Support sharing of guest IRQ levels for assigned devices

Support sharing of guest IRQ levels for assigned devices

[Amit Shah]

I've tested Sheng's latest patches and they work fine. This series
is essentially the same, rebased on top of HEAD after the move of
the device assignment code to virt/kvm/kvm_main.c from arch/x86/kvm/x86.c

Also contains a small fix to request the irq source only when we register
the interrupt handler.

Please apply for 2.6.28.

Amit.

[PATCH 1/2] KVM: Handle multiple interrupt sources

[Amit Shah]

From: Sheng Yang <sheng@linux.intel.com>

Keep a record of current interrupt state before injecting. Don't
assert/deassert repeatedly, so that every caller of kvm_set_irq()
can be identified as a separate interrupt source for the IOAPIC/PIC
to implement logical OR of level triggered interrupts on one IRQ line.

Notice that userspace devices are treated as one device for each IRQ
line. The correctness of sharing interrupt for each IRQ line should be
ensured by the userspace program (QEmu).

[Amit: rebase to kvm.git HEAD]

Signed-off-by: Sheng Yang <sheng@linux.intel.com>
Signed-off-by: Amit Shah <amit.shah@redhat.com>
---
 arch/x86/kvm/x86.c       |   13 ++++++++++++-
 include/linux/kvm_host.h |    3 +++
 virt/kvm/kvm_main.c      |   12 +++++++++---
 3 files changed, 24 insertions(+), 4 deletions(-)

diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index dda478e..6f45428 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1816,7 +1816,18 @@ long kvm_arch_vm_ioctl(struct file *filp,
 			goto out;
 		if (irqchip_in_kernel(kvm)) {
 			mutex_lock(&kvm->lock);
-			kvm_set_irq(kvm, irq_event.irq, irq_event.level);
+			/*
+			 * Take one IRQ line as from one device, shared IRQ
+			 * line should also be handled in the userspace before
+			 * use KVM_IRQ_LINE ioctl to change IRQ line state.
+			 */
+			if (kvm->userspace_intrsource_states[irq_event.irq]
+					!= irq_event.level) {
+				kvm_set_irq(kvm, irq_event.irq,
+					    irq_event.level);
+				kvm->userspace_intrsource_states[irq_event.irq]
+						= irq_event.level;
+			}
 			mutex_unlock(&kvm->lock);
 			r = 0;
 		}
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index 3833c48..d392e31 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -129,6 +129,8 @@ struct kvm {
 	unsigned long mmu_notifier_seq;
 	long mmu_notifier_count;
 #endif
+
+	int userspace_intrsource_states[KVM_IOAPIC_NUM_PINS];
 };
 
 /* The guest did something we don't support. */
@@ -306,6 +308,7 @@ struct kvm_assigned_dev_kernel {
 	int host_irq;
 	int guest_irq;
 	int irq_requested;
+	int irq_state;
 	struct pci_dev *dev;
 	struct kvm *kvm;
 };
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index cf0ab8e..faa56fb 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -104,8 +104,11 @@ static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
 	 * finer-grained lock, update this
 	 */
 	mutex_lock(&assigned_dev->kvm->lock);
-	kvm_set_irq(assigned_dev->kvm,
-		    assigned_dev->guest_irq, 1);
+	if (assigned_dev->irq_state == 0) {
+		kvm_set_irq(assigned_dev->kvm,
+			    assigned_dev->guest_irq, 1);
+		assigned_dev->irq_state = 1;
+	}
 	mutex_unlock(&assigned_dev->kvm->lock);
 	kvm_put_kvm(assigned_dev->kvm);
 }
@@ -134,7 +137,10 @@ static void kvm_assigned_dev_ack_irq(struct kvm_irq_ack_notifier *kian)
 
 	dev = container_of(kian, struct kvm_assigned_dev_kernel,
 			   ack_notifier);
-	kvm_set_irq(dev->kvm, dev->guest_irq, 0);
+	if (dev->irq_state == 1) {
+		kvm_set_irq(dev->kvm, dev->guest_irq, 0);
+		dev->irq_state = 0;
+	}
 	enable_irq(dev->host_irq);
 }
 
-- 
1.5.4.3

[PATCH 2/2] KVM: Device Assignment: Fix guest shared interrupts with in-kernel irqchip

[Amit Shah]

From: Sheng Yang <sheng@linux.intel.com>

Derived from Avi's suggestion, now every call of kvm_set_irq() should offer
an irq_source_id, which is allocated by kvm_request_irq_source_id(). We based
on irq_source_id to identify irq source and implement logical OR for shared
level interrupts.

