[PATCH v2 0/6] Handle forwarded level-triggered interrupts

This series illustrates an alternative approach to Eric Auger's direct EOI
setup patches [1] in terms of the KVM VGIC support.

The idea is to maintain existing semantics for the VGIC for mapped
level-triggered IRQs and think support for the timer into it.

Patch 1 is necessary to align the timer and VFIO ways of signaling the
VGIC.  Patch 2 is stolen from Eric's series and is necessary for these
patches to compile as well.  Patch 3 includes the core support for
mapped level-triggered interrupts.  Patch 4 handles guest MMIO acess to
the virtual distributor.  Patch 5 moves some code around for patch 6.
Patch 6 implements an optimization for the timer.  The last two patches
could be deferred until the timer optimization series.

Based on v4.13-rc7.

Thanks,
-Christoffer

Changes since v1:
 - Added necessary changes to the timer (Patch 1)
 - Added handling of guest MMIO accesses to the virtual distributor
   (Patch 4)
 - Addressed Marc's comments from the initial RFC (mostly renames)

Christoffer Dall (5):
  KVM: arm/arm64: Don't cache the timer IRQ level
  KVM: arm/arm64: vgic: Support level-triggered mapped interrupts
  KVM: arm/arm64: Support VGIC dist pend/active changes for mapped IRQs
  KVM: arm/arm64: Rearrange kvm_vgic_[un]map_phys code in vgic.c
  KVM: arm/arm64: Provide a vgic interrupt line level sample function

Eric Auger (1):
  KVM: arm/arm64: vgic: restructure kvm_vgic_(un)map_phys_irq

 include/kvm/arm_vgic.h        | 19 +++++++++--
 virt/kvm/arm/arch_timer.c     | 52 +++++++++++++-----------------
 virt/kvm/arm/vgic/vgic-mmio.c | 27 ++++++++++++++++
 virt/kvm/arm/vgic/vgic-v2.c   | 29 +++++++++++++++++
 virt/kvm/arm/vgic/vgic-v3.c   | 29 +++++++++++++++++
 virt/kvm/arm/vgic/vgic.c      | 75 ++++++++++++++++++++++++++++++++++++-------
 virt/kvm/arm/vgic/vgic.h      |  8 +++++
 7 files changed, 196 insertions(+), 43 deletions(-)

-- 
2.9.0

The timer was modeled after a strict idea of modelling an interrupt line
level in software, meaning that only transitions in the level needed to
be reported to the VGIC.  This works well for the timer, because the
arch timer code is in complete control of the device and can track the
transitions of the line.

However, as we are about to support using the HW bit in the VGIC not
just for the timer, but also for VFIO which cannot track transitions of
the interrupt line, we have to decide on an interface for level
triggered mapped interrupts to the GIC, which both the timer and VFIO
can use.

VFIO only sees an asserting transition of the physical interrupt line,
and tells the VGIC when that happens.  That means that part of the
interrupt flow is offloaded to the hardware.

To use the same interface for VFIO devices and the timer, we therefore
have to change the timer (we cannot change VFIO because it doesn't know
the details of the device it is assigning to a VM).

Luckily, changing the timer is simple, we just need to stop 'caching'
the line level, but instead let the VGIC know the state of the timer on
every entry to the guest, and the VGIC can ignore notifications using
its validate mechanism.

Signed-off-by: Christoffer Dall <redacted>
---
 virt/kvm/arm/arch_timer.c | 14 ++++++++------
 1 file changed, 8 insertions(+), 6 deletions(-)

diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 8e89d63..2a5f877 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c

@@ -219,9 +219,10 @@ static void kvm_timer_update_irq(struct kvm_vcpu *vcpu, bool new_level,
 	int ret;
 
 	timer_ctx->active_cleared_last = false;
+	if (timer_ctx->irq.level != new_level)
+		trace_kvm_timer_update_irq(vcpu->vcpu_id, timer_ctx->irq.irq,
+					   new_level);
 	timer_ctx->irq.level = new_level;
-	trace_kvm_timer_update_irq(vcpu->vcpu_id, timer_ctx->irq.irq,
-				   timer_ctx->irq.level);
 
 	if (likely(irqchip_in_kernel(vcpu->kvm))) {
 		ret = kvm_vgic_inject_irq(vcpu->kvm, vcpu->vcpu_id,

@@ -241,6 +242,7 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
 	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
 	struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
 	struct arch_timer_context *ptimer = vcpu_ptimer(vcpu);
+	bool level;
 
 	/*
 	 * If userspace modified the timer registers via SET_ONE_REG before

@@ -251,11 +253,11 @@ static void kvm_timer_update_state(struct kvm_vcpu *vcpu)
 	if (unlikely(!timer->enabled))
 		return;
 
-	if (kvm_timer_should_fire(vtimer) != vtimer->irq.level)
-		kvm_timer_update_irq(vcpu, !vtimer->irq.level, vtimer);
+	level = kvm_timer_should_fire(vtimer);
+	kvm_timer_update_irq(vcpu, level, vtimer);
 
-	if (kvm_timer_should_fire(ptimer) != ptimer->irq.level)
-		kvm_timer_update_irq(vcpu, !ptimer->irq.level, ptimer);
+	level = kvm_timer_should_fire(ptimer);
+	kvm_timer_update_irq(vcpu, level, ptimer);
 }
 
 /* Schedule the background timer for the emulated timer. */

-- 
2.9.0

From: Eric Auger <eric.auger@redhat.com>

We want to reuse the core of the map/unmap functions for IRQ forwarding.
Let's move the computation of the hwirq in kvm_vgic_map_phys_irq and
pass the linux IRQ as parameter.

