++++---------------------
 arch/arm64/kvm/sys_regs.c             | 17 +++---
 arch/arm64/kvm/vgic/vgic-kvm-device.c |  9 +--
 3 files changed, 26 insertions(+), 82 deletions(-)

diff --git a/arch/arm64/kvm/hyp/vgic-v5-sr.c b/arch/arm64/kvm/hyp/vgic-v5-sr.c
index 47e6bcd437029..6d69dfe89a96c 100644
--- a/arch/arm64/kvm/hyp/vgic-v5-sr.c
+++ b/arch/arm64/kvm/hyp/vgic-v5-sr.c
@@ -30,10 +30,9 @@ void __vgic_v5_save_ppi_state(struct vgic_v5_cpu_if *cpu_if)
 {
 	/*
 	 * The following code assumes that the bitmap storage that we have for
-	 * PPIs is either 64 (architected PPIs, only) or 128 bits (architected &
-	 * impdef PPIs).
+	 * PPIs is either 64 (architected PPIs, only).
 	 */
-	BUILD_BUG_ON(VGIC_V5_NR_PRIVATE_IRQS % 64);
+	BUILD_BUG_ON(VGIC_V5_NR_PRIVATE_IRQS != 64);
 
 	bitmap_write(host_data_ptr(vgic_v5_ppi_state)->activer_exit,
 		     read_sysreg_s(SYS_ICH_PPI_ACTIVER0_EL2), 0, 64);
@@ -49,22 +48,6 @@ void __vgic_v5_save_ppi_state(struct vgic_v5_cpu_if *cpu_if)
 	cpu_if->vgic_ppi_priorityr[6] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR6_EL2);
 	cpu_if->vgic_ppi_priorityr[7] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR7_EL2);
 
-	if (VGIC_V5_NR_PRIVATE_IRQS == 128) {
-		bitmap_write(host_data_ptr(vgic_v5_ppi_state)->activer_exit,
-			     read_sysreg_s(SYS_ICH_PPI_ACTIVER1_EL2), 64, 64);
-		bitmap_write(host_data_ptr(vgic_v5_ppi_state)->pendr,
-			     read_sysreg_s(SYS_ICH_PPI_PENDR1_EL2), 64, 64);
-
-		cpu_if->vgic_ppi_priorityr[8] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR8_EL2);
-		cpu_if->vgic_ppi_priorityr[9] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR9_EL2);
-		cpu_if->vgic_ppi_priorityr[10] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR10_EL2);
-		cpu_if->vgic_ppi_priorityr[11] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR11_EL2);
-		cpu_if->vgic_ppi_priorityr[12] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR12_EL2);
-		cpu_if->vgic_ppi_priorityr[13] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR13_EL2);
-		cpu_if->vgic_ppi_priorityr[14] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR14_EL2);
-		cpu_if->vgic_ppi_priorityr[15] = read_sysreg_s(SYS_ICH_PPI_PRIORITYR15_EL2);
-	}
-
 	/* Now that we are done, disable DVI */
 	write_sysreg_s(0, SYS_ICH_PPI_DVIR0_EL2);
 	write_sysreg_s(0, SYS_ICH_PPI_DVIR1_EL2);
@@ -74,9 +57,6 @@ void __vgic_v5_restore_ppi_state(struct vgic_v5_cpu_if *cpu_if)
 {
 	DECLARE_BITMAP(pendr, VGIC_V5_NR_PRIVATE_IRQS);
 
-	/* We assume 64 or 128 PPIs - see above comment */
-	BUILD_BUG_ON(VGIC_V5_NR_PRIVATE_IRQS % 64);
-
 	/* Enable DVI so that the guest's interrupt config takes over */
 	write_sysreg_s(bitmap_read(cpu_if->vgic_ppi_dvir, 0, 64),
 		       SYS_ICH_PPI_DVIR0_EL2);
@@ -108,50 +88,20 @@ void __vgic_v5_restore_ppi_state(struct vgic_v5_cpu_if *cpu_if)
 	write_sysreg_s(cpu_if->vgic_ppi_priorityr[7],
 		       SYS_ICH_PPI_PRIORITYR7_EL2);
 
