rpms/kernel/devel linux-2.6-kvm-77.patch, NONE, 1.1.4.2 kernel.spec, 1.1071, 1.1071.2.1

[Juan Quintela]

Author: quintela

Update of /cvs/pkgs/rpms/kernel/devel
In directory cvs1.fedora.phx.redhat.com:/tmp/cvs-serv30456

Modified Files:
      Tag: private-quintela-kvm-77b-branch
	kernel.spec 
Added Files:
      Tag: private-quintela-kvm-77b-branch
	linux-2.6-kvm-77.patch 
Log Message:
add kvm-77 patch

linux-2.6-kvm-77.patch:

--- NEW FILE linux-2.6-kvm-77.patch ---
diff --git a/MAINTAINERS b/MAINTAINERS
index 8dae455..c28aebf 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2409,7 +2409,14 @@ S:	Supported
 
 KERNEL VIRTUAL MACHINE (KVM)
 P:	Avi Kivity
-M:	avi at qumranet.com
+M:	avi at redhat.com
+L:	kvm at vger.kernel.org
+W:	http://kvm.qumranet.com
+S:	Supported
+
+KERNEL VIRTUAL MACHINE (KVM) FOR AMD-V
+P:	Joerg Roedel
+M:	joerg.roedel at amd.com
 L:	kvm at vger.kernel.org
 W:	http://kvm.qumranet.com
 S:	Supported
diff --git a/arch/ia64/include/asm/kvm_host.h b/arch/ia64/include/asm/kvm_host.h
index 1efe513..85db124 100644
--- a/arch/ia64/include/asm/kvm_host.h
+++ b/arch/ia64/include/asm/kvm_host.h
@@ -132,7 +132,7 @@
 #define GPFN_IOSAPIC        (4UL << 60) /* IOSAPIC base */
 #define GPFN_LEGACY_IO      (5UL << 60) /* Legacy I/O base */
 #define GPFN_GFW        (6UL << 60) /* Guest Firmware */
-#define GPFN_HIGH_MMIO      (7UL << 60) /* High MMIO range */
+#define GPFN_PHYS_MMIO      (7UL << 60) /* Directed MMIO Range */
 
 #define GPFN_IO_MASK        (7UL << 60) /* Guest pfn is I/O type */
 #define GPFN_INV_MASK       (1UL << 63) /* Guest pfn is invalid */
@@ -413,6 +413,10 @@ struct kvm_arch {
 	struct kvm_ioapic *vioapic;
 	struct kvm_vm_stat stat;
 	struct kvm_sal_data rdv_sal_data;
+
+	struct list_head assigned_dev_head;
+	struct dmar_domain *intel_iommu_domain;
+	struct hlist_head irq_ack_notifier_list;
 };
 
 union cpuid3_t {
diff --git a/arch/ia64/kvm/Kconfig b/arch/ia64/kvm/Kconfig
index 7914e48..8e99fed 100644
--- a/arch/ia64/kvm/Kconfig
+++ b/arch/ia64/kvm/Kconfig
@@ -46,4 +46,6 @@ config KVM_INTEL
 config KVM_TRACE
        bool
 
+source drivers/virtio/Kconfig
+
 endif # VIRTUALIZATION
diff --git a/arch/ia64/kvm/Makefile b/arch/ia64/kvm/Makefile
index bf22fb9..cf37f8f 100644
--- a/arch/ia64/kvm/Makefile
+++ b/arch/ia64/kvm/Makefile
@@ -44,7 +44,11 @@ EXTRA_CFLAGS += -Ivirt/kvm -Iarch/ia64/kvm/
 EXTRA_AFLAGS += -Ivirt/kvm -Iarch/ia64/kvm/
 
 common-objs = $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \
-		coalesced_mmio.o)
+		coalesced_mmio.o irq_comm.o)
+
+ifeq ($(CONFIG_DMAR),y)
+common-objs += $(addprefix ../../../virt/kvm/, vtd.o)
+endif
 