The host_irq is added to struct vgic_irq because it is needed in later
patches which manipulate the physical GIC state to support forwarded
IRQs.

We introduce kvm_vgic_map/unmap_irq which take a struct vgic_irq handle
as a parameter.

Signed-off-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Christoffer Dall <redacted>
Acked-by: Marc Zyngier <redacted>
---
 include/kvm/arm_vgic.h    |  8 ++++---
 virt/kvm/arm/arch_timer.c | 24 +------------------
 virt/kvm/arm/vgic/vgic.c  | 60 +++++++++++++++++++++++++++++++++++------------
 3 files changed, 51 insertions(+), 41 deletions(-)

diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
index 34dba51..53f631b 100644
--- a/include/kvm/arm_vgic.h
+++ b/include/kvm/arm_vgic.h

@@ -116,6 +116,7 @@ struct vgic_irq {
 	bool hw;			/* Tied to HW IRQ */
 	struct kref refcount;		/* Used for LPIs */
 	u32 hwintid;			/* HW INTID number */
+	unsigned int host_irq;		/* linux irq corresponding to hwintid */
 	union {
 		u8 targets;			/* GICv2 target VCPUs mask */
 		u32 mpidr;			/* GICv3 target VCPU */

@@ -307,9 +308,10 @@ void kvm_vgic_init_cpu_hardware(void);
 
 int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid,
 			bool level, void *owner);
-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_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
+			  u32 vintid);
+int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int vintid);
+bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int vintid);
 
 int kvm_vgic_vcpu_pending_irq(struct kvm_vcpu *vcpu);
 

diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index 2a5f877..c4fa675 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c

@@ -649,9 +649,6 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
 {
 	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;
 	struct arch_timer_context *vtimer = vcpu_vtimer(vcpu);
-	struct irq_desc *desc;
-	struct irq_data *data;
-	int phys_irq;
 	int ret;
 
 	if (timer->enabled)

@@ -669,26 +666,7 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
 		return -EINVAL;
 	}
 
-	/*
-	 * 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, vtimer->irq.irq, phys_irq);
+	ret = kvm_vgic_map_phys_irq(vcpu, host_vtimer_irq, vtimer->irq.irq);
 	if (ret)
 		return ret;
 

diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c
index fed717e..9d557efd 100644
--- a/virt/kvm/arm/vgic/vgic.c
+++ b/virt/kvm/arm/vgic/vgic.c

@@ -17,6 +17,8 @@
 #include <linux/kvm.h>
 #include <linux/kvm_host.h>
 #include <linux/list_sort.h>
+#include <linux/interrupt.h>
+#include <linux/irq.h>
 
 #include "vgic.h"
 

@@ -403,38 +405,66 @@ int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid,
 	return 0;
 }
 
-int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, u32 virt_irq, u32 phys_irq)
+/* @irq->irq_lock must be held */
+static int kvm_vgic_map_irq(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
+			    unsigned int host_irq)
 {
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, virt_irq);
+	struct irq_desc *desc;
+	struct irq_data *data;
 
-	BUG_ON(!irq);
-
-	spin_lock(&irq->irq_lock);
+	/*
+	 * Find the physical IRQ number corresponding to @host_irq
+	 */
+	desc = irq_to_desc(host_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;
 
 	irq->hw = true;
-	irq->hwintid = phys_irq;
+	irq->host_irq = host_irq;
+	irq->hwintid = data->hwirq;
+	return 0;
+}
+
+/* @irq->irq_lock must be held */
+static inline void kvm_vgic_unmap_irq(struct vgic_irq *irq)
+{
+	irq->hw = false;
+	irq->hwintid = 0;
+}
+
+int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
+			  u32 vintid)
+{
+	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
+	int ret;
 
+	BUG_ON(!irq);
+
+	spin_lock(&irq->irq_lock);
+	ret = kvm_vgic_map_irq(vcpu, irq, host_irq);
 	spin_unlock(&irq->irq_lock);
 	vgic_put_irq(vcpu->kvm, irq);
 
-	return 0;
+	return ret;
 }
 
-int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int virt_irq)
+int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int vintid)
 {
 	struct vgic_irq *irq;
 
 	if (!vgic_initialized(vcpu->kvm))
 		return -EAGAIN;
 
-	irq = vgic_get_irq(vcpu->kvm, vcpu, virt_irq);
+	irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
 	BUG_ON(!irq);
 
 	spin_lock(&irq->irq_lock);
-
-	irq->hw = false;
-	irq->hwintid = 0;
-
+	kvm_vgic_unmap_irq(irq);
 	spin_unlock(&irq->irq_lock);
 	vgic_put_irq(vcpu->kvm, irq);
 

@@ -772,9 +802,9 @@ void vgic_kick_vcpus(struct kvm *kvm)
 	}
 }
 
-bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int virt_irq)
+bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int vintid)
 {
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, virt_irq);
+	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
 	bool map_is_active;
 
 	spin_lock(&irq->irq_lock);

-- 
2.9.0

Level-triggered mapped IRQs are special because we only observe rising
edges as input to the VGIC, and we don't set the EOI flag and therefore
are not told when the level goes down, so that we can re-queue a new
interrupt when the level goes up.