-	if (VGIC_V5_NR_PRIVATE_IRQS == 128) {
-		/* Enable DVI so that the guest's interrupt config takes over */
-		write_sysreg_s(bitmap_read(cpu_if->vgic_ppi_dvir, 64, 64),
-			       SYS_ICH_PPI_DVIR1_EL2);
-
-		write_sysreg_s(bitmap_read(cpu_if->vgic_ppi_activer, 64, 64),
-			       SYS_ICH_PPI_ACTIVER1_EL2);
-		write_sysreg_s(bitmap_read(cpu_if->vgic_ppi_enabler, 64, 64),
-			       SYS_ICH_PPI_ENABLER1_EL2);
-		write_sysreg_s(bitmap_read(pendr, 64, 64),
-			       SYS_ICH_PPI_PENDR1_EL2);
-
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[8],
-			       SYS_ICH_PPI_PRIORITYR8_EL2);
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[9],
-			       SYS_ICH_PPI_PRIORITYR9_EL2);
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[10],
-			       SYS_ICH_PPI_PRIORITYR10_EL2);
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[11],
-			       SYS_ICH_PPI_PRIORITYR11_EL2);
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[12],
-			       SYS_ICH_PPI_PRIORITYR12_EL2);
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[13],
-			       SYS_ICH_PPI_PRIORITYR13_EL2);
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[14],
-			       SYS_ICH_PPI_PRIORITYR14_EL2);
-		write_sysreg_s(cpu_if->vgic_ppi_priorityr[15],
-			       SYS_ICH_PPI_PRIORITYR15_EL2);
-	} else {
-		write_sysreg_s(0, SYS_ICH_PPI_DVIR1_EL2);
-
-		write_sysreg_s(0, SYS_ICH_PPI_ACTIVER1_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_ENABLER1_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PENDR1_EL2);
-
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR8_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR9_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR10_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR11_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR12_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR13_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR14_EL2);
-		write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR15_EL2);
-	}
+	write_sysreg_s(0, SYS_ICH_PPI_DVIR1_EL2);
+
+	write_sysreg_s(0, SYS_ICH_PPI_ACTIVER1_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_ENABLER1_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PENDR1_EL2);
+
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR8_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR9_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR10_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR11_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR12_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR13_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR14_EL2);
+	write_sysreg_s(0, SYS_ICH_PPI_PRIORITYR15_EL2);
 }
 
 void __vgic_v5_save_state(struct vgic_v5_cpu_if *cpu_if)
diff --git a/arch/arm64/kvm/sys_regs.c b/arch/arm64/kvm/sys_regs.c
index 513f5f1429b5f..6083a1b23dbf9 100644
--- a/arch/arm64/kvm/sys_regs.c
+++ b/arch/arm64/kvm/sys_regs.c
@@ -724,6 +724,7 @@ static bool access_gicv5_ppi_enabler(struct kvm_vcpu *vcpu,
 {
 	unsigned long *mask = vcpu->kvm->arch.vgic.gicv5_vm.vgic_ppi_mask;
 	struct vgic_v5_cpu_if *cpu_if = &vcpu->arch.vgic_cpu.vgic_v5;
+	unsigned long reg = p->regval;
 	int i;
 
 	/* We never expect to get here with a read! */
@@ -731,21 +732,17 @@ static bool access_gicv5_ppi_enabler(struct kvm_vcpu *vcpu,
 		return undef_access(vcpu, p, r);
 
 	/*
-	 * If we're only handling architected PPIs and the guest writes to the
-	 * enable for the non-architected PPIs, we just return as there's
-	 * nothing to do at all. We don't even allocate the storage for them in
-	 * this case.
+	 * As we're only handling architected PPIs, the guest writes to the
+	 * enable for the non-architected PPIs just return as there's
+	 * nothing to do at all. We don't even allocate the storage for them.
 	 */
-	if (VGIC_V5_NR_PRIVATE_IRQS == 64 && p->Op2 % 2)
+	if (p->Op2 % 2)
 		return true;
 