 kvm-objs := $(common-objs) kvm-ia64.o kvm_fw.o
 obj-$(CONFIG_KVM) += kvm.o
diff --git a/arch/ia64/kvm/irq.h b/arch/ia64/kvm/irq.h
new file mode 100644
index 0000000..c6786e8
--- /dev/null
+++ b/arch/ia64/kvm/irq.h
@@ -0,0 +1,31 @@
+/*
+ * irq.h: In-kernel interrupt controller related definitions
+ * Copyright (c) 2008, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
+ *
+ * Authors:
+ *   Xiantao Zhang <xiantao.zhang at intel.com>
+ *
+ */
+
+#ifndef __IRQ_H
+#define __IRQ_H
+
+static inline int irqchip_in_kernel(struct kvm *kvm)
+{
+	return 1;
+}
+
+#endif
diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c
index cd0d1a7..c0699f0 100644
--- a/arch/ia64/kvm/kvm-ia64.c
+++ b/arch/ia64/kvm/kvm-ia64.c
@@ -31,6 +31,7 @@
 #include <linux/bitops.h>
 #include <linux/hrtimer.h>
 #include <linux/uaccess.h>
+#include <linux/intel-iommu.h>
 
 #include <asm/pgtable.h>
 #include <asm/gcc_intrin.h>
@@ -45,6 +46,7 @@
 #include "iodev.h"
 #include "ioapic.h"
 #include "lapic.h"
+#include "irq.h"
 
 static unsigned long kvm_vmm_base;
 static unsigned long kvm_vsa_base;
@@ -179,12 +181,16 @@ int kvm_dev_ioctl_check_extension(long ext)
 	switch (ext) {
 	case KVM_CAP_IRQCHIP:
 	case KVM_CAP_USER_MEMORY:
+	case KVM_CAP_MP_STATE:
 
 		r = 1;
 		break;
 	case KVM_CAP_COALESCED_MMIO:
 		r = KVM_COALESCED_MMIO_PAGE_OFFSET;
 		break;
+	case KVM_CAP_IOMMU:
+		r = intel_iommu_found();
+		break;
 	default:
 		r = 0;
 	}
@@ -771,6 +777,7 @@ static void kvm_init_vm(struct kvm *kvm)
 	 */
 	kvm_build_io_pmt(kvm);
 
+	INIT_LIST_HEAD(&kvm->arch.assigned_dev_head);
 }
 
 struct  kvm *kvm_arch_create_vm(void)
@@ -1334,6 +1341,10 @@ static void kvm_release_vm_pages(struct kvm *kvm)
 
 void kvm_arch_destroy_vm(struct kvm *kvm)
 {
+	kvm_iommu_unmap_guest(kvm);
+#ifdef  KVM_CAP_DEVICE_ASSIGNMENT
+	kvm_free_all_assigned_devices(kvm);
+#endif
 	kfree(kvm->arch.vioapic);
 	kvm_release_vm_pages(kvm);
 	kvm_free_physmem(kvm);
@@ -1435,17 +1446,24 @@ int kvm_arch_set_memory_region(struct kvm *kvm,
 		int user_alloc)
 {
 	unsigned long i;
-	struct page *page;
+	unsigned long pfn;
 	int npages = mem->memory_size >> PAGE_SHIFT;
 	struct kvm_memory_slot *memslot = &kvm->memslots[mem->slot];
 	unsigned long base_gfn = memslot->base_gfn;
 
 	for (i = 0; i < npages; i++) {
-		page = gfn_to_page(kvm, base_gfn + i);
-		kvm_set_pmt_entry(kvm, base_gfn + i,
-				page_to_pfn(page) << PAGE_SHIFT,
-				_PAGE_AR_RWX|_PAGE_MA_WB);
-		memslot->rmap[i] = (unsigned long)page;
+		pfn = gfn_to_pfn(kvm, base_gfn + i);
+		if (!kvm_is_mmio_pfn(pfn)) {
+			kvm_set_pmt_entry(kvm, base_gfn + i,
+					pfn << PAGE_SHIFT,
+				_PAGE_AR_RWX | _PAGE_MA_WB);
+			memslot->rmap[i] = (unsigned long)pfn_to_page(pfn);
+		} else {
+			kvm_set_pmt_entry(kvm, base_gfn + i,
+					GPFN_PHYS_MMIO | (pfn << PAGE_SHIFT),
+					_PAGE_MA_UC);
+			memslot->rmap[i] = 0;
+			}
 	}
 
