From 01497bb352ef1b67c28b1276909b97033606f9cc Mon Sep 17 00:00:00 2001
From: Weidong Han <weidong.han@intel.com>
Date: Tue, 2 Dec 2008 21:24:23 +0800
Subject: [PATCH] KVM: support device assignment

Support device assignment, it can be used in device hotplug.


Signed-off-by: Weidong Han <weidong.han@intel.com>
---
 include/linux/kvm.h      |    5 +++++
 include/linux/kvm_host.h |    8 ++++++++
 virt/kvm/kvm_main.c      |   42 ++++++++++++++++++++++++++++++++++++++++++
 virt/kvm/vtd.c           |   24 ++++++++++++++++++++++++
 4 files changed, 79 insertions(+), 0 deletions(-)

diff --git a/include/linux/kvm.h b/include/linux/kvm.h
index 0997e6f..2904276 100644
--- a/include/linux/kvm.h
+++ b/include/linux/kvm.h
@@ -395,6 +395,9 @@ struct kvm_trace_rec {
 #if defined(CONFIG_X86)
 #define KVM_CAP_DEVICE_MSI 20
 #endif
+#if defined(CONFIG_X86)||defined(CONFIG_IA64)
+#define KVM_CAP_DEVICE_DEASSIGNMENT 21
+#endif
 
 /*
  * ioctls for VM fds
@@ -428,6 +431,8 @@ struct kvm_trace_rec {
 				   struct kvm_assigned_pci_dev)
 #define KVM_ASSIGN_IRQ _IOR(KVMIO, 0x70, \
 			    struct kvm_assigned_irq)
+#define KVM_DEASSIGN_PCI_DEVICE _IOR(KVMIO, 0x71, \
+				     struct kvm_assigned_pci_dev)
 
 /*
  * ioctls for vcpu fds
diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h
index aeabd32..cb1d404 100644
--- a/include/linux/kvm_host.h
+++ b/include/linux/kvm_host.h
@@ -333,6 +333,8 @@ int kvm_iommu_map_guest(struct kvm *kvm);
 int kvm_iommu_unmap_guest(struct kvm *kvm);
 int kvm_assign_device(struct kvm *kvm,
 		      struct kvm_assigned_dev_kernel *assigned_dev);
+int kvm_deassign_device(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *assigned_dev);
 #else /* CONFIG_DMAR */
 static inline int kvm_iommu_map_pages(struct kvm *kvm,
 				      gfn_t base_gfn,
@@ -356,6 +358,12 @@ static inline int kvm_assign_device(struct kvm *kvm,
 {
 	return 0;
 }
+
+static inline int kvm_deassign_device(struct kvm *kvm,
+		struct kvm_assigned_dev_kernel *assigned_dev)
+{
+	return 0;
+}
 #endif /* CONFIG_DMAR */
 
 static inline void kvm_guest_enter(void)
diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c
index 43a5236..fe6aba0 100644
--- a/virt/kvm/kvm_main.c
+++ b/virt/kvm/kvm_main.c
@@ -499,6 +499,35 @@ out_free:
 }
 #endif
 
+#ifdef KVM_CAP_DEVICE_DEASSIGNMENT
+static int kvm_vm_ioctl_deassign_device(struct kvm *kvm,
+		struct kvm_assigned_pci_dev *assigned_dev)
+{
+	int r = 0;
+	struct kvm_assigned_dev_kernel *match;
+
+	mutex_lock(&kvm->lock);
+
+	match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head,
+				      assigned_dev->assigned_dev_id);
+	if (!match) {
+		printk(KERN_INFO "%s: device hasn't been assigned before, "
+		  "so cannot be deassigned\n", __func__);
+		r = -EINVAL;
+		goto out;
+	}
+
+	if (assigned_dev->flags & KVM_DEV_ASSIGN_ENABLE_IOMMU)
+		kvm_deassign_device(kvm, match);
+
+	kvm_free_assigned_device(kvm, match);
+
+out:
+	mutex_unlock(&kvm->lock);
+	return r;
+}
+#endif
+
 static inline int valid_vcpu(int n)
 {
 	return likely(n >= 0 && n < KVM_MAX_VCPUS);
@@ -1838,6 +1867,19 @@ static long kvm_vm_ioctl(struct file *filp,
 		break;
 	}
 #endif
+#ifdef KVM_CAP_DEVICE_DEASSIGNMENT
+	case KVM_DEASSIGN_PCI_DEVICE: {
+		struct kvm_assigned_pci_dev assigned_dev;
+
+		r = -EFAULT;
+		if (copy_from_user(&assigned_dev, argp, sizeof assigned_dev))
+			goto out;
+		r = kvm_vm_ioctl_deassign_device(kvm, &assigned_dev);
+		if (r)
+			goto out;
+		break;
+	}
+#endif
 	default:
 		r = kvm_arch_vm_ioctl(filp, ioctl, arg);
 	}
diff --git a/virt/kvm/vtd.c b/virt/kvm/vtd.c
index 44bb58a..174ea1f 100644
--- a/virt/kvm/vtd.c
+++ b/virt/kvm/vtd.c
@@ -116,6 +116,30 @@ int kvm_assign_device(struct kvm *kvm,
 	return 0;
 }
 
+int kvm_deassign_device(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *assigned_dev)
+{
+	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	struct pci_dev *pdev = NULL;
+
+	/* check if iommu exists and in use */
+	if (!domain)
+		return 0;
+
+	pdev = assigned_dev->dev;
+	if (pdev == NULL)
+		return -ENODEV;
+
+	intel_iommu_detach_device(domain, pdev);
+
+	printk(KERN_DEBUG "deassign device: host bdf = %x:%x:%x\n",
+		assigned_dev->host_busnr,
+		PCI_SLOT(assigned_dev->host_devfn),
+		PCI_FUNC(assigned_dev->host_devfn));
+
+	return 0;
+}
+
 int kvm_iommu_map_guest(struct kvm *kvm)
 {
 	int r;
-- 
1.5.1