 	/*
-	 * Merge the raw guest write into out bitmap at an offset of either 0 or
-	 * 64, then and it with our PPI mask.
+	 * Merge the raw guest write into out bitmap, anded with our PPI mask.
 	 */
-	bitmap_write(cpu_if->vgic_ppi_enabler, p->regval, 64 * (p->Op2 % 2), 64);
-	bitmap_and(cpu_if->vgic_ppi_enabler, cpu_if->vgic_ppi_enabler, mask,
-		   VGIC_V5_NR_PRIVATE_IRQS);
+	bitmap_and(cpu_if->vgic_ppi_enabler, &reg, mask, VGIC_V5_NR_PRIVATE_IRQS);
 
 	/*
 	 * Sync the change in enable states to the vgic_irqs. We consider all
diff --git a/arch/arm64/kvm/vgic/vgic-kvm-device.c b/arch/arm64/kvm/vgic/vgic-kvm-device.c
index a96c77dccf353..90be99443df3b 100644
--- a/arch/arm64/kvm/vgic/vgic-kvm-device.c
+++ b/arch/arm64/kvm/vgic/vgic-kvm-device.c
@@ -730,18 +730,15 @@ static int vgic_v5_get_userspace_ppis(struct kvm_device *dev,
 	guard(mutex)(&dev->kvm->arch.config_lock);
 
 	/*
-	 * We either support 64 or 128 PPIs. In the former case, we need to
-	 * return 0s for the second 64 bits as we have no storage backing those.
+	 * We only support 64 PPIs, so, we need to return 0s for the
+	 * second 64 bits as we have no storage backing those.
 	 */
 	ret = put_user(bitmap_read(gicv5_vm->userspace_ppis, 0, 64), uaddr);
 	if (ret)
 		return ret;
 	uaddr++;
 
-	if (VGIC_V5_NR_PRIVATE_IRQS == 128)
-		ret = put_user(bitmap_read(gicv5_vm->userspace_ppis, 64, 128), uaddr);
-	else
-		ret = put_user(0, uaddr);
+	ret = put_user(0, uaddr);
 
 	return ret;
 }
-- 
2.47.3

[PATCH v2 10/18] KVM: arm64: vgic-v5: Add missing trap handing for NV triage

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

As things stand, there is no support for Nested Virt with GICv5 guests
yet. However, this is coming and therefore we need to be able to
correctly triage the traps when running with NV.

Add the missing fgtreg lookups required for that to
triage_sysreg_trap(). These are specific to the FGT regs added as part
of GICv5:
   * ICH_HFGRTR_EL2
   * ICH_HFGWTR_EL2
   * ICH_HFGITR_EL2

Fixes: 9d6d9514c08f "(KVM: arm64: gic-v5: Support GICv5 FGTs & FGUs")
Link: https://sashiko.dev/#/patchset/20260319154937.3619520-1-sascha.bischoff%40arm.com
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/emulate-nested.c | 8 ++++++++
 1 file changed, 8 insertions(+)

diff --git a/arch/arm64/kvm/emulate-nested.c b/arch/arm64/kvm/emulate-nested.c
index dba7ced74ca5e..a4eb36b4c4421 100644
--- a/arch/arm64/kvm/emulate-nested.c
+++ b/arch/arm64/kvm/emulate-nested.c
@@ -2631,6 +2631,14 @@ bool triage_sysreg_trap(struct kvm_vcpu *vcpu, int *sr_index)
 		fgtreg = HFGITR2_EL2;
 		break;
 
+	case ICH_HFGRTR_GROUP:
+		fgtreg = is_read ? ICH_HFGRTR_EL2 : ICH_HFGWTR_EL2;
+		break;
+
+	case ICH_HFGITR_GROUP:
+		fgtreg = ICH_HFGITR_EL2;
+		break;
+
 	default:
 		/* Something is really wrong, bail out */
 		WARN_ONCE(1, "Bad FGT group (encoding %08x, config %016llx)\n",
-- 
2.47.3

[PATCH v2 11/18] KVM: arm64: vgic-v5: Atomically assign bits to PPI DVI bitmap

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

For GICv5 guests we make use of the DVI mechanism for PPIs where
possible.  When mapping a virtual irq to a physical one for a GICv5
guest, the corresponding bit in the DVI bitmap is set. When unmapping,
said bit is cleared again. The key user of this mechanism is the arch
timer.

The existing code used the non-atomic __assign_bit() rather than doing
the update atomically. This could technically result in losing state
if a second PPI's DVI bit were being manipulated concurrently. Each
individual bit within the DVI bitmap is guarded using
vgic_irq->irq_lock, but there's no locking for the overall
bitmap. Therefore, switch to using the atomic assign_bit() function
instead.