One way to solve this problem is to side-step the logic of the VGIC and
special case the validation in the injection path, but it has the
unfortunate drawback of having to peak into the physical GIC state
whenever we want to know if the interrupt is pending on the virtual
distributor.

Instead, we can maintain the current semantics of a level triggered
interrupt by sort of treating it as an edge-triggered interrupt,
following from the fact that we only observe an asserting edge.  This
requires us to be a bit careful when populating the LRs and when folding
the state back in though:

 * We lower the line level when populating the LR, so that when
   subsequently observing an asserting edge, the VGIC will do the right
   thing.

 * If the guest never acked the interrupt while running (for example if
   it had masked interrupts at the CPU level while running), we have
   to preserve the pending state of the LR and move it back to the
   line_level field of the struct irq when folding LR state.

   If the guest never acked the interrupt while running, but changed the
   device state and lowered the line (again with interrupts masked) then
   we need to observe this change in the line_level.

   Both of the above situations are solved by sampling the physical line
   and set the line level when folding the LR back.

 * Finally, if the guest never acked the interrupt while running and
   sampling the line reveals that the device state has changed and the
   line has been lowered, we must clear the physical active state, since
   we will otherwise never be told when the interrupt becomes asserted
   again.

This has the added benefit of making the timer optimization patches
(https://lists.cs.columbia.edu/pipermail/kvmarm/2017-July/026343.html) a
bit simpler, because the timer code doesn't have to clear the active
state on the sync anymore.  It also potentially improves the performance
of the timer implementation because the GIC knows the state or the LR
and only needs to clear the
active state when the pending bit in the LR is still set, where the
timer has to always clear it when returning from running the guest with
an injected timer interrupt.

Signed-off-by: Christoffer Dall <redacted>
Reviewed-by: Marc Zyngier <redacted>
---
 virt/kvm/arm/vgic/vgic-v2.c | 29 +++++++++++++++++++++++++++++
 virt/kvm/arm/vgic/vgic-v3.c | 29 +++++++++++++++++++++++++++++
 virt/kvm/arm/vgic/vgic.c    | 23 +++++++++++++++++++++++
 virt/kvm/arm/vgic/vgic.h    |  7 +++++++
 4 files changed, 88 insertions(+)

diff --git a/virt/kvm/arm/vgic/vgic-v2.c b/virt/kvm/arm/vgic/vgic-v2.c
index e4187e5..618ed3f 100644
--- a/virt/kvm/arm/vgic/vgic-v2.c
+++ b/virt/kvm/arm/vgic/vgic-v2.c

@@ -104,6 +104,26 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu)
 				irq->pending_latch = false;
 		}
 
+		/*
+		 * Level-triggered mapped IRQs are special because we only
+		 * observe rising edges as input to the VGIC.
+		 *
+		 * If the guest never acked the interrupt we have to sample
+		 * the physical line and set the line level, because the
+		 * device state could have changed or we simply need to
+		 * process the still pending interrupt later.
+		 *
+		 * If this causes us to lower the level, we have to also clear
+		 * the physical active state, since we will otherwise never be
+		 * told when the interrupt becomes asserted again.
+		 */
+		if (vgic_irq_is_mapped_level(irq) && (val & GICH_LR_PENDING_BIT)) {
+			irq->line_level = vgic_get_phys_line_level(irq);
+
+			if (!irq->line_level)
+				vgic_irq_set_phys_active(irq, false);
+		}
+
 		spin_unlock(&irq->irq_lock);
 		vgic_put_irq(vcpu->kvm, irq);
 	}

@@ -161,6 +181,15 @@ void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr)
 			val |= GICH_LR_EOI;
 	}
 
+	/*
+	 * Level-triggered mapped IRQs are special because we only observe
+	 * rising edges as input to the VGIC.  We therefore lower the line
+	 * level here, so that we can take new virtual IRQs.  See
+	 * vgic_v2_fold_lr_state for more info.
+	 */
+	if (vgic_irq_is_mapped_level(irq) && (val & GICH_LR_PENDING_BIT))
+		irq->line_level = false;
+
 	/* The GICv2 LR only holds five bits of priority. */
 	val |= (irq->priority >> 3) << GICH_LR_PRIORITY_SHIFT;
 

diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c
index 96ea597..0f72bcb 100644
--- a/virt/kvm/arm/vgic/vgic-v3.c
+++ b/virt/kvm/arm/vgic/vgic-v3.c

@@ -94,6 +94,26 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu)
 				irq->pending_latch = false;
 		}
 
+		/*
+		 * Level-triggered mapped IRQs are special because we only
+		 * observe rising edges as input to the VGIC.
+		 *
+		 * If the guest never acked the interrupt we have to sample
+		 * the physical line and set the line level, because the
+		 * device state could have changed or we simply need to
+		 * process the still pending interrupt later.
+		 *
+		 * If this causes us to lower the level, we have to also clear
+		 * the physical active state, since we will otherwise never be
+		 * told when the interrupt becomes asserted again.
+		 */
+		if (vgic_irq_is_mapped_level(irq) && (val & ICH_LR_PENDING_BIT)) {
+			irq->line_level = vgic_get_phys_line_level(irq);
+
+			if (!irq->line_level)
+				vgic_irq_set_phys_active(irq, false);
+		}
+
 		spin_unlock(&irq->irq_lock);
 		vgic_put_irq(vcpu->kvm, irq);
 	}