The allocated irq_source_id can be freed by kvm_free_irq_sources().

As of now we support sizeof(unsigned long) different irq sources at most.

[Amit: - rebase to kvm.git HEAD
       - move kvm_request_irq_source_id to the update_irq ioctl]

    Signed-off-by: Sheng Yang <sheng@linux.intel.com>
    Signed-off-by: Amit Shah <amit.shah@redhat.com>

Signed-off-by: Amit Shah <amit.shah@redhat.com>
---
 arch/x86/kvm/i8254.c       |   11 +++++++++--
 arch/x86/kvm/i8254.h       |    1 +
 arch/x86/kvm/irq.h         |    5 +++++
 arch/x86/kvm/x86.c         |    7 +++++--
 include/asm-x86/kvm_host.h |    3 +++
 include/linux/kvm_host.h   |    3 ++-
 virt/kvm/irq_comm.c        |   42 +++++++++++++++++++++++++++++++++++++++---
 virt/kvm/kvm_main.c        |   13 ++++++++-----
 8 files changed, 72 insertions(+), 13 deletions(-)

diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c
index 6144d3f..da06919 100644
--- a/arch/x86/kvm/i8254.c
+++ b/arch/x86/kvm/i8254.c
@@ -545,6 +545,12 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm)
 	if (!pit)
 		return NULL;
 
+	mutex_lock(&kvm->lock);
+	pit->irq_source_id = kvm_request_irq_source_id(kvm);
+	mutex_unlock(&kvm->lock);
+	if (pit->irq_source_id < 0)
+		return NULL;
+
 	mutex_init(&pit->pit_state.lock);
 	mutex_lock(&pit->pit_state.lock);
 	spin_lock_init(&pit->pit_state.inject_lock);
@@ -587,6 +593,7 @@ void kvm_free_pit(struct kvm *kvm)
 		mutex_lock(&kvm->arch.vpit->pit_state.lock);
 		timer = &kvm->arch.vpit->pit_state.pit_timer.timer;
 		hrtimer_cancel(timer);
+		kvm_free_irq_source_id(kvm, kvm->arch.vpit->irq_source_id);
 		mutex_unlock(&kvm->arch.vpit->pit_state.lock);
 		kfree(kvm->arch.vpit);
 	}
@@ -598,8 +605,8 @@ static void __inject_pit_timer_intr(struct kvm *kvm)
 	int i;
 
 	mutex_lock(&kvm->lock);
-	kvm_set_irq(kvm, 0, 1);
-	kvm_set_irq(kvm, 0, 0);
+	kvm_set_irq(kvm, kvm->arch.vpit->irq_source_id, 0, 1);
+	kvm_set_irq(kvm, kvm->arch.vpit->irq_source_id, 0, 0);
 	mutex_unlock(&kvm->lock);
 
 	/*
diff --git a/arch/x86/kvm/i8254.h b/arch/x86/kvm/i8254.h
index e436d49..4178022 100644
--- a/arch/x86/kvm/i8254.h
+++ b/arch/x86/kvm/i8254.h
@@ -44,6 +44,7 @@ struct kvm_pit {
 	struct kvm_io_device speaker_dev;
 	struct kvm *kvm;
 	struct kvm_kpit_state pit_state;
+	int irq_source_id;
 };
 
 #define KVM_PIT_BASE_ADDRESS	    0x40
diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h
index 71e37a5..9c6d9a5 100644
--- a/arch/x86/kvm/irq.h
+++ b/arch/x86/kvm/irq.h
@@ -32,6 +32,8 @@
 
 #define PIC_NUM_PINS 16
 
+#define KVM_USERSPACE_IRQ_SOURCE_ID	0
+
 struct kvm;
 struct kvm_vcpu;
 