Fixes: 5a98d0e17e59 ("KVM: arm64: gic-v5: Implement direct injection of PPIs")
Link: https://sashiko.dev/#/patchset/20260319154937.3619520-1-sascha.bischoff%40arm.com
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/vgic/vgic-v5.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arm64/kvm/vgic/vgic-v5.c b/arch/arm64/kvm/vgic/vgic-v5.c
index 7916bd8d564ef..d4789ff3e7402 100644
--- a/arch/arm64/kvm/vgic/vgic-v5.c
+++ b/arch/arm64/kvm/vgic/vgic-v5.c
@@ -272,7 +272,7 @@ void vgic_v5_set_ppi_dvi(struct kvm_vcpu *vcpu, struct vgic_irq *irq, bool dvi)
 	lockdep_assert_held(&irq->irq_lock);
 
 	ppi = vgic_v5_get_hwirq_id(irq->intid);
-	__assign_bit(ppi, cpu_if->vgic_ppi_dvir, dvi);
+	assign_bit(ppi, cpu_if->vgic_ppi_dvir, dvi);
 }
 
 static const struct irq_ops vgic_v5_ppi_irq_ops = {
-- 
2.47.3

[PATCH v2 12/18] KVM: arm64: selftests: Add missing GIC CDEN to no-vgic-v5 selftest

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

The selftest mistakenly omitted the GIC CDEN instruction from the
testing. Add it in.

Fixes: ce29261ec648 ("KVM: arm64: selftests: Add no-vgic-v5 selftest")
Reviewed-by: Joey Gouly <joey.gouly@arm.com>
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 tools/testing/selftests/kvm/arm64/no-vgic.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/tools/testing/selftests/kvm/arm64/no-vgic.c b/tools/testing/selftests/kvm/arm64/no-vgic.c
index 25b2e3222f685..ab57902ce4293 100644
--- a/tools/testing/selftests/kvm/arm64/no-vgic.c
+++ b/tools/testing/selftests/kvm/arm64/no-vgic.c
@@ -159,6 +159,7 @@ static void guest_code_gicv5(void)
 	check_gicv5_gic_op(CDAFF);
 	check_gicv5_gic_op(CDDI);
 	check_gicv5_gic_op(CDDIS);
+	check_gicv5_gic_op(CDEN);
 	check_gicv5_gic_op(CDEOI);
 	check_gicv5_gic_op(CDHM);
 	check_gicv5_gic_op(CDPEND);
-- 
2.47.3

[PATCH v2 13/18] KVM: arm64: selftests: Cleanup unused vars in GICv5 PPI selftest

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

Clean up a set of unused variables around the size of the guest's PA
space as they are completely irrelevant for GICv5 when only
considering PPIs.

Fixes: 0a9f38bf612b ("KVM: arm64: selftests: Introduce a minimal GICv5 PPI selftest")
Link: https://sashiko.dev/#/patchset/20260319154937.3619520-1-sascha.bischoff%40arm.com
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 tools/testing/selftests/kvm/arm64/vgic_v5.c | 6 ------
 1 file changed, 6 deletions(-)

diff --git a/tools/testing/selftests/kvm/arm64/vgic_v5.c b/tools/testing/selftests/kvm/arm64/vgic_v5.c
index d785b660d8476..a8707120de0d8 100644
--- a/tools/testing/selftests/kvm/arm64/vgic_v5.c
+++ b/tools/testing/selftests/kvm/arm64/vgic_v5.c
@@ -20,8 +20,6 @@ struct vm_gic {
 	u32 gic_dev_type;
 };
 
-static u64 max_phys_size;
-
 #define GUEST_CMD_IRQ_CDIA	10
 #define GUEST_CMD_IRQ_DIEOI	11
 #define GUEST_CMD_IS_AWAKE	12
@@ -208,13 +206,9 @@ void run_tests(u32 gic_dev_type)
 int main(int ac, char **av)
 {
 	int ret;
-	int pa_bits;
 
 	test_disable_default_vgic();
 
-	pa_bits = vm_guest_mode_params[VM_MODE_DEFAULT].pa_bits;
-	max_phys_size = 1ULL << pa_bits;
-
 	ret = test_kvm_device(KVM_DEV_TYPE_ARM_VGIC_V5);
 	if (ret) {
 		pr_info("No GICv5 support; Not running GIC_v5 tests.\n");
-- 
2.47.3

[PATCH v2 14/18] KVM: arm64: selftests: Improve error handling for GICv5 PPI selftest

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

Cases where the KVM_RUN ioctl returned an error were wrongly reported
as incorrect ucalls. Furthermore, potential failures when calling
KVM_IRQ_LINE were being hidden.