 	return 0;
@@ -1789,11 +1807,43 @@ int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu)
 int kvm_arch_vcpu_ioctl_get_mpstate(struct kvm_vcpu *vcpu,
 				    struct kvm_mp_state *mp_state)
[...9781 lines suppressed...]
 	relay_write(kt->rchan, &rec, size);
 }
 
diff --git a/virt/kvm/vtd.c b/virt/kvm/vtd.c
new file mode 100644
index 0000000..a770874
--- /dev/null
+++ b/virt/kvm/vtd.c
@@ -0,0 +1,191 @@
+/*
+ * Copyright (c) 2006, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
+ *
+ * Copyright (C) 2006-2008 Intel Corporation
+ * Copyright IBM Corporation, 2008
+ * Author: Allen M. Kay <allen.m.kay at intel.com>
+ * Author: Weidong Han <weidong.han at intel.com>
+ * Author: Ben-Ami Yassour <benami at il.ibm.com>
+ */
+
+#include <linux/list.h>
+#include <linux/kvm_host.h>
+#include <linux/pci.h>
+#include <linux/dmar.h>
+#include <linux/intel-iommu.h>
+
+static int kvm_iommu_unmap_memslots(struct kvm *kvm);
+static void kvm_iommu_put_pages(struct kvm *kvm,
+				gfn_t base_gfn, unsigned long npages);
+
+int kvm_iommu_map_pages(struct kvm *kvm,
+			gfn_t base_gfn, unsigned long npages)
+{
+	gfn_t gfn = base_gfn;
+	pfn_t pfn;
+	int i, r = 0;
+	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+
+	/* check if iommu exists and in use */
+	if (!domain)
+		return 0;
+
+	for (i = 0; i < npages; i++) {
+		/* check if already mapped */
+		pfn = (pfn_t)intel_iommu_iova_to_pfn(domain,
+						     gfn_to_gpa(gfn));
+		if (pfn)
+			continue;
+
+		pfn = gfn_to_pfn(kvm, gfn);
+		r = intel_iommu_page_mapping(domain,
+					     gfn_to_gpa(gfn),
+					     pfn_to_hpa(pfn),
+					     PAGE_SIZE,
+					     DMA_PTE_READ |
+					     DMA_PTE_WRITE);
+		if (r) {
+			printk(KERN_ERR "kvm_iommu_map_pages:"
+			       "iommu failed to map pfn=%lx\n", pfn);
+			goto unmap_pages;
+		}
+		gfn++;
+	}
+	return 0;
+
+unmap_pages:
+	kvm_iommu_put_pages(kvm, base_gfn, i);
+	return r;
+}
+
+static int kvm_iommu_map_memslots(struct kvm *kvm)
+{
+	int i, r;
+
+	down_read(&kvm->slots_lock);
+	for (i = 0; i < kvm->nmemslots; i++) {
+		r = kvm_iommu_map_pages(kvm, kvm->memslots[i].base_gfn,
+					kvm->memslots[i].npages);
+		if (r)
+			break;
+	}
+	up_read(&kvm->slots_lock);
+	return r;
+}
+
+int kvm_iommu_map_guest(struct kvm *kvm,
+			struct kvm_assigned_dev_kernel *assigned_dev)
+{
+	struct pci_dev *pdev = NULL;
+	int r;
+
+	if (!intel_iommu_found()) {
+		printk(KERN_ERR "%s: intel iommu not found\n", __func__);
+		return -ENODEV;
+	}
+
+	printk(KERN_DEBUG "VT-d direct map: host bdf = %x:%x:%x\n",
+	       assigned_dev->host_busnr,
+	       PCI_SLOT(assigned_dev->host_devfn),
+	       PCI_FUNC(assigned_dev->host_devfn));
+
+	pdev = assigned_dev->dev;
+
+	if (pdev == NULL) {
+		if (kvm->arch.intel_iommu_domain) {
+			intel_iommu_domain_exit(kvm->arch.intel_iommu_domain);
+			kvm->arch.intel_iommu_domain = NULL;
+		}
+		return -ENODEV;
+	}
+
+	kvm->arch.intel_iommu_domain = intel_iommu_domain_alloc(pdev);
+	if (!kvm->arch.intel_iommu_domain)
+		return -ENODEV;
+
+	r = kvm_iommu_map_memslots(kvm);
+	if (r)
+		goto out_unmap;
+
+	intel_iommu_detach_dev(kvm->arch.intel_iommu_domain,
+			       pdev->bus->number, pdev->devfn);
+
+	r = intel_iommu_context_mapping(kvm->arch.intel_iommu_domain,
+					pdev);
+	if (r) {
+		printk(KERN_ERR "Domain context map for %s failed",
+		       pci_name(pdev));
+		goto out_unmap;
+	}
+	return 0;
+
+out_unmap:
+	kvm_iommu_unmap_memslots(kvm);
+	return r;
+}
+
+static void kvm_iommu_put_pages(struct kvm *kvm,
+			       gfn_t base_gfn, unsigned long npages)
+{
+	gfn_t gfn = base_gfn;
+	pfn_t pfn;
+	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+	int i;
+
+	for (i = 0; i < npages; i++) {
+		pfn = (pfn_t)intel_iommu_iova_to_pfn(domain,
+						     gfn_to_gpa(gfn));
+		kvm_release_pfn_clean(pfn);
+		gfn++;
+	}
+}
+
+static int kvm_iommu_unmap_memslots(struct kvm *kvm)
+{
+	int i;
+	down_read(&kvm->slots_lock);
+	for (i = 0; i < kvm->nmemslots; i++) {
+		kvm_iommu_put_pages(kvm, kvm->memslots[i].base_gfn,
+				    kvm->memslots[i].npages);
+	}
+	up_read(&kvm->slots_lock);
+
+	return 0;
+}
+
+int kvm_iommu_unmap_guest(struct kvm *kvm)
+{
+	struct kvm_assigned_dev_kernel *entry;
+	struct dmar_domain *domain = kvm->arch.intel_iommu_domain;
+
+	/* check if iommu exists and in use */
+	if (!domain)
+		return 0;
+
+	list_for_each_entry(entry, &kvm->arch.assigned_dev_head, list) {
+		printk(KERN_DEBUG "VT-d unmap: host bdf = %x:%x:%x\n",
+		       entry->host_busnr,
+		       PCI_SLOT(entry->host_devfn),
+		       PCI_FUNC(entry->host_devfn));
+
+		/* detach kvm dmar domain */
+		intel_iommu_detach_dev(domain, entry->host_busnr,
+				       entry->host_devfn);
+	}
+	kvm_iommu_unmap_memslots(kvm);
+	intel_iommu_domain_exit(domain);
+	return 0;
+}


