From 283d0c65e4c277cf81d71418341c74eee199d149 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 20 Dec 2009 14:25:19 +0200 Subject: KVM: Disentangle mmu notifiers and coalesced_mmio registration They aren't related. Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 18 +++++++----------- 1 file changed, 7 insertions(+), 11 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index a944be392d6e..92f5c98b6c4d 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -374,17 +374,16 @@ static struct kvm *kvm_create_vm(void) #endif #if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER) - { - kvm->mmu_notifier.ops = &kvm_mmu_notifier_ops; - r = mmu_notifier_register(&kvm->mmu_notifier, current->mm); - if (r) { + kvm->mmu_notifier.ops = &kvm_mmu_notifier_ops; + r = mmu_notifier_register(&kvm->mmu_notifier, current->mm); +#endif + + if (r) { #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET - put_page(page); + put_page(page); #endif - goto out_err; - } + goto out_err; } -#endif kvm->mm = current->mm; atomic_inc(&kvm->mm->mm_count); @@ -406,11 +405,8 @@ static struct kvm *kvm_create_vm(void) out: return kvm; -#if defined(KVM_COALESCED_MMIO_PAGE_OFFSET) || \ - (defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER)) out_err: hardware_disable_all(); -#endif out_err_nodisable: kfree(kvm); return ERR_PTR(r); -- cgit v1.2.3 From 3f0852e0aca9de2d2f8bc756630a82352cbfde68 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 20 Dec 2009 14:42:19 +0200 Subject: KVM: Add include guards for coalesced_mmio.h Signed-off-by: Avi Kivity --- virt/kvm/coalesced_mmio.h | 5 +++++ 1 file changed, 5 insertions(+) (limited to 'virt') diff --git a/virt/kvm/coalesced_mmio.h b/virt/kvm/coalesced_mmio.h index 4b49f27fa31e..e7033469733d 100644 --- a/virt/kvm/coalesced_mmio.h +++ b/virt/kvm/coalesced_mmio.h @@ -1,3 +1,6 @@ +#ifndef __KVM_COALESCED_MMIO_H__ +#define __KVM_COALESCED_MMIO_H__ + /* * KVM coalesced MMIO * @@ -22,3 +25,5 @@ int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, struct kvm_coalesced_mmio_zone *zone); int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, struct kvm_coalesced_mmio_zone *zone); + +#endif -- cgit v1.2.3 From 4c07b0a4b6df45fd8c6d07d838b4faf65eed2ba4 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 20 Dec 2009 14:54:04 +0200 Subject: KVM: Remove ifdefs from mmu notifier initialization Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 92f5c98b6c4d..c10d117c4317 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -341,6 +341,20 @@ static const struct mmu_notifier_ops kvm_mmu_notifier_ops = { .change_pte = kvm_mmu_notifier_change_pte, .release = kvm_mmu_notifier_release, }; + +static int kvm_init_mmu_notifier(struct kvm *kvm) +{ + kvm->mmu_notifier.ops = &kvm_mmu_notifier_ops; + return mmu_notifier_register(&kvm->mmu_notifier, current->mm); +} + +#else /* !(CONFIG_MMU_NOTIFIER && KVM_ARCH_WANT_MMU_NOTIFIER) */ + +static int kvm_init_mmu_notifier(struct kvm *kvm) +{ + return 0; +} + #endif /* CONFIG_MMU_NOTIFIER && KVM_ARCH_WANT_MMU_NOTIFIER */ static struct kvm *kvm_create_vm(void) @@ -373,11 +387,7 @@ static struct kvm *kvm_create_vm(void) (struct kvm_coalesced_mmio_ring *)page_address(page); #endif -#if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER) - kvm->mmu_notifier.ops = &kvm_mmu_notifier_ops; - r = mmu_notifier_register(&kvm->mmu_notifier, current->mm); -#endif - + r = kvm_init_mmu_notifier(kvm); if (r) { #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET put_page(page); -- cgit v1.2.3 From 50eb2a3cd0f50d912b26d0b79b7f443344608390 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 20 Dec 2009 15:00:10 +0200 Subject: KVM: Add KVM_MMIO kconfig item s390 doesn't have mmio, this will simplify ifdefing it out. Signed-off-by: Avi Kivity --- virt/kvm/Kconfig | 3 +++ 1 file changed, 3 insertions(+) (limited to 'virt') diff --git a/virt/kvm/Kconfig b/virt/kvm/Kconfig index daece36c0a57..7f1178f6b839 100644 --- a/virt/kvm/Kconfig +++ b/virt/kvm/Kconfig @@ -12,3 +12,6 @@ config HAVE_KVM_EVENTFD config KVM_APIC_ARCHITECTURE bool + +config KVM_MMIO + bool -- cgit v1.2.3 From 980da6ce573b7c40886406674ff8f022a975e65e Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Sun, 20 Dec 2009 15:13:43 +0200 Subject: KVM: Simplify coalesced mmio initialization - add destructor function - move related allocation into constructor - add stubs for !CONFIG_KVM_MMIO Signed-off-by: Avi Kivity --- virt/kvm/coalesced_mmio.c | 25 +++++++++++++++++++++++-- virt/kvm/coalesced_mmio.h | 10 ++++++++++ virt/kvm/kvm_main.c | 7 +------ 3 files changed, 