Improve the error handling to correctly propagate the error in both
cases.

Fixes: 0a9f38bf612b ("KVM: arm64: selftests: Introduce a minimal GICv5 PPI selftest")
Link: https://sashiko.dev/#/patchset/20260319154937.3619520-1-sascha.bischoff%40arm.com
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 tools/testing/selftests/kvm/arm64/vgic_v5.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/kvm/arm64/vgic_v5.c b/tools/testing/selftests/kvm/arm64/vgic_v5.c
index a8707120de0d8..96cfd6bb32f6f 100644
--- a/tools/testing/selftests/kvm/arm64/vgic_v5.c
+++ b/tools/testing/selftests/kvm/arm64/vgic_v5.c
@@ -129,6 +129,8 @@ static void test_vgic_v5_ppis(u32 gic_dev_type)
 
 	while (1) {
 		ret = run_vcpu(vcpus[0]);
+		if (ret)
+			break;
 
 		switch (get_ucall(vcpus[0], &uc)) {
 		case UCALL_SYNC:
@@ -144,7 +146,7 @@ static void test_vgic_v5_ppis(u32 gic_dev_type)
 				irq = FIELD_PREP(KVM_ARM_IRQ_NUM_MASK, 3);
 				irq |= KVM_ARM_IRQ_TYPE_PPI << KVM_ARM_IRQ_TYPE_SHIFT;
 
-				_kvm_irq_line(v.vm, irq, level);
+				kvm_irq_line(v.vm, irq, level);
 			} else if (uc.args[1] == GUEST_CMD_IS_AWAKE) {
 				pr_info("Guest skipping WFI due to pending IRQ\n");
 			} else if (uc.args[1] == GUEST_CMD_IRQ_CDIA) {
-- 
2.47.3

[PATCH v2 15/18] Documentation: KVM: Fix typos in VGICv5 documentation

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

Fix two typos in the VGICv5 documentation.

Fixes: d51c978b7d3e ("KVM: arm64: gic-v5: Communicate userspace-driveable PPIs via a UAPI")
Fixes: eb3c4d2c9a4d ("Documentation: KVM: Introduce documentation for VGICv5")
Link: https://sashiko.dev/#/patchset/20260319154937.3619520-1-sascha.bischoff%40arm.com
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 Documentation/virt/kvm/devices/arm-vgic-v5.rst | 6 +++---
 1 file changed, 3 insertions(+), 3 deletions(-)

diff --git a/Documentation/virt/kvm/devices/arm-vgic-v5.rst b/Documentation/virt/kvm/devices/arm-vgic-v5.rst
index 29335ea823fc5..70b9162755c7e 100644
--- a/Documentation/virt/kvm/devices/arm-vgic-v5.rst
+++ b/Documentation/virt/kvm/devices/arm-vgic-v5.rst
@@ -12,8 +12,8 @@ Only one VGIC instance may be instantiated through this API.  The created VGIC
 will act as the VM interrupt controller, requiring emulated user-space devices
 to inject interrupts to the VGIC instead of directly to CPUs.
 
-Creating a guest GICv5 device requires a host GICv5 host.  The current VGICv5
-device only supports PPI interrupts.  These can either be injected from emulated
+Creating a guest GICv5 device requires a GICv5 host.  The current VGICv5 device
+only supports PPI interrupts.  These can either be injected from emulated
 in-kernel devices (such as the Arch Timer, or PMU), or via the KVM_IRQ_LINE
 ioctl.
 
@@ -25,7 +25,7 @@ Groups:
       request the initialization of the VGIC, no additional parameter in
       kvm_device_attr.addr. Must be called after all VCPUs have been created.
 