Index: kernel.spec
===================================================================
RCS file: /cvs/pkgs/rpms/kernel/devel/kernel.spec,v
retrieving revision 1.1071
retrieving revision 1.1071.2.1
diff -u -r1.1071 -r1.1071.2.1
--- kernel.spec	21 Oct 2008 21:08:57 -0000	1.1071
+++ kernel.spec	21 Oct 2008 21:19:34 -0000	1.1071.2.1
@@ -12,7 +12,7 @@
 # that the kernel isn't the stock distribution kernel, for example,
 # by setting the define to ".local" or ".bz123456"
 #
-#% define buildid .local
+%define buildid .kvm
 
 # fedora_build defines which build revision of this kernel version we're
 # building. Rather than incrementing forever, as with the prior versioning
@@ -703,6 +703,8 @@
 # Sony Vaio suspend fix
 Patch3100: linux-2.6.27-sony-laptop-suspend-fix.patch
 
+# kvm patch
+Patch4000: linux-2.6-kvm-77.patch
 %endif
 
 BuildRoot: %{_tmppath}/kernel-%{KVERREL}-root
@@ -1269,6 +1271,7 @@
 # hush pci bar allocation failures
 ApplyPatch linux-2.6.27-pci-hush-allocation-failures.patch
 
+ApplyPatch linux-2.6-kvm-77.patch
 # END OF PATCH APPLICATIONS
 
 %endif
@@ -1842,6 +1845,9 @@
 %kernel_variant_files -k vmlinux %{with_kdump} kdump
 
 %changelog
+* Mon Oct 21 2008 Juan Quintela <quintela at redhat.com>
+- Add kvm-77 patch.
+
 * Tue Oct 21 2008 Jarod Wilson <jarod at redhat.com> 2.6.27.3-35.rc1
 - improved lirc support for iMon LCD/IR devices
 - lirc support for additional MCE transceivers