@@ -95,4 +97,7 @@ void __kvm_migrate_timers(struct kvm_vcpu *vcpu);
 int pit_has_pending_timer(struct kvm_vcpu *vcpu);
 int apic_has_pending_timer(struct kvm_vcpu *vcpu);
 
+int kvm_request_irq_source_id(struct kvm *kvm);
+void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id);
+
 #endif
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 6f45428..4388c44 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -1823,8 +1823,8 @@ long kvm_arch_vm_ioctl(struct file *filp,
 			 */
 			if (kvm->userspace_intrsource_states[irq_event.irq]
 					!= irq_event.level) {
-				kvm_set_irq(kvm, irq_event.irq,
-					    irq_event.level);
+				kvm_set_irq(kvm, KVM_USERSPACE_IRQ_SOURCE_ID,
+					    irq_event.irq, irq_event.level);
 				kvm->userspace_intrsource_states[irq_event.irq]
 						= irq_event.level;
 			}
@@ -4126,6 +4126,9 @@ struct  kvm *kvm_arch_create_vm(void)
 	INIT_LIST_HEAD(&kvm->arch.active_mmu_pages);
 	INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
 
+	/* Reserve bit 0 of irq_sources_bitmap for userspace irq source */
+	kvm->arch.irq_sources_bitmap = 1;
+
 	return kvm;
 }
 
diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h
index 4b06ca8..f01e92e 100644
--- a/include/asm-x86/kvm_host.h
+++ b/include/asm-x86/kvm_host.h
@@ -368,6 +368,9 @@ struct kvm_arch{
 
 	struct page *ept_identity_pagetable;
 	bool ept_identity_pagetable_done;
+
+	unsigned long irq_sources_bitmap;
+	unsigned long irq_states[KVM_IOAPIC_NUM_PINS];
 };
 
 struct kvm_vm_stat {
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index d392e31..c6e2cc3 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -308,11 +308,12 @@ struct kvm_assigned_dev_kernel {
 	int host_irq;
 	int guest_irq;
 	int irq_requested;
+	int irq_source_id;
 	int irq_state;
 	struct pci_dev *dev;
 	struct kvm *kvm;
 };
-void kvm_set_irq(struct kvm *kvm, int irq, int level);
+void kvm_set_irq(struct kvm *kvm, int irq_source_id, int irq, int level);
 void kvm_notify_acked_irq(struct kvm *kvm, unsigned gsi);
 void kvm_register_irq_ack_notifier(struct kvm *kvm,
 				   struct kvm_irq_ack_notifier *kian);
diff --git a/virt/kvm/irq_comm.c b/virt/kvm/irq_comm.c
index d0169f5..a014831 100644
--- a/virt/kvm/irq_comm.c
+++ b/virt/kvm/irq_comm.c
@@ -25,15 +25,23 @@
 #include "ioapic.h"
 
 /* This should be called with the kvm->lock mutex held */
-void kvm_set_irq(struct kvm *kvm, int irq, int level)
+void kvm_set_irq(struct kvm *kvm, int irq_source_id, int irq, int level)
 {
+	unsigned long *irq_state = &kvm->arch.irq_states[irq];
+
+	/* Logical OR for level trig interrupt */
+	if (level)
+		set_bit(irq_source_id, irq_state);
+	else
+		clear_bit(irq_source_id, irq_state);
+
 	/* Not possible to detect if the guest uses the PIC or the
 	 * IOAPIC.  So set the bit in both. The guest will ignore
 	 * writes to the unused one.
 	 */
-	kvm_ioapic_set_irq(kvm->arch.vioapic, irq, level);
+	kvm_ioapic_set_irq(kvm->arch.vioapic, irq, !!(*irq_state));
 #ifdef CONFIG_X86
-	kvm_pic_set_irq(pic_irqchip(kvm), irq, level);
+	kvm_pic_set_irq(pic_irqchip(kvm), irq, !!(*irq_state));
 #endif
 }
 