-   KVM_DEV_ARM_VGIC_USERPSPACE_PPIs
+   KVM_DEV_ARM_VGIC_USERSPACE_PPIS
       request the mask of userspace-drivable PPIs. Only a subset of the PPIs can
       be directly driven from userspace with GICv5, and the returned mask
       informs userspace of which it is allowed to drive via KVM_IRQ_LINE.
-- 
2.47.3

[PATCH v2 16/18] Documentation: KVM: Clarify that PMU_V3_IRQ IntID requirements for GICv5

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

When running a GICv5-based guest, the PMU must use PPI 23. This,
however, must be communicated via the
KVM_ARM_VCPU_PMU_V3_CTRL->KVM_ARM_VCPU_PMU_V3_IRQ ioctl as a full
GICv5-style Interrupt ID. That is, 0x20000017. Optionally, the whole
ioctl can be skipped for GICv5.

This was previously not clearly documented, so bump the documentation
accordingly.

Fixes: 7c31c06e2d2d ("KVM: arm64: gic-v5: Mandate architected PPI for PMU emulation on GICv5")
Link: https://sashiko.dev/#/patchset/20260319154937.3619520-1-sascha.bischoff%40arm.com
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 Documentation/virt/kvm/devices/vcpu.rst | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)

diff --git a/Documentation/virt/kvm/devices/vcpu.rst b/Documentation/virt/kvm/devices/vcpu.rst
index 5e38058200105..66e714f2fcfa7 100644
--- a/Documentation/virt/kvm/devices/vcpu.rst
+++ b/Documentation/virt/kvm/devices/vcpu.rst
@@ -37,8 +37,11 @@ Returns:
 A value describing the PMUv3 (Performance Monitor Unit v3) overflow interrupt
 number for this vcpu. This interrupt could be a PPI or SPI, but 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. For
-GICv5-based guests, the architected PPI (23) must be used.
+all vcpus, while as an SPI it must be a separate number per vcpu.
+
+For GICv5-based guests, the architected PPI (23) must be used, and must be
+communicated as the full GICv5-style Interrupt ID, i.e., 0x20000017. This ioctl
+can be omitted altogether for a GICv5-based guest.
 
 1.2 ATTRIBUTE: KVM_ARM_VCPU_PMU_V3_INIT
 ---------------------------------------
-- 
2.47.3

[PATCH v2 17/18] irqchip/gic-v5: Immediately exec priority drop following activate

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

With GICv5 an interrupt of equal or lower priority cannot be signalled
until there has been a priority drop. This is done via the GIC CDEOI
system instruction. Once this has been executed, the hardware is able
to signal the next interrupt if there is one.

As all interrupts are programmed to have the same priority, no new
interrupts can be signalled until the priority drop has happened. This
can cause issues when, for example, an interrupt remains active while
a long running process takes place, such as when injecting a physical
interrupt into a guest VM in software.

The GICv5 driver has so far done the priority drop as part of
irq_eoi(), i.e., at the same time as deactivating the interrupt. This
means that any long running process (or VM) could block incoming
interrupts, effectively causing a denial of service for all other
interrupts.

Rather than doing the EOI as part of irq_eoi() (which the name would
suggest would be a good place for it), move it to happen immediately
after acknowledging an interrupt in the main GICv5 interrupt
handler. The deactivation of interrupts (GIC CDDI) remains implemented
as part of irq_eoi(), which means that the same interrupt cannot be
signalled a second time until deactivated by software.

Suggested-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 drivers/irqchip/irq-gic-v5.c | 13 ++++++++-----
 1 file changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/irqchip/irq-gic-v5.c b/drivers/irqchip/irq-gic-v5.c
index 6b0903be8ebfd..58e457d4c1476 100644
--- a/drivers/irqchip/irq-gic-v5.c
+++ b/drivers/irqchip/irq-gic-v5.c
@@ -218,17 +218,13 @@ static void gicv5_hwirq_eoi(u32 hwirq_id, u8 hwirq_type)
 	       FIELD_PREP(GICV5_GIC_CDDI_TYPE_MASK, hwirq_type);
 
 	gic_insn(cddi, CDDI);
-
-	gic_insn(0, CDEOI);
 }
 
 static void gicv5_ppi_irq_eoi(struct irq_data *d)
 {
 	/* Skip deactivate for forwarded PPI interrupts */
-	if (irqd_is_forwarded_to_vcpu(d)) {
-		gic_insn(0, CDEOI);
+	if (irqd_is_forwarded_to_vcpu(d))
 		return;
-	}
 
 	gicv5_hwirq_eoi(d->hwirq, GICV5_HWIRQ_TYPE_PPI);
 }
@@ -963,6 +959,13 @@ static void __exception_irq_entry gicv5_handle_irq(struct pt_regs *regs)
 	 */
 	isb();
 