@@ -144,6 +164,15 @@ void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr)
 	}
 
 	/*
+	 * Level-triggered mapped IRQs are special because we only observe
+	 * rising edges as input to the VGIC.  We therefore lower the line
+	 * level here, so that we can take new virtual IRQs.  See
+	 * vgic_v3_fold_lr_state for more info.
+	 */
+	if (vgic_irq_is_mapped_level(irq) && (val & ICH_LR_PENDING_BIT))
+		irq->line_level = false;
+
+	/*
 	 * We currently only support Group1 interrupts, which is a
 	 * known defect. This needs to be addressed at some point.
 	 */

diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c
index 9d557efd..8072969 100644
--- a/virt/kvm/arm/vgic/vgic.c
+++ b/virt/kvm/arm/vgic/vgic.c

@@ -140,6 +140,29 @@ void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq)
 	kfree(irq);
 }
 
+/* Get the input level of a mapped IRQ directly from the physical GIC */
+bool vgic_get_phys_line_level(struct vgic_irq *irq)
+{
+	bool line_level;
+
+	BUG_ON(!irq->hw);
+
+	WARN_ON(irq_get_irqchip_state(irq->host_irq,
+				      IRQCHIP_STATE_PENDING,
+				      &line_level));
+	return line_level;
+}
+
+/* Set/Clear the physical active state */
+void vgic_irq_set_phys_active(struct vgic_irq *irq, bool active)
+{
+
+	BUG_ON(!irq->hw);
+	WARN_ON(irq_set_irqchip_state(irq->host_irq,
+				      IRQCHIP_STATE_ACTIVE,
+				      active));
+}
+
 /**
  * kvm_vgic_target_oracle - compute the target vcpu for an irq
  *

diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h
index bba7fa2..7bdcda2 100644
--- a/virt/kvm/arm/vgic/vgic.h
+++ b/virt/kvm/arm/vgic/vgic.h

@@ -104,6 +104,11 @@ static inline bool irq_is_pending(struct vgic_irq *irq)
 		return irq->pending_latch || irq->line_level;
 }
 
+static inline bool vgic_irq_is_mapped_level(struct vgic_irq *irq)
+{
+	return irq->config == VGIC_CONFIG_LEVEL && irq->hw;
+}
+
 /*
  * This struct provides an intermediate representation of the fields contained
  * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC

@@ -140,6 +145,8 @@ vgic_get_mmio_region(struct kvm_vcpu *vcpu, struct vgic_io_device *iodev,
 struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 			      u32 intid);
 void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq);
+bool vgic_get_phys_line_level(struct vgic_irq *irq);
+void vgic_irq_set_phys_active(struct vgic_irq *irq, bool active);
 bool vgic_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq);
 void vgic_kick_vcpus(struct kvm *kvm);
 

-- 
2.9.0

For mapped IRQs (with the HW bit set in the LR) we have to follow some
rules of the architecture.  One of these rules is that VM must not be
allowed to deactivate a virtual interrupt with the HW bit set unless the
physical interrupt is also active.

This works fine when injecting mapped interrupts, because we leave it up
to the injector to either set EOImode==1 or manually set the active
state of the physical interrupt.

However, the guest can set virtual interrupt to be pending or active by
writing to the virtual distributor, which could lead to deactivating a
virtual interrupt with the HW bit set without the physical interrupt
being active.

We could set the physical interrupt to active whenever we are about to
eneter the VM with a HW interrupt either pending or active, but that
would be really slow, especially on GICv2.  So we take the long way
around and do the hard work when needed, which is expected to be
extremely rare.

When the VM sets the pending state for a HW interrupt on the virtual
distributor we set the active state on the physical distributor, because
the virtual interrupt can become active and then the guest can
deactivate it.

When the VM clears the pending state we also clear it on the physical
side, because the injector might otherwise raise the interrupt.

Signed-off-by: Christoffer Dall <redacted>
---
 virt/kvm/arm/vgic/vgic-mmio.c | 27 +++++++++++++++++++++++++++
 virt/kvm/arm/vgic/vgic.c      |  7 +++++++
 virt/kvm/arm/vgic/vgic.h      |  1 +
 3 files changed, 35 insertions(+)

diff --git a/virt/kvm/arm/vgic/vgic-mmio.c b/virt/kvm/arm/vgic/vgic-mmio.c
index c1e4bdd..c825d7c 100644
--- a/virt/kvm/arm/vgic/vgic-mmio.c
+++ b/virt/kvm/arm/vgic/vgic-mmio.c

@@ -131,6 +131,12 @@ void vgic_mmio_write_spending(struct kvm_vcpu *vcpu,
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 
 		spin_lock(&irq->irq_lock);
+		if (irq->hw) {
+			vgic_irq_set_phys_active(irq, true);
+			spin_unlock(&irq->irq_lock);
+			continue;
+		}
+
 		irq->pending_latch = true;
 
 		vgic_queue_irq_unlock(vcpu->kvm, irq);

@@ -149,6 +155,11 @@ void vgic_mmio_write_cpending(struct kvm_vcpu *vcpu,
 		struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, intid + i);
 
 		spin_lock(&irq->irq_lock);
+		if (irq->hw) {
+			vgic_irq_set_phys_pending(irq, false);
+			spin_unlock(&irq->irq_lock);
+			continue;
+		}
 
 		irq->pending_latch = false;
 