@@ -58,3 +66,31 @@ void kvm_unregister_irq_ack_notifier(struct kvm *kvm,
 {
 	hlist_del(&kian->link);
 }
+
+/* The caller must hold kvm->lock mutex */
+int kvm_request_irq_source_id(struct kvm *kvm)
+{
+	unsigned long *bitmap = &kvm->arch.irq_sources_bitmap;
+	int irq_source_id = find_first_zero_bit(bitmap,
+				sizeof(kvm->arch.irq_sources_bitmap));
+	if (irq_source_id >= sizeof(kvm->arch.irq_sources_bitmap)) {
+		printk(KERN_WARNING "kvm: exhaust allocatable IRQ sources!\n");
+		irq_source_id = -EFAULT;
+	} else
+		set_bit(irq_source_id, bitmap);
+	return irq_source_id;
+}
+
+void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id)
+{
+	int i;
+
+	if (irq_source_id <= 0 ||
+	    irq_source_id >= sizeof(kvm->arch.irq_sources_bitmap)) {
+		printk(KERN_ERR "kvm: IRQ source ID out of range!\n");
+		return;
+	}
+	for (i = 0; i < KVM_IOAPIC_NUM_PINS; i++)
+		clear_bit(irq_source_id, &kvm->arch.irq_states[i]);
+	clear_bit(irq_source_id, &kvm->arch.irq_sources_bitmap);
+}
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index faa56fb..0479898 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -106,6 +106,7 @@ static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
 	mutex_lock(&assigned_dev->kvm->lock);
 	if (assigned_dev->irq_state == 0) {
 		kvm_set_irq(assigned_dev->kvm,
+			    assigned_dev->irq_source_id,
 			    assigned_dev->guest_irq, 1);
 		assigned_dev->irq_state = 1;
 	}
@@ -113,9 +114,6 @@ static void kvm_assigned_dev_interrupt_work_handler(struct work_struct *work)
 	kvm_put_kvm(assigned_dev->kvm);
 }
 
-/* FIXME: Implement the OR logic needed to make shared interrupts on
- * this line behave properly
- */
 static irqreturn_t kvm_assigned_dev_intr(int irq, void *dev_id)
 {
 	struct kvm_assigned_dev_kernel *assigned_dev =
@@ -138,7 +136,7 @@ static void kvm_assigned_dev_ack_irq(struct kvm_irq_ack_notifier *kian)
 	dev = container_of(kian, struct kvm_assigned_dev_kernel,
 			   ack_notifier);
 	if (dev->irq_state == 1) {
-		kvm_set_irq(dev->kvm, dev->guest_irq, 0);
+		kvm_set_irq(dev->kvm, dev->irq_source_id, dev->guest_irq, 0);
 		dev->irq_state = 0;
 	}
 	enable_irq(dev->host_irq);
@@ -152,6 +150,7 @@ static void kvm_free_assigned_device(struct kvm *kvm,
 		free_irq(assigned_dev->host_irq, (void *)assigned_dev);
 
 	kvm_unregister_irq_ack_notifier(kvm, &assigned_dev->ack_notifier);
+	kvm_free_irq_source_id(kvm, assigned_dev->irq_source_id);
 
 	if (cancel_work_sync(&assigned_dev->interrupt_work))
 		/* We had pending work. That means we will have to take
@@ -221,6 +220,11 @@ static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 		match->ack_notifier.gsi = assigned_irq->guest_irq;
 		match->ack_notifier.irq_acked = kvm_assigned_dev_ack_irq;
 		kvm_register_irq_ack_notifier(kvm, &match->ack_notifier);
+		r = kvm_request_irq_source_id(kvm);
+		if (r < 0)
+			goto out_release;
+		else
+			match->irq_source_id = r;
 
 		/* Even though this is PCI, we don't want to use shared
 		 * interrupts. Sharing host devices with guest-assigned devices
@@ -233,7 +237,6 @@ static int kvm_vm_ioctl_assign_irq(struct kvm *kvm,
 			goto out_release;
 		}
 	}
-
 	match->irq_requested = true;
 	mutex_unlock(&kvm->lock);
 	return r;
-- 
1.5.4.3

Re: Support sharing of guest IRQ levels for assigned devices

[Amit Shah]

* On Saturday 11 Oct 2008 13:40:50 Amit Shah wrote:
> I've tested Sheng's latest patches and they work fine. This series
> is essentially the same, rebased on top of HEAD after the move of
> the device assignment code to virt/kvm/kvm_main.c from arch/x86/kvm/x86.c
>
> Also contains a small fix to request the irq source only when we register
> the interrupt handler.