+	/*
+	 * Ensure that we can receive the next interrupts in the event that we
+	 * have a long running handler or directly enter a guest by doing the
+	 * priority drop immediately.
+	 */
+	gic_insn(0, CDEOI);
+
 	hwirq = FIELD_GET(GICV5_HWIRQ_INTID, ia);
 
 	handle_irq_per_domain(hwirq);
-- 
2.47.3

[PATCH v2 18/18] KVM: arm64: Fix arch timer interrupts for GICv3-on-GICv5 guests

[Marc Zyngier]

From: Sascha Bischoff <sascha.bischoff@arm.com>

When running on a GICv5 host, we push an arch-timer-specific interrupt
domain for the timer interrupts. This interrupt domain is used to mask
the host interrupt when a GICv5 guest is running. However, this
interrupt domain is still in place when running with a GICv3 guest on
GICv5 hardware. The result is that some interrupt state changes are
not correctly propragated to the host irqchip driver for legacy
guests.

Explicitly pass irqchip state changes though to the host irqchip
driver when running a GICv3-based guest on a GICv5 host. This bypasses
all masking, and thereby operates just as a native GICv3 guest would,
with the exception of having an additional irq domain in the
hierarchy.

Fixes: 9491c63b6cd7 ("KVM: arm64: gic-v5: Enlighten arch timer for GICv5")
Suggested-by: Marc Zyngier <maz@kernel.org>
Signed-off-by: Sascha Bischoff <sascha.bischoff@arm.com>
Signed-off-by: Marc Zyngier <maz@kernel.org>
---
 arch/arm64/kvm/arch_timer.c | 17 +++++++----------
 1 file changed, 7 insertions(+), 10 deletions(-)

diff --git a/arch/arm64/kvm/arch_timer.c b/arch/arm64/kvm/arch_timer.c
index f003df76fdda7..53b67b4d0bf24 100644
--- a/arch/arm64/kvm/arch_timer.c
+++ b/arch/arm64/kvm/arch_timer.c
@@ -1294,7 +1294,12 @@ static int timer_irq_set_vcpu_affinity(struct irq_data *d, void *vcpu)
 static int timer_irq_set_irqchip_state(struct irq_data *d,
 				       enum irqchip_irq_state which, bool val)
 {
-	if (which != IRQCHIP_STATE_ACTIVE || !irqd_is_forwarded_to_vcpu(d))
+	bool passthrough = which != IRQCHIP_STATE_ACTIVE ||
+		!irqd_is_forwarded_to_vcpu(d) ||
+		(kvm_vgic_global_state.type == VGIC_V5 &&
+		 vgic_is_v3(kvm_get_running_vcpu()->kvm));
+
+	if (passthrough)
 		return irq_chip_set_parent_state(d, which, val);
 
 	if (val)
@@ -1307,15 +1312,7 @@ static int timer_irq_set_irqchip_state(struct irq_data *d,
 
 static void timer_irq_eoi(struct irq_data *d)
 {
-	/*
-	 * On a GICv5 host, we still need to call EOI on the parent for
-	 * PPIs. The host driver already handles irqs which are forwarded to
-	 * vcpus, and skips the GIC CDDI while still doing the GIC CDEOI. This
-	 * is required to emulate the EOIMode=1 on GICv5 hardware. Failure to
-	 * call EOI unsurprisingly results in *BAD* lock-ups.
-	 */
-	if (!irqd_is_forwarded_to_vcpu(d) ||
-	    kvm_vgic_global_state.type == VGIC_V5)
+	if (!irqd_is_forwarded_to_vcpu(d))
 		irq_chip_eoi_parent(d);
 }
 
-- 
2.47.3