@@ -214,6 +225,22 @@ static void vgic_mmio_change_active(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
 	       irq->vcpu->cpu != -1) /* VCPU thread is running */
 		cond_resched_lock(&irq->irq_lock);
 
+	if (irq->hw) {
+		/*
+		 * We cannot support setting the physical active state for
+		 * private interrupts from another CPU than the one running
+		 * the VCPU which identifies which private interrupt it is
+		 * trying to modify.
+		 */
+		if (irq->intid < VGIC_NR_PRIVATE_IRQS &&
+		    irq->target_vcpu != requester_vcpu) {
+			spin_unlock(&irq->irq_lock);
+			return;
+		}
+
+		vgic_irq_set_phys_active(irq, new_active_state);
+	}
+
 	irq->active = new_active_state;
 	if (new_active_state)
 		vgic_queue_irq_unlock(vcpu->kvm, irq);

diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c
index 8072969..7aec730 100644
--- a/virt/kvm/arm/vgic/vgic.c
+++ b/virt/kvm/arm/vgic/vgic.c

@@ -140,6 +140,13 @@ void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq)
 	kfree(irq);
 }
 
+void vgic_irq_set_phys_pending(struct vgic_irq *irq, bool pending)
+{
+	WARN_ON(irq_set_irqchip_state(irq->host_irq,
+				      IRQCHIP_STATE_PENDING,
+				      pending));
+}
+
 /* Get the input level of a mapped IRQ directly from the physical GIC */
 bool vgic_get_phys_line_level(struct vgic_irq *irq)
 {

diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h
index 7bdcda2..498ee05 100644
--- a/virt/kvm/arm/vgic/vgic.h
+++ b/virt/kvm/arm/vgic/vgic.h

@@ -146,6 +146,7 @@ struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 			      u32 intid);
 void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq);
 bool vgic_get_phys_line_level(struct vgic_irq *irq);
+void vgic_irq_set_phys_pending(struct vgic_irq *irq, bool pending);
 void vgic_irq_set_phys_active(struct vgic_irq *irq, bool active);
 bool vgic_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq);
 void vgic_kick_vcpus(struct kvm *kvm);

-- 
2.9.0

The small indirection of a static function made the locking very obvious
but becomes pretty ugly as we start passing function pointer around.
Let's inline these two functions first to make the following patch more
readable.

Signed-off-by: Christoffer Dall <redacted>
Acked-by: Marc Zyngier <redacted>
---
 virt/kvm/arm/vgic/vgic.c | 38 +++++++++++++-------------------------
 1 file changed, 13 insertions(+), 25 deletions(-)

diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c
index 7aec730..b704ff5 100644
--- a/virt/kvm/arm/vgic/vgic.c
+++ b/virt/kvm/arm/vgic/vgic.c

@@ -435,12 +435,17 @@ int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid,
 	return 0;
 }
 
-/* @irq->irq_lock must be held */
-static int kvm_vgic_map_irq(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
-			    unsigned int host_irq)
+int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
+			  u32 vintid)
 {
+	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
 	struct irq_desc *desc;
 	struct irq_data *data;
+	int ret = 0;
+
+	BUG_ON(!irq);
+
+	spin_lock(&irq->irq_lock);
 
 	/*
 	 * Find the physical IRQ number corresponding to @host_irq

@@ -448,7 +453,8 @@ static int kvm_vgic_map_irq(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
 	desc = irq_to_desc(host_irq);
 	if (!desc) {
 		kvm_err("%s: no interrupt descriptor\n", __func__);
-		return -EINVAL;
+		ret = -EINVAL;
+		goto out;
 	}
 	data = irq_desc_get_irq_data(desc);
 	while (data->parent_data)

@@ -457,29 +463,10 @@ static int kvm_vgic_map_irq(struct kvm_vcpu *vcpu, struct vgic_irq *irq,
 	irq->hw = true;
 	irq->host_irq = host_irq;
 	irq->hwintid = data->hwirq;
-	return 0;
-}
-
-/* @irq->irq_lock must be held */
-static inline void kvm_vgic_unmap_irq(struct vgic_irq *irq)
-{
-	irq->hw = false;
-	irq->hwintid = 0;
-}
-
-int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
-			  u32 vintid)
-{
-	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
-	int ret;
 
-	BUG_ON(!irq);
-
-	spin_lock(&irq->irq_lock);
-	ret = kvm_vgic_map_irq(vcpu, irq, host_irq);
+out:
 	spin_unlock(&irq->irq_lock);
 	vgic_put_irq(vcpu->kvm, irq);
-
 	return ret;
 }
 

@@ -494,7 +481,8 @@ int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int vintid)
 	BUG_ON(!irq);
 
 	spin_lock(&irq->irq_lock);
-	kvm_vgic_unmap_irq(irq);
+	irq->hw = false;
+	irq->hwintid = 0;
 	spin_unlock(&irq->irq_lock);
 	vgic_put_irq(vcpu->kvm, irq);
 

-- 
2.9.0

The GIC sometimes need to sample the physical line of a mapped
interrupt.  As we know this to be notoriously slow, provide a callback
function for devices (such as the timer) which can do this much faster
than talking to the distributor, for example by comparing a few
in-memory values.  Fall back to the good old method of poking the
physical GIC if no callback is provided.