I forgot to mention: Xiantao, please check the IA64 build and also if any 
other changes might be necessary.

Amit

Re: [PATCH 1/2] KVM: Handle multiple interrupt sources

[Sheng Yang]

On Saturday 11 October 2008 16:10:51 Amit Shah wrote:
> From: Sheng Yang <sheng@linux.intel.com>
>
> Keep a record of current interrupt state before injecting. Don't
> assert/deassert repeatedly, so that every caller of kvm_set_irq()
> can be identified as a separate interrupt source for the IOAPIC/PIC
> to implement logical OR of level triggered interrupts on one IRQ line.
>
> Notice that userspace devices are treated as one device for each IRQ
> line. The correctness of sharing interrupt for each IRQ line should be
> ensured by the userspace program (QEmu).
>
> [Amit: rebase to kvm.git HEAD]

Hi, Amit

Thanks for your work! 

But maybe I miss something. I suppose my later patch can work indepently? I 
think the second patch should solve the whole problem (sorry to reply it to 
the second rather than [0/2] which made confusion...). Can you have a check?

Thanks!
--
regards
Yang, Sheng
>
> Signed-off-by: Sheng Yang <sheng@linux.intel.com>
> Signed-off-by: Amit Shah <amit.shah@redhat.com>
> ---
>  arch/x86/kvm/x86.c       |   13 ++++++++++++-
>  include/linux/kvm_host.h |    3 +++
>  virt/kvm/kvm_main.c      |   12 +++++++++---
>  3 files changed, 24 insertions(+), 4 deletions(-)
>
> diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
> index dda478e..6f45428 100644
> --- a/arch/x86/kvm/x86.c
> +++ b/arch/x86/kvm/x86.c
> @@ -1816,7 +1816,18 @@ long kvm_arch_vm_ioctl(struct file *filp,
>  			goto out;
>  		if (irqchip_in_kernel(kvm)) {
>  			mutex_lock(&kvm->lock);
> -			kvm_set_irq(kvm, irq_event.irq, irq_event.level);
> +			/*
> +			 * Take one IRQ line as from one device, shared IRQ
> +			 * line should also be handled in the userspace before
> +			 * use KVM_IRQ_LINE ioctl to change IRQ line state.
> +			 */
> +			if (kvm->userspace_intrsource_states[irq_event.irq]
> +					!= irq_event.level) {
> +				kvm_set_irq(kvm, irq_event.irq,
> +					    irq_event.level);
> +				kvm->userspace_intrsource_states[irq_event.irq]
> +						= irq_event.level;
> +			}
>  			mutex_unlock(&kvm->lock);
>  			r = 0;
>  		}
> diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
> index 3833c48..d392e31 100644
> --- a/include/linux/kvm_host.h
> +++ b/include/linux/kvm_host.h
> @@ -129,6 +129,8 @@ struct kvm {
>  	unsigned long mmu_notifier_seq;
>  	long mmu_notifier_count;
>  #endif
> +
> +	int userspace_intrsource_states[KVM_IOAPIC_NUM_PINS];
>  };
>
>  /* The guest did something we don't support. */
> @@ -306,6 +308,7 @@ struct kvm_assigned_dev_kernel {
>  	int host_irq;
>  	int guest_irq;
>  	int irq_requested;
> +	int irq_state;
>  	struct pci_dev *dev;
>  	struct kvm *kvm;
>  };
> diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
> index cf0ab8e..faa56fb 100644
> --- a/virt/kvm/kvm_main.c
> +++ b/virt/kvm/kvm_main.c
> @@ -104,8 +104,11 @@ static void
> kvm_assigned_dev_interrupt_work_handler(struct work_struct *work) *
> finer-grained lock, update this
>  	 */
>  	mutex_lock(&assigned_dev->kvm->lock);
> -	kvm_set_irq(assigned_dev->kvm,
> -		    assigned_dev->guest_irq, 1);
> +	if (assigned_dev->irq_state == 0) {
> +		kvm_set_irq(assigned_dev->kvm,
> +			    assigned_dev->guest_irq, 1);
> +		assigned_dev->irq_state = 1;
> +	}
>  	mutex_unlock(&assigned_dev->kvm->lock);
>  	kvm_put_kvm(assigned_dev->kvm);
>  }
> @@ -134,7 +137,10 @@ static void kvm_assigned_dev_ack_irq(struct
> kvm_irq_ack_notifier *kian)
>
>  	dev = container_of(kian, struct kvm_assigned_dev_kernel,
>  			   ack_notifier);
> -	kvm_set_irq(dev->kvm, dev->guest_irq, 0);
> +	if (dev->irq_state == 1) {
> +		kvm_set_irq(dev->kvm, dev->guest_irq, 0);
> +		dev->irq_state = 0;
> +	}
>  	enable_irq(dev->host_irq);
>  }