34 insertions(+), 8 deletions(-) (limited to 'virt') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index 04d69cd7049b..d68e6c68e0ff 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -92,11 +92,19 @@ static const struct kvm_io_device_ops coalesced_mmio_ops = { int kvm_coalesced_mmio_init(struct kvm *kvm) { struct kvm_coalesced_mmio_dev *dev; + struct page *page; int ret; + ret = -ENOMEM; + page = alloc_page(GFP_KERNEL | __GFP_ZERO); + if (!page) + goto out_err; + kvm->coalesced_mmio_ring = page_address(page); + + ret = -ENOMEM; dev = kzalloc(sizeof(struct kvm_coalesced_mmio_dev), GFP_KERNEL); if (!dev) - return -ENOMEM; + goto out_free_page; spin_lock_init(&dev->lock); kvm_iodevice_init(&dev->dev, &coalesced_mmio_ops); dev->kvm = kvm; @@ -104,11 +112,24 @@ int kvm_coalesced_mmio_init(struct kvm *kvm) ret = kvm_io_bus_register_dev(kvm, &kvm->mmio_bus, &dev->dev); if (ret < 0) - kfree(dev); + goto out_free_dev; + + return ret; +out_free_dev: + kfree(dev); +out_free_page: + __free_page(page); +out_err: return ret; } +void kvm_coalesced_mmio_free(struct kvm *kvm) +{ + if (kvm->coalesced_mmio_ring) + free_page((unsigned long)kvm->coalesced_mmio_ring); +} + int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, struct kvm_coalesced_mmio_zone *zone) { diff --git a/virt/kvm/coalesced_mmio.h b/virt/kvm/coalesced_mmio.h index e7033469733d..8a5959e3535f 100644 --- a/virt/kvm/coalesced_mmio.h +++ b/virt/kvm/coalesced_mmio.h @@ -10,6 +10,8 @@ * */ +#ifdef CONFIG_KVM_MMIO + #define KVM_COALESCED_MMIO_ZONE_MAX 100 struct kvm_coalesced_mmio_dev { @@ -21,9 +23,17 @@ struct kvm_coalesced_mmio_dev { }; int kvm_coalesced_mmio_init(struct kvm *kvm); +void kvm_coalesced_mmio_free(struct kvm *kvm); int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, struct kvm_coalesced_mmio_zone *zone); int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, struct kvm_coalesced_mmio_zone *zone); +#else + +static inline int kvm_coalesced_mmio_init(struct kvm *kvm) { return 0; } +static inline void kvm_coalesced_mmio_free(struct kvm *kvm) { } + +#endif + #endif diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index c10d117c4317..bc23b8e0609b 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -51,9 +51,7 @@ #include #include -#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET #include "coalesced_mmio.h" -#endif #define CREATE_TRACE_POINTS #include @@ -468,10 +466,7 @@ static void kvm_destroy_vm(struct kvm *kvm) kvm_free_irq_routing(kvm); kvm_io_bus_destroy(&kvm->pio_bus); kvm_io_bus_destroy(&kvm->mmio_bus); -#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET - if (kvm->coalesced_mmio_ring != NULL) - free_page((unsigned long)kvm->coalesced_mmio_ring); -#endif + kvm_coalesced_mmio_free(kvm); #if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER) mmu_notifier_unregister(&kvm->mmu_notifier, kvm->mm); #else -- cgit v1.2.3 From 46a26bf55714c1e2f17e34683292a389acb8e601 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:16 -0200 Subject: KVM: modify memslots layout in struct kvm Have a pointer to an allocated region inside struct kvm. [alex: fix ppc book 3s] Signed-off-by: Alexander Graf Signed-off-by: Marcelo Tosatti --- virt/kvm/iommu.c | 18 ++++++++++++------ virt/kvm/kvm_main.c | 36 +++++++++++++++++++++++------------- 2 files changed, 35 insertions(+), 19 deletions(-) (limited to 'virt') diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index 15147583abd1..bc697a66a883 100644 --- a/virt/kvm/iommu.c +++ b/virt/kvm/iommu.c @@ -76,10 +76,13 @@ unmap_pages: static int kvm_iommu_map_memslots(struct kvm *kvm) { int i, r = 0; + struct kvm_memslots *slots; - for (i = 0; i < kvm->nmemslots; i++) { - r = kvm_iommu_map_pages(kvm, kvm->memslots[i].base_gfn, - kvm->memslots[i].npages); + slots = kvm->memslots; + + for (i = 0; i < slots->nmemslots; i++) { + r = kvm_iommu_map_pages(kvm, slots->memslots[i].base_gfn, + slots->memslots[i].npages); if (r) break; } @@ -210,10 +213,13 @@ static void kvm_iommu_put_pages(struct kvm *kvm, static int kvm_iommu_unmap_memslots(struct kvm *kvm) { int i; + struct kvm_memslots *slots; + + slots = kvm->memslots; - for (i = 0; i < kvm->nmemslots; i++) { - kvm_iommu_put_pages(kvm, kvm->memslots[i].base_gfn, - kvm->memslots[i].npages); + for (i = 0; i < slots->nmemslots; i++) { + kvm_iommu_put_pages(kvm, slots->memslots[i].base_gfn, + slots->memslots[i].npages); } return 0; diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index bc23b8e0609b..86dd8f3d29c9 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -375,12 +375,16 @@ static struct