Signed-off-by: Christoffer Dall <redacted>
Reviewed-by: Marc Zyngier <redacted>
---
 include/kvm/arm_vgic.h    | 13 ++++++++++++-
 virt/kvm/arm/arch_timer.c | 16 +++++++++++++++-
 virt/kvm/arm/vgic/vgic.c  |  7 ++++++-
 3 files changed, 33 insertions(+), 3 deletions(-)

diff --git a/include/kvm/arm_vgic.h b/include/kvm/arm_vgic.h
index 53f631b..7f76c9e 100644
--- a/include/kvm/arm_vgic.h
+++ b/include/kvm/arm_vgic.h

@@ -125,6 +125,17 @@ struct vgic_irq {
 	u8 priority;
 	enum vgic_irq_config config;	/* Level or edge */
 
+	/*
+	 * Callback function pointer to in-kernel devices that can tell us the
+	 * state of the input level of mapped level-triggered IRQ faster than
+	 * peaking into the physical GIC.
+	 *
+	 * Always called in non-preemptible section and the functions can use
+	 * kvm_arm_get_running_vcpu() to get the vcpu pointer for private
+	 * IRQs.
+	 */
+	bool (*get_input_level)(int vintid);
+
 	void *owner;			/* Opaque pointer to reserve an interrupt
 					   for in-kernel devices. */
 };

@@ -309,7 +320,7 @@ void kvm_vgic_init_cpu_hardware(void);
 int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid,
 			bool level, void *owner);
 int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
-			  u32 vintid);
+			  u32 vintid, bool (*get_input_level)(int vindid));
 int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int vintid);
 bool kvm_vgic_map_is_active(struct kvm_vcpu *vcpu, unsigned int vintid);
 

diff --git a/virt/kvm/arm/arch_timer.c b/virt/kvm/arm/arch_timer.c
index c4fa675..5cb96ca 100644
--- a/virt/kvm/arm/arch_timer.c
+++ b/virt/kvm/arm/arch_timer.c

@@ -645,6 +645,19 @@ static bool timer_irqs_are_valid(struct kvm_vcpu *vcpu)
 	return true;
 }
 
+static bool timer_get_input_level(int vintid)
+{
+	struct kvm_vcpu *vcpu = kvm_arm_get_running_vcpu();
+	struct arch_timer_context *timer;
+
+	if (vintid == vcpu_vtimer(vcpu)->irq.irq)
+		timer = vcpu_vtimer(vcpu);
+	else
+		BUG(); /* We only map the vtimer so far */
+
+	return kvm_timer_should_fire(timer);
+}
+
 int kvm_timer_enable(struct kvm_vcpu *vcpu)
 {
 	struct arch_timer_cpu *timer = &vcpu->arch.timer_cpu;

@@ -666,7 +679,8 @@ int kvm_timer_enable(struct kvm_vcpu *vcpu)
 		return -EINVAL;
 	}
 
-	ret = kvm_vgic_map_phys_irq(vcpu, host_vtimer_irq, vtimer->irq.irq);
+	ret = kvm_vgic_map_phys_irq(vcpu, host_vtimer_irq, vtimer->irq.irq,
+				    timer_get_input_level);
 	if (ret)
 		return ret;
 

diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c
index b704ff5..202938c 100644
--- a/virt/kvm/arm/vgic/vgic.c
+++ b/virt/kvm/arm/vgic/vgic.c

@@ -154,6 +154,9 @@ bool vgic_get_phys_line_level(struct vgic_irq *irq)
 
 	BUG_ON(!irq->hw);
 
+	if (irq->get_input_level)
+		return irq->get_input_level(irq->intid);
+
 	WARN_ON(irq_get_irqchip_state(irq->host_irq,
 				      IRQCHIP_STATE_PENDING,
 				      &line_level));

@@ -436,7 +439,7 @@ int kvm_vgic_inject_irq(struct kvm *kvm, int cpuid, unsigned int intid,
 }
 
 int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
-			  u32 vintid)
+			  u32 vintid, bool (*get_input_level)(int vindid))
 {
 	struct vgic_irq *irq = vgic_get_irq(vcpu->kvm, vcpu, vintid);
 	struct irq_desc *desc;

@@ -463,6 +466,7 @@ int kvm_vgic_map_phys_irq(struct kvm_vcpu *vcpu, unsigned int host_irq,
 	irq->hw = true;
 	irq->host_irq = host_irq;
 	irq->hwintid = data->hwirq;
+	irq->get_input_level = get_input_level;
 
 out:
 	spin_unlock(&irq->irq_lock);

@@ -483,6 +487,7 @@ int kvm_vgic_unmap_phys_irq(struct kvm_vcpu *vcpu, unsigned int vintid)
 	spin_lock(&irq->irq_lock);
 	irq->hw = false;
 	irq->hwintid = 0;
+	irq->get_input_level = NULL;
 	spin_unlock(&irq->irq_lock);
 	vgic_put_irq(vcpu->kvm, irq);
 

-- 
2.9.0

Hi Christoffer,

On 04/09/2017 12:24, Christoffer Dall wrote:

I am still unclear wrt ISPENDR case. Assume the guest writes into the
ISPENDR. So now we set the pending state on the phys distributor. The
SPI is acked by the host. I understand it becomes active (ie. not
pending and active). Then it gets injected into the guest and in the
above fold code the LR state is observed GICH_LR_PENDING_BIT. We are
going to read the physical pending state which is: not pending. So the
line level is low, the latch_pending is not used in mapped case and is
low. So to me the vIRQ is missed.