Re: [PATCH 1/2] KVM: Handle multiple interrupt sources

[Amit Shah]

----- "Sheng Yang" <sheng@linux.intel.com> wrote:

> On Saturday 11 October 2008 16:10:51 Amit Shah wrote:
> > From: Sheng Yang <sheng@linux.intel.com>
> >
> > Keep a record of current interrupt state before injecting. Don't
> > assert/deassert repeatedly, so that every caller of kvm_set_irq()
> > can be identified as a separate interrupt source for the IOAPIC/PIC
> > to implement logical OR of level triggered interrupts on one IRQ
> line.
> >
> > Notice that userspace devices are treated as one device for each
> IRQ
> > line. The correctness of sharing interrupt for each IRQ line should
> be
> > ensured by the userspace program (QEmu).
> >
> > [Amit: rebase to kvm.git HEAD]
> 
> Hi, Amit
> 
> Thanks for your work! 
> 
> But maybe I miss something. I suppose my later patch can work
> indepently? I 
> think the second patch should solve the whole problem (sorry to reply
> it to 
> the second rather than [0/2] which made confusion...). Can you have a
> check?

I'm not sure I understand. Which concern are you talking about?

I used the latest patch that you sent and I also verified that it works.

Amit

Re: [PATCH 1/2] KVM: Handle multiple interrupt sources

[Sheng Yang]

On Monday 13 October 2008 13:06:18 Amit Shah wrote:
> ----- "Sheng Yang" <sheng@linux.intel.com> wrote:
> > On Saturday 11 October 2008 16:10:51 Amit Shah wrote:
> > > From: Sheng Yang <sheng@linux.intel.com>
> > >
> > > Keep a record of current interrupt state before injecting. Don't
> > > assert/deassert repeatedly, so that every caller of kvm_set_irq()
> > > can be identified as a separate interrupt source for the IOAPIC/PIC
> > > to implement logical OR of level triggered interrupts on one IRQ
> >
> > line.
> >
> > > Notice that userspace devices are treated as one device for each
> >
> > IRQ
> >
> > > line. The correctness of sharing interrupt for each IRQ line should
> >
> > be
> >
> > > ensured by the userspace program (QEmu).
> > >
> > > [Amit: rebase to kvm.git HEAD]
> >
> > Hi, Amit
> >
> > Thanks for your work!
> >
> > But maybe I miss something. I suppose my later patch can work
> > indepently? I
> > think the second patch should solve the whole problem (sorry to reply
> > it to
> > the second rather than [0/2] which made confusion...). Can you have a
> > check?
>
> I'm not sure I understand. Which concern are you talking about?
>
> I used the latest patch that you sent and I also verified that it works.
>
Well, at least I meant to replace all of my first two patches with my later 
one... I suppose the second patch(my later one, derived from Avi's 
suggestion) should work alone without the first one...
--
regards
Yang, Sheng

Re: [PATCH 1/2] KVM: Handle multiple interrupt sources

[Amit Shah]

----- "Sheng Yang" <sheng@linux.intel.com> wrote:
> Well, at least I meant to replace all of my first two patches with my
> later 
> one... I suppose the second patch(my later one, derived from Avi's 
> suggestion) should work alone without the first one...

OK; I just saw that. I'll rebase and test again.