kvm *kvm_create_vm(void) INIT_HLIST_HEAD(&kvm->irq_ack_notifier_list); #endif + r = -ENOMEM; + kvm->memslots = kzalloc(sizeof(struct kvm_memslots), GFP_KERNEL); + if (!kvm->memslots) + goto out_err; + #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET page = alloc_page(GFP_KERNEL | __GFP_ZERO); - if (!page) { - r = -ENOMEM; + if (!page) goto out_err; - } + kvm->coalesced_mmio_ring = (struct kvm_coalesced_mmio_ring *)page_address(page); #endif @@ -416,6 +420,7 @@ out: out_err: hardware_disable_all(); out_err_nodisable: + kfree(kvm->memslots); kfree(kvm); return ERR_PTR(r); } @@ -450,9 +455,12 @@ static void kvm_free_physmem_slot(struct kvm_memory_slot *free, void kvm_free_physmem(struct kvm *kvm) { int i; + struct kvm_memslots *slots = kvm->memslots; + + for (i = 0; i < slots->nmemslots; ++i) + kvm_free_physmem_slot(&slots->memslots[i], NULL); - for (i = 0; i < kvm->nmemslots; ++i) - kvm_free_physmem_slot(&kvm->memslots[i], NULL); + kfree(kvm->memslots); } static void kvm_destroy_vm(struct kvm *kvm) @@ -533,7 +541,7 @@ int __kvm_set_memory_region(struct kvm *kvm, if (mem->guest_phys_addr + mem->memory_size < mem->guest_phys_addr) goto out; - memslot = &kvm->memslots[mem->slot]; + memslot = &kvm->memslots->memslots[mem->slot]; base_gfn = mem->guest_phys_addr >> PAGE_SHIFT; npages = mem->memory_size >> PAGE_SHIFT; @@ -554,7 +562,7 @@ int __kvm_set_memory_region(struct kvm *kvm, /* Check for overlaps */ r = -EEXIST; for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { - struct kvm_memory_slot *s = &kvm->memslots[i]; + struct kvm_memory_slot *s = &kvm->memslots->memslots[i]; if (s == memslot || !s->npages) continue; @@ -656,8 +664,8 @@ skip_lpage: kvm_arch_flush_shadow(kvm); spin_lock(&kvm->mmu_lock); - if (mem->slot >= kvm->nmemslots) - kvm->nmemslots = mem->slot + 1; + if (mem->slot >= kvm->memslots->nmemslots) + kvm->memslots->nmemslots = mem->slot + 1; *memslot = new; spin_unlock(&kvm->mmu_lock); @@ -727,7 +735,7 @@ int kvm_get_dirty_log(struct kvm *kvm, if (log->slot >= KVM_MEMORY_SLOTS) goto out; - memslot = &kvm->memslots[log->slot]; + memslot = &kvm->memslots->memslots[log->slot]; r = -ENOENT; if (!memslot->dirty_bitmap) goto out; @@ -781,9 +789,10 @@ EXPORT_SYMBOL_GPL(kvm_is_error_hva); struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn) { int i; + struct kvm_memslots *slots = kvm->memslots; - for (i = 0; i < kvm->nmemslots; ++i) { - struct kvm_memory_slot *memslot = &kvm->memslots[i]; + for (i = 0; i < slots->nmemslots; ++i) { + struct kvm_memory_slot *memslot = &slots->memslots[i]; if (gfn >= memslot->base_gfn && gfn < memslot->base_gfn + memslot->npages) @@ -802,10 +811,11 @@ struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn) int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn) { int i; + struct kvm_memslots *slots = kvm->memslots; gfn = unalias_gfn(kvm, gfn); for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { - struct kvm_memory_slot *memslot = &kvm->memslots[i]; + struct kvm_memory_slot *memslot = &slots->memslots[i]; if (gfn >= memslot->base_gfn && gfn < memslot->base_gfn + memslot->npages) -- cgit v1.2.3 From f7784b8ec9b6a041fa828cfbe9012fe51933f5ac Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:18 -0200 Subject: KVM: split kvm_arch_set_memory_region into prepare and commit Required for SRCU convertion later. Signed-off-by: Marcelo Tosatti --- virt/kvm/kvm_main.c | 12 +++++------- 1 file changed, 5 insertions(+), 7 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 86dd8f3d29c9..c9f6cfe83120 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -663,6 +663,10 @@ skip_lpage: if (!npages) kvm_arch_flush_shadow(kvm); + r = kvm_arch_prepare_memory_region(kvm, &new, old, mem, user_alloc); + if (r) + goto out_free; + spin_lock(&kvm->mmu_lock); if (mem->slot >= kvm->memslots->nmemslots) kvm->memslots->nmemslots = mem->slot + 1; @@ -670,13 +674,7 @@ skip_lpage: *memslot = new; spin_unlock(&kvm->mmu_lock); - r = kvm_arch_set_memory_region(kvm, mem, old, user_alloc); - if (r) { - spin_lock(&kvm->mmu_lock); - *memslot = old; - spin_unlock(&kvm->mmu_lock); - goto out_free; - } + kvm_arch_commit_memory_region(kvm, mem, old, user_alloc); kvm_free_physmem_slot(&old, npages ? &new : NULL); /* Slot deletion case: we have to update the current slot */ -- cgit v1.2.3 From 506f0d6f9c40ae7d9634acf3c26358810f42c24a Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:19 -0200 Subject: KVM: introduce gfn_to_pfn_memslot Which takes a memslot pointer instead of using kvm->memslots. To be used by SRCU convertion later. Signed-off-by: Marcelo Tosatti --- virt/kvm/kvm_main.c | 33 +++++++++++++++++++++++++-------- 1 file changed, 25 insertions(+), 8 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index c9f6cfe83120..4e2321c733f7 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -835,21 +835,14 @@ unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn) } EXPORT_SYMBOL_GPL(gfn_to_hva); -pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn) +static pfn_t hva_to_pfn(struct kvm *kvm, unsigned long addr) { struct page *page[1]; - unsigned long addr; int npages; pfn_t pfn; might_sleep(); - addr = gfn_to_hva(kvm, gfn); - if (kvm_is_error_hva(addr)) { - get_page(bad_page); - return page_to_pfn(bad_page); - } - npages = get_user_pages_fast(addr, 1, 1, page); if (unlikely(npages != 1)) { @@ -874,8 +867,32 @@ pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn) return pfn; } +pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn) +{ + unsigned long addr; + + addr = gfn_to_hva(kvm, gfn); + if (kvm_is_error_hva(addr)) { + get_page(bad_page); + return page_to_pfn(bad_page); + } + + return hva_to_pfn(kvm, addr); +} EXPORT_SYMBOL_GPL(gfn_to_pfn); +static unsigned long gfn_to_hva_memslot(struct kvm_memory_slot *slot, gfn_t gfn) +{ + return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE); +} + +pfn_t gfn_to_pfn_memslot(struct kvm *kvm, + struct kvm_memory_slot *slot, gfn_t gfn) +{ + unsigned long addr = gfn_to_hva_memslot(slot, gfn); + return hva_to_pfn(kvm, addr); +} + struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn) { pfn_t pfn; -- cgit v1.2.3 From 3ad26d8139a82b0510b1e0435ee82ae461d33401 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:20 -0200 Subject: KVM: use gfn_to_pfn_memslot in kvm_iommu_map_pages So its possible to iommu map a memslot before making it visible to kvm. Signed-off-by: Marcelo Tosatti --- virt/kvm/iommu.c | 13 ++++++------- virt/kvm/kvm_main.c | 2 +- 2 files changed, 7 insertions(+), 8 deletions(-) (limited to 'virt') diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index bc697a66a883..cf567d8033db 100644 --- a/virt/kvm/iommu.c +++ b/virt/kvm/iommu.c @@ -32,10 +32,10 @@ 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) +int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot) { - gfn_t gfn = base_gfn; + gfn_t gfn = slot->base_gfn; + unsigned long npages = slot->npages; pfn_t pfn; int i, r = 0; struct iommu_domain *domain = kvm->arch.iommu_domain; @@ -54,7 +54,7 @@ int kvm_iommu_map_pages(struct kvm *kvm, if (iommu_iova_to_phys(domain, gfn_to_gpa(gfn))) continue; - pfn = gfn_to_pfn(kvm, gfn); + pfn = gfn_to_pfn_memslot(kvm, slot, gfn); r = iommu_map_range(domain, gfn_to_gpa(gfn), pfn_to_hpa(pfn), @@ -69,7 +69,7 @@ int kvm_iommu_map_pages(struct kvm *kvm, return 0; unmap_pages: - kvm_iommu_put_pages(kvm, base_gfn, i); + kvm_iommu_put_pages(kvm, slot->base_gfn, i); return r; } @@ -81,8 +81,7 @@ static int kvm_iommu_map_memslots(struct kvm *kvm) slots = kvm->memslots; for (i = 0; i < slots->nmemslots; i++) { - r = kvm_iommu_map_pages(kvm, slots->memslots[i].base_gfn, - slots->memslots[i].npages); + r = kvm_iommu_map_pages(kvm, &slots->memslots[i]); if (r) break; } diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 4e2321c733f7..87d296d8b270 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -684,7 +684,7 @@ skip_lpage: spin_unlock(&kvm->mmu_lock); #ifdef CONFIG_DMAR /* map the pages in iommu page table */ - r = kvm_iommu_map_pages(kvm, base_gfn, npages); + r = kvm_iommu_map_pages(kvm, memslot); if (r) goto out; #endif -- cgit v1.2.3 From bc6678a33d9b952981a8e44a4f876c3ad64ca4d8 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:21 -0200 Subject: KVM: introduce kvm->srcu and convert kvm_set_memory_region to SRCU update Use two steps for memslot deletion: mark the slot invalid (which stops instantiation of new shadow pages for that slot, but allows destruction), then instantiate the new empty slot. Also simplifies kvm_handle_hva locking. Signed-off-by: Marcelo Tosatti --- virt/kvm/assigned-dev.c | 8 +-- virt/kvm/iommu.c | 4 +- virt/kvm/kvm_main.c | 141 +++++++++++++++++++++++++++++++++++------------- 3 files changed, 111 insertions(+), 42 deletions(-) (limited to 'virt') diff --git a/virt/kvm/assigned-dev.c b/virt/kvm/assigned-dev.c index f73de631e3ee..f51e684dd238 100644 --- a/virt/kvm/assigned-dev.c +++ b/virt/kvm/assigned-dev.c @@ -504,12 +504,12 @@ out: static int kvm_vm_ioctl_assign_device(struct kvm *kvm, struct kvm_assigned_pci_dev *assigned_dev) { - int r = 0; + int r = 0, idx; struct kvm_assigned_dev_kernel *match; struct pci_dev *dev; mutex_lock(&kvm->lock); - down_read(&kvm->slots_lock); + idx = srcu_read_lock(&kvm->srcu); match = kvm_find_assigned_dev(&kvm->arch.assigned_dev_head, assigned_dev->assigned_dev_id); @@ -573,7 +573,7 @@ static int kvm_vm_ioctl_assign_device(struct kvm *kvm, } out: - up_read(&kvm->slots_lock); + srcu_read_unlock(&kvm->srcu, idx); mutex_unlock(&kvm->lock); return r; out_list_del: @@ -585,7 +585,7 @@ out_put: pci_dev_put(dev); out_free: kfree(match); - up_read(&kvm->slots_lock); + srcu_read_unlock(&kvm->srcu, idx); mutex_unlock(&kvm->lock); return r; } diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index cf567d8033db..65a51432c8e5 100644 --- a/virt/kvm/iommu.c +++ b/virt/kvm/iommu.c @@ -78,7 +78,7 @@ static int kvm_iommu_map_memslots(struct kvm *kvm) int i, r = 0; struct kvm_memslots *slots; - slots = kvm->memslots; + slots = rcu_dereference(kvm->memslots); for (i = 0; i < slots->nmemslots; i++) { r = kvm_iommu_map_pages(kvm, &slots->memslots[i]); @@ -214,7 +214,7 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm) int i; struct kvm_memslots *slots; - slots = kvm->memslots; + slots = rcu_dereference(kvm->memslots); for (i = 0; i < slots->nmemslots; i++) { kvm_iommu_put_pages(kvm, slots->memslots[i].base_gfn, diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 87d296d8b270..2bb24a814fdf 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -44,6 +44,7 @@ #include #include #include +#include #include #include @@ -213,7 +214,7 @@ static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn, unsigned long address) { struct kvm *kvm = mmu_notifier_to_kvm(mn); - int need_tlb_flush; + int need_tlb_flush, idx; /* * When ->invalidate_page runs, the linux pte has been zapped @@ -233,10 +234,12 @@ static void kvm_mmu_notifier_invalidate_page(struct mmu_notifier *mn, * pte after kvm_unmap_hva returned, without noticing the page * is going to be freed. */ + idx = srcu_read_lock(&kvm->srcu); spin_lock(&kvm->mmu_lock); kvm->mmu_notifier_seq++; need_tlb_flush = kvm_unmap_hva(kvm, address); spin_unlock(&kvm->mmu_lock); + srcu_read_unlock(&kvm->srcu, idx); /* we've to flush the tlb before the pages can be freed */ if (need_tlb_flush) @@ -250,11 +253,14 @@ static void kvm_mmu_notifier_change_pte(struct mmu_notifier *mn, pte_t pte) { struct kvm *kvm = mmu_notifier_to_kvm(mn); + int idx; + idx = srcu_read_lock(&kvm->srcu); spin_lock(&kvm->mmu_lock); kvm->mmu_notifier_seq++; kvm_set_spte_hva(kvm, address, pte); spin_unlock(&kvm->mmu_lock); + srcu_read_unlock(&kvm->srcu, idx); } static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn, @@ -263,8 +269,9 @@ static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn, unsigned long end) { struct kvm *kvm = mmu_notifier_to_kvm(mn); - int need_tlb_flush = 0; + int need_tlb_flush = 0, idx; + idx = srcu_read_lock(&kvm->srcu); spin_lock(&kvm->mmu_lock); /* * The count increase must become visible at unlock time as no @@ -275,6 +282,7 @@ static void kvm_mmu_notifier_invalidate_range_start(struct mmu_notifier *mn, for (; start < end; start += PAGE_SIZE) need_tlb_flush |= kvm_unmap_hva(kvm, start); spin_unlock(&kvm->mmu_lock); + srcu_read_unlock(&kvm->srcu, idx); /* we've to flush the tlb before the pages can be freed */ if (need_tlb_flush) @@ -312,11 +320,13 @@ static int kvm_mmu_notifier_clear_flush_young(struct mmu_notifier *mn, unsigned long address) { struct kvm *kvm = mmu_notifier_to_kvm(mn); - int young; + int young, idx; + idx = srcu_read_lock(&kvm->srcu); spin_lock(&kvm->mmu_lock); young = kvm_age_hva(kvm, address); spin_unlock(&kvm->mmu_lock); + srcu_read_unlock(&kvm->srcu, idx); if (young) kvm_flush_remote_tlbs(kvm); @@ -379,11 +389,15 @@ static struct kvm *kvm_create_vm(void) kvm->memslots = kzalloc(sizeof(struct kvm_memslots), GFP_KERNEL); if (!kvm->memslots) goto out_err; + if (init_srcu_struct(&kvm->srcu)) + goto out_err; #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET page = alloc_page(GFP_KERNEL | __GFP_ZERO); - if (!page) + if (!page) { + cleanup_srcu_struct(&kvm->srcu); goto out_err; + } kvm->coalesced_mmio_ring = (struct kvm_coalesced_mmio_ring *)page_address(page); @@ -391,6 +405,7 @@ static struct kvm *kvm_create_vm(void) r = kvm_init_mmu_notifier(kvm); if (r) { + cleanup_srcu_struct(&kvm->srcu); #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET put_page(page); #endif @@ -480,6 +495,7 @@ static void kvm_destroy_vm(struct kvm *kvm) #else kvm_arch_flush_shadow(kvm); #endif + cleanup_srcu_struct(&kvm->srcu); kvm_arch_destroy_vm(kvm); hardware_disable_all(); mmdrop(mm); @@ -521,12 +537,13 @@ int __kvm_set_memory_region(struct kvm *kvm, struct kvm_userspace_memory_region *mem, int user_alloc) { - int r; + int r, flush_shadow = 0; gfn_t base_gfn; unsigned long npages; unsigned long i; struct kvm_memory_slot *memslot; struct kvm_memory_slot old, new; + struct kvm_memslots *slots, *old_memslots; r = -EINVAL; /* General sanity checks */ @@ -588,15 +605,7 @@ int __kvm_set_memory_region(struct kvm *kvm, memset(new.rmap, 0, npages * sizeof(*new.rmap)); new.user_alloc = user_alloc; - /* - * hva_to_rmmap() serialzies with the mmu_lock and to be - * safe it has to ignore memslots with !user_alloc && - * !userspace_addr. - */ - if (user_alloc) - new.userspace_addr = mem->userspace_addr; - else - new.userspace_addr = 0; + new.userspace_addr = mem->userspace_addr; } if (!npages) goto skip_lpage; @@ -651,8 +660,9 @@ skip_lpage: if (!new.dirty_bitmap) goto out_free; memset(new.dirty_bitmap, 0, dirty_bytes); + /* destroy any largepage mappings for dirty tracking */ if (old.npages) - kvm_arch_flush_shadow(kvm); + flush_shadow = 1; } #else /* not defined CONFIG_S390 */ new.user_alloc = user_alloc; @@ -660,34 +670,72 @@ skip_lpage: new.userspace_addr = mem->userspace_addr; #endif /* not defined CONFIG_S390 */ - if (!npages) + if (!npages) { + r = -ENOMEM; + slots = kzalloc(sizeof(struct kvm_memslots), GFP_KERNEL); + if (!slots) + goto out_free; + memcpy(slots, kvm->memslots, sizeof(struct kvm_memslots)); + if (mem->slot >= slots->nmemslots) + slots->nmemslots = mem->slot + 1; + slots->memslots[mem->slot].flags |= KVM_MEMSLOT_INVALID; + + old_memslots = kvm->memslots; + rcu_assign_pointer(kvm->memslots, slots); + synchronize_srcu_expedited(&kvm->srcu); + /* From this point no new shadow pages pointing to a deleted + * memslot will be created. + * + * validation of sp->gfn happens in: + * - gfn_to_hva (kvm_read_guest, gfn_to_pfn) + * - kvm_is_visible_gfn (mmu_check_roots) + */ kvm_arch_flush_shadow(kvm); + kfree(old_memslots); + } r = kvm_arch_prepare_memory_region(kvm, &new, old, mem, user_alloc); if (r) goto out_free; - spin_lock(&kvm->mmu_lock); - if (mem->slot >= kvm->memslots->nmemslots) - kvm->memslots->nmemslots = mem->slot + 1; +#ifdef CONFIG_DMAR + /* map the pages in iommu page table */ + if (npages) { + r = kvm_iommu_map_pages(kvm, &new); + if (r) + goto out_free; + } +#endif - *memslot = new; - spin_unlock(&kvm->mmu_lock); + r = -ENOMEM; + slots = kzalloc(sizeof(struct kvm_memslots), GFP_KERNEL); + if (!slots) + goto out_free; + memcpy(slots, kvm->memslots, sizeof(struct kvm_memslots)); + if (mem->slot >= slots->nmemslots) + slots->nmemslots = mem->slot + 1; + + /* actual memory is freed via old in kvm_free_physmem_slot below */ + if (!npages) { + new.rmap = NULL; + new.dirty_bitmap = NULL; + for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) + new.lpage_info[i] = NULL; + } + + slots->memslots[mem->slot] = new; + old_memslots = kvm->memslots; + rcu_assign_pointer(kvm->memslots, slots); + synchronize_srcu_expedited(&kvm->srcu); kvm_arch_commit_memory_region(kvm, mem, old, user_alloc); - kvm_free_physmem_slot(&old, npages ? &new : NULL); - /* Slot deletion case: we have to update the current slot */ - spin_lock(&kvm->mmu_lock); - if (!npages) - *memslot = old; - spin_unlock(&kvm->mmu_lock); -#ifdef CONFIG_DMAR - /* map the pages in iommu page table */ - r = kvm_iommu_map_pages(kvm, memslot); - if (r) - goto out; -#endif + kvm_free_physmem_slot(&old, &new); + kfree(old_memslots); + + if (flush_shadow) + kvm_arch_flush_shadow(kvm); + return 0; out_free: @@ -787,7 +835,7 @@ EXPORT_SYMBOL_GPL(kvm_is_error_hva); struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn) { int i; - struct kvm_memslots *slots = kvm->memslots; + struct kvm_memslots *slots = rcu_dereference(kvm->memslots); for (i = 0; i < slots->nmemslots; ++i) { struct kvm_memory_slot *memslot = &slots->memslots[i]; @@ -809,12 +857,15 @@ struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn) int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn) { int i; - struct kvm_memslots *slots = kvm->memslots; + struct kvm_memslots *slots = rcu_dereference(kvm->memslots); gfn = unalias_gfn(kvm, gfn); for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { struct kvm_memory_slot *memslot = &slots->memslots[i]; + if (memslot->flags & KVM_MEMSLOT_INVALID) + continue; + if (gfn >= memslot->base_gfn && gfn < memslot->base_gfn + memslot->npages) return 1; @@ -823,13 +874,31 @@ int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn) } EXPORT_SYMBOL_GPL(kvm_is_visible_gfn); +int memslot_id(struct kvm *kvm, gfn_t gfn) +{ + int i; + struct kvm_memslots *slots = rcu_dereference(kvm->memslots); + struct kvm_memory_slot *memslot = NULL; + + gfn = unalias_gfn(kvm, gfn); + for (i = 0; i < slots->nmemslots; ++i) { + memslot = &slots->memslots[i]; + + if (gfn >= memslot->base_gfn + && gfn < memslot->base_gfn + memslot->npages) + break; + } + + return memslot - slots->memslots; +} + unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn) { struct kvm_memory_slot *slot; gfn = unalias_gfn(kvm, gfn); slot = gfn_to_memslot_unaliased(kvm, gfn); - if (!slot) + if (!slot || slot->flags & KVM_MEMSLOT_INVALID) return bad_hva(); return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE); } -- cgit v1.2.3 From a983fb238728e1123177e8058d4f644b949a7d05 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:23 -0200 Subject: KVM: x86: switch kvm_set_memory_alias to SRCU update Using a similar two-step procedure as for memslots. Signed-off-by: Marcelo Tosatti --- virt/kvm/kvm_main.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 2bb24a814fdf..c680f7b64c6f 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -859,7 +859,7 @@ int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn) int i; struct kvm_memslots *slots = rcu_dereference(kvm->memslots); - gfn = unalias_gfn(kvm, gfn); + gfn = unalias_gfn_instantiation(kvm, gfn); for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { struct kvm_memory_slot *memslot = &slots->memslots[i]; @@ -896,7 +896,7 @@ unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn) { struct kvm_memory_slot *slot; - gfn = unalias_gfn(kvm, gfn); + gfn = unalias_gfn_instantiation(kvm, gfn); slot = gfn_to_memslot_unaliased(kvm, gfn); if (!slot || slot->flags & KVM_MEMSLOT_INVALID) return bad_hva(); -- cgit v1.2.3 From e93f8a0f821e290ac5149830110a5f704db7a1fc Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:24 -0200 Subject: KVM: convert io_bus to SRCU Signed-off-by: Marcelo Tosatti --- virt/kvm/coalesced_mmio.c | 4 +- virt/kvm/eventfd.c | 8 ++-- virt/kvm/ioapic.c | 4 +- virt/kvm/kvm_main.c | 106 +++++++++++++++++++++++++++------------------- 4 files changed, 73 insertions(+), 49 deletions(-) (limited to 'virt') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index d68e6c68e0ff..a736a93ca7b7 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -110,7 +110,9 @@ int kvm_coalesced_mmio_init(struct kvm *kvm) dev->kvm = kvm; kvm->coalesced_mmio_dev = dev; - ret = kvm_io_bus_register_dev(kvm, &kvm->mmio_bus, &dev->dev); + down_write(&kvm->slots_lock); + ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, &dev->dev); + up_write(&kvm->slots_lock); if (ret < 0) goto out_free_dev; diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c index a9d3fc6c681c..315a586ec4d5 100644 --- a/virt/kvm/eventfd.c +++ b/virt/kvm/eventfd.c @@ -463,7 +463,7 @@ static int kvm_assign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) { int pio = args->flags & KVM_IOEVENTFD_FLAG_PIO; - struct kvm_io_bus *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus; + enum kvm_bus bus_idx = pio ? KVM_PIO_BUS : KVM_MMIO_BUS; struct _ioeventfd *p; struct eventfd_ctx *eventfd; int ret; @@ -518,7 +518,7 @@ kvm_assign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) kvm_iodevice_init(&p->dev, &ioeventfd_ops); - ret = __kvm_io_bus_register_dev(bus, &p->dev); + ret = kvm_io_bus_register_dev(kvm, bus_idx, &p->dev); if (ret < 0) goto unlock_fail; @@ -542,7 +542,7 @@ static int kvm_deassign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) { int pio = args->flags & KVM_IOEVENTFD_FLAG_PIO; - struct kvm_io_bus *bus = pio ? &kvm->pio_bus : &kvm->mmio_bus; + enum kvm_bus bus_idx = pio ? KVM_PIO_BUS : KVM_MMIO_BUS; struct _ioeventfd *p, *tmp; struct eventfd_ctx *eventfd; int ret = -ENOENT; @@ -565,7 +565,7 @@ kvm_deassign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) if (!p->wildcard && p->datamatch != args->datamatch) continue; - __kvm_io_bus_unregister_dev(bus, &p->dev); + kvm_io_bus_unregister_dev(kvm, bus_idx, &p->dev); ioeventfd_release(p); ret = 0; break; diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index 38a2d20b89de..f326a6f301cc 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -372,7 +372,9 @@ int kvm_ioapic_init(struct kvm *kvm) kvm_ioapic_reset(ioapic); kvm_iodevice_init(&ioapic->dev, &ioapic_mmio_ops); ioapic->kvm = kvm; - ret = kvm_io_bus_register_dev(kvm, &kvm->mmio_bus, &ioapic->dev); + down_write(&kvm->slots_lock); + ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, &ioapic->dev); + up_write(&kvm->slots_lock); if (ret < 