What do I miss?

Thanks

Eric

Hi Christoffer,
On 04/09/2017 12:24, Christoffer Dall wrote:

I chose to hold the lock outside of kvm_vgic_map/unmap_irq because in
kvm_vgic_set_forwarding(see https://lkml.org/lkml/2017/6/15/278) I was
also testing hw and target_vcpu fields. As you pointed out maybe I am
not obliged to check them but that was the rationale.

Thanks

Eric

On Tue, Sep 05, 2017 at 11:38:48AM +0200, Auger Eric wrote:

Hi Christoffer,

On 04/09/2017 12:24, Christoffer Dall wrote:

quoted

Level-triggered mapped IRQs are special because we only observe rising
edges as input to the VGIC, and we don't set the EOI flag and therefore
are not told when the level goes down, so that we can re-queue a new
interrupt when the level goes up.

One way to solve this problem is to side-step the logic of the VGIC and
special case the validation in the injection path, but it has the
unfortunate drawback of having to peak into the physical GIC state
whenever we want to know if the interrupt is pending on the virtual
distributor.

Instead, we can maintain the current semantics of a level triggered
interrupt by sort of treating it as an edge-triggered interrupt,
following from the fact that we only observe an asserting edge.  This
requires us to be a bit careful when populating the LRs and when folding
the state back in though:

 * We lower the line level when populating the LR, so that when
   subsequently observing an asserting edge, the VGIC will do the right
   thing.

 * If the guest never acked the interrupt while running (for example if
   it had masked interrupts at the CPU level while running), we have
   to preserve the pending state of the LR and move it back to the
   line_level field of the struct irq when folding LR state.

   If the guest never acked the interrupt while running, but changed the
   device state and lowered the line (again with interrupts masked) then
   we need to observe this change in the line_level.

   Both of the above situations are solved by sampling the physical line
   and set the line level when folding the LR back.

 * Finally, if the guest never acked the interrupt while running and
   sampling the line reveals that the device state has changed and the
   line has been lowered, we must clear the physical active state, since
   we will otherwise never be told when the interrupt becomes asserted
   again.

This has the added benefit of making the timer optimization patches
(https://lists.cs.columbia.edu/pipermail/kvmarm/2017-July/026343.html) a
bit simpler, because the timer code doesn't have to clear the active
state on the sync anymore.  It also potentially improves the performance
of the timer implementation because the GIC knows the state or the LR
and only needs to clear the
active state when the pending bit in the LR is still set, where the
timer has to always clear it when returning from running the guest with
an injected timer interrupt.

Signed-off-by: Christoffer Dall <redacted>
Reviewed-by: Marc Zyngier <redacted>
---
 virt/kvm/arm/vgic/vgic-v2.c | 29 +++++++++++++++++++++++++++++
 virt/kvm/arm/vgic/vgic-v3.c | 29 +++++++++++++++++++++++++++++
 virt/kvm/arm/vgic/vgic.c    | 23 +++++++++++++++++++++++
 virt/kvm/arm/vgic/vgic.h    |  7 +++++++
 4 files changed, 88 insertions(+)

diff --git a/virt/kvm/arm/vgic/vgic-v2.c b/virt/kvm/arm/vgic/vgic-v2.c
index e4187e5..618ed3f 100644
--- a/virt/kvm/arm/vgic/vgic-v2.c
+++ b/virt/kvm/arm/vgic/vgic-v2.c

@@ -104,6 +104,26 @@ void vgic_v2_fold_lr_state(struct kvm_vcpu *vcpu)
 				irq->pending_latch = false;
 		}
 
+		/*
+		 * Level-triggered mapped IRQs are special because we only
+		 * observe rising edges as input to the VGIC.
+		 *
+		 * If the guest never acked the interrupt we have to sample
+		 * the physical line and set the line level, because the
+		 * device state could have changed or we simply need to
+		 * process the still pending interrupt later.
+		 *
+		 * If this causes us to lower the level, we have to also clear
+		 * the physical active state, since we will otherwise never be
+		 * told when the interrupt becomes asserted again.
+		 */
+		if (vgic_irq_is_mapped_level(irq) && (val & GICH_LR_PENDING_BIT)) {
+			irq->line_level = vgic_get_phys_line_level(irq);
+
+			if (!irq->line_level)
+				vgic_irq_set_phys_active(irq, false);

I am still unclear wrt ISPENDR case. Assume the guest writes into the
ISPENDR. So now we set the pending state on the phys distributor. The
SPI is acked by the host. I understand it becomes active (ie. not
pending and active). Then it gets injected into the guest and in the
above fold code the LR state is observed GICH_LR_PENDING_BIT. We are
going to read the physical pending state which is: not pending. So the
line level is low, the latch_pending is not used in mapped case and is
low. So to me the vIRQ is missed.

What do I miss?

You're not missing anything.  It was supposed to set pending first then
followed by active.  I can fix that.