0) kfree(ioapic); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index c680f7b64c6f..659bc12ad16a 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -85,6 +85,8 @@ static long kvm_vcpu_ioctl(struct file *file, unsigned int ioctl, static int hardware_enable_all(void); static void hardware_disable_all(void); +static void kvm_io_bus_destroy(struct kvm_io_bus *bus); + static bool kvm_rebooting; static bool largepages_enabled = true; @@ -367,7 +369,7 @@ static int kvm_init_mmu_notifier(struct kvm *kvm) static struct kvm *kvm_create_vm(void) { - int r = 0; + int r = 0, i; struct kvm *kvm = kvm_arch_create_vm(); #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET struct page *page; @@ -391,6 +393,14 @@ static struct kvm *kvm_create_vm(void) goto out_err; if (init_srcu_struct(&kvm->srcu)) goto out_err; + for (i = 0; i < KVM_NR_BUSES; i++) { + kvm->buses[i] = kzalloc(sizeof(struct kvm_io_bus), + GFP_KERNEL); + if (!kvm->buses[i]) { + cleanup_srcu_struct(&kvm->srcu); + goto out_err; + } + } #ifdef KVM_COALESCED_MMIO_PAGE_OFFSET page = alloc_page(GFP_KERNEL | __GFP_ZERO); @@ -416,11 +426,9 @@ static struct kvm *kvm_create_vm(void) atomic_inc(&kvm->mm->mm_count); spin_lock_init(&kvm->mmu_lock); spin_lock_init(&kvm->requests_lock); - kvm_io_bus_init(&kvm->pio_bus); kvm_eventfd_init(kvm); mutex_init(&kvm->lock); mutex_init(&kvm->irq_lock); - kvm_io_bus_init(&kvm->mmio_bus); init_rwsem(&kvm->slots_lock); atomic_set(&kvm->users_count, 1); spin_lock(&kvm_lock); @@ -435,6 +443,8 @@ out: out_err: hardware_disable_all(); out_err_nodisable: + for (i = 0; i < KVM_NR_BUSES; i++) + kfree(kvm->buses[i]); kfree(kvm->memslots); kfree(kvm); return ERR_PTR(r); @@ -480,6 +490,7 @@ void kvm_free_physmem(struct kvm *kvm) static void kvm_destroy_vm(struct kvm *kvm) { + int i; struct mm_struct *mm = kvm->mm; kvm_arch_sync_events(kvm); @@ -487,8 +498,8 @@ static void kvm_destroy_vm(struct kvm *kvm) list_del(&kvm->vm_list); spin_unlock(&kvm_lock); kvm_free_irq_routing(kvm); - kvm_io_bus_destroy(&kvm->pio_bus); - kvm_io_bus_destroy(&kvm->mmio_bus); + for (i = 0; i < KVM_NR_BUSES; i++) + kvm_io_bus_destroy(kvm->buses[i]); kvm_coalesced_mmio_free(kvm); #if defined(CONFIG_MMU_NOTIFIER) && defined(KVM_ARCH_WANT_MMU_NOTIFIER) mmu_notifier_unregister(&kvm->mmu_notifier, kvm->mm); @@ -1949,12 +1960,7 @@ static struct notifier_block kvm_reboot_notifier = { .priority = 0, }; -void kvm_io_bus_init(struct kvm_io_bus *bus) -{ - memset(bus, 0, sizeof(*bus)); -} - -void kvm_io_bus_destroy(struct kvm_io_bus *bus) +static void kvm_io_bus_destroy(struct kvm_io_bus *bus) { int i; @@ -1963,13 +1969,15 @@ void kvm_io_bus_destroy(struct kvm_io_bus *bus) kvm_iodevice_destructor(pos); } + kfree(bus); } /* kvm_io_bus_write - called under kvm->slots_lock */ -int kvm_io_bus_write(struct kvm_io_bus *bus, gpa_t addr, +int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len, const void *val) { int i; + struct kvm_io_bus *bus = rcu_dereference(kvm->buses[bus_idx]); for (i = 0; i < bus->dev_count; i++) if (!kvm_iodevice_write(bus->devs[i], addr, len, val)) return 0; @@ -1977,59 +1985,71 @@ int kvm_io_bus_write(struct kvm_io_bus *bus, gpa_t addr, } /* kvm_io_bus_read - called under kvm->slots_lock */ -int kvm_io_bus_read(struct kvm_io_bus *bus, gpa_t addr, int len, void *val) +int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, + int len, void *val) { int i; + struct kvm_io_bus *bus = rcu_dereference(kvm->buses[bus_idx]); + for (i = 0; i < bus->dev_count; i++) if (!kvm_iodevice_read(bus->devs[i], addr, len, val)) return 0; return -EOPNOTSUPP; } -int kvm_io_bus_register_dev(struct kvm *kvm, struct kvm_io_bus *bus, - struct kvm_io_device *dev) +/* Caller must have write lock on slots_lock. */ +int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, + struct kvm_io_device *dev) { - int ret; - - down_write(&kvm->slots_lock); - ret = __kvm_io_bus_register_dev(bus, dev); - up_write(&kvm->slots_lock); + struct kvm_io_bus *new_bus, *bus; - return ret; -} - -/* An unlocked version. Caller must have write lock on slots_lock. */ -int __kvm_io_bus_register_dev(struct kvm_io_bus *bus, - struct kvm_io_device *dev) -{ + bus = kvm->buses[bus_idx]; if (bus->dev_count > NR_IOBUS_DEVS-1) return -ENOSPC; - bus->devs[bus->dev_count++] = dev; + new_bus = kzalloc(sizeof(struct kvm_io_bus), GFP_KERNEL); + if (!new_bus) + return -ENOMEM; + memcpy(new_bus, bus, sizeof(struct kvm_io_bus)); + new_bus->devs[new_bus->dev_count++] = dev; + rcu_assign_pointer(kvm->buses[bus_idx], new_bus); + synchronize_srcu_expedited(&kvm->srcu); + kfree(bus); return 0; } -void kvm_io