Thanks,
-Christoffer

quoted

+		}
+
 		spin_unlock(&irq->irq_lock);
 		vgic_put_irq(vcpu->kvm, irq);
 	}

@@ -161,6 +181,15 @@ void vgic_v2_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr)
 			val |= GICH_LR_EOI;
 	}
 
+	/*
+	 * Level-triggered mapped IRQs are special because we only observe
+	 * rising edges as input to the VGIC.  We therefore lower the line
+	 * level here, so that we can take new virtual IRQs.  See
+	 * vgic_v2_fold_lr_state for more info.
+	 */
+	if (vgic_irq_is_mapped_level(irq) && (val & GICH_LR_PENDING_BIT))
+		irq->line_level = false;
+
 	/* The GICv2 LR only holds five bits of priority. */
 	val |= (irq->priority >> 3) << GICH_LR_PRIORITY_SHIFT;
 

diff --git a/virt/kvm/arm/vgic/vgic-v3.c b/virt/kvm/arm/vgic/vgic-v3.c
index 96ea597..0f72bcb 100644
--- a/virt/kvm/arm/vgic/vgic-v3.c
+++ b/virt/kvm/arm/vgic/vgic-v3.c

@@ -94,6 +94,26 @@ void vgic_v3_fold_lr_state(struct kvm_vcpu *vcpu)
 				irq->pending_latch = false;
 		}
 
+		/*
+		 * Level-triggered mapped IRQs are special because we only
+		 * observe rising edges as input to the VGIC.
+		 *
+		 * If the guest never acked the interrupt we have to sample
+		 * the physical line and set the line level, because the
+		 * device state could have changed or we simply need to
+		 * process the still pending interrupt later.
+		 *
+		 * If this causes us to lower the level, we have to also clear
+		 * the physical active state, since we will otherwise never be
+		 * told when the interrupt becomes asserted again.
+		 */
+		if (vgic_irq_is_mapped_level(irq) && (val & ICH_LR_PENDING_BIT)) {
+			irq->line_level = vgic_get_phys_line_level(irq);
+
+			if (!irq->line_level)
+				vgic_irq_set_phys_active(irq, false);
+		}
+
 		spin_unlock(&irq->irq_lock);
 		vgic_put_irq(vcpu->kvm, irq);
 	}

@@ -144,6 +164,15 @@ void vgic_v3_populate_lr(struct kvm_vcpu *vcpu, struct vgic_irq *irq, int lr)
 	}
 
 	/*
+	 * Level-triggered mapped IRQs are special because we only observe
+	 * rising edges as input to the VGIC.  We therefore lower the line
+	 * level here, so that we can take new virtual IRQs.  See
+	 * vgic_v3_fold_lr_state for more info.
+	 */
+	if (vgic_irq_is_mapped_level(irq) && (val & ICH_LR_PENDING_BIT))
+		irq->line_level = false;
+
+	/*
 	 * We currently only support Group1 interrupts, which is a
 	 * known defect. This needs to be addressed at some point.
 	 */

diff --git a/virt/kvm/arm/vgic/vgic.c b/virt/kvm/arm/vgic/vgic.c
index 9d557efd..8072969 100644
--- a/virt/kvm/arm/vgic/vgic.c
+++ b/virt/kvm/arm/vgic/vgic.c

@@ -140,6 +140,29 @@ void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq)
 	kfree(irq);
 }
 
+/* Get the input level of a mapped IRQ directly from the physical GIC */
+bool vgic_get_phys_line_level(struct vgic_irq *irq)
+{
+	bool line_level;
+
+	BUG_ON(!irq->hw);
+
+	WARN_ON(irq_get_irqchip_state(irq->host_irq,
+				      IRQCHIP_STATE_PENDING,
+				      &line_level));
+	return line_level;
+}
+
+/* Set/Clear the physical active state */
+void vgic_irq_set_phys_active(struct vgic_irq *irq, bool active)
+{
+
+	BUG_ON(!irq->hw);
+	WARN_ON(irq_set_irqchip_state(irq->host_irq,
+				      IRQCHIP_STATE_ACTIVE,
+				      active));
+}
+
 /**
  * kvm_vgic_target_oracle - compute the target vcpu for an irq
  *

diff --git a/virt/kvm/arm/vgic/vgic.h b/virt/kvm/arm/vgic/vgic.h
index bba7fa2..7bdcda2 100644
--- a/virt/kvm/arm/vgic/vgic.h
+++ b/virt/kvm/arm/vgic/vgic.h

@@ -104,6 +104,11 @@ static inline bool irq_is_pending(struct vgic_irq *irq)
 		return irq->pending_latch || irq->line_level;
 }
 
+static inline bool vgic_irq_is_mapped_level(struct vgic_irq *irq)
+{
+	return irq->config == VGIC_CONFIG_LEVEL && irq->hw;
+}
+
 /*
  * This struct provides an intermediate representation of the fields contained
  * in the GICH_VMCR and ICH_VMCR registers, such that code exporting the GIC

@@ -140,6 +145,8 @@ vgic_get_mmio_region(struct kvm_vcpu *vcpu, struct vgic_io_device *iodev,
 struct vgic_irq *vgic_get_irq(struct kvm *kvm, struct kvm_vcpu *vcpu,
 			      u32 intid);
 void vgic_put_irq(struct kvm *kvm, struct vgic_irq *irq);
+bool vgic_get_phys_line_level(struct vgic_irq *irq);
+void vgic_irq_set_phys_active(struct vgic_irq *irq, bool active);
 bool vgic_queue_irq_unlock(struct kvm *kvm, struct vgic_irq *irq);
 void vgic_kick_vcpus(struct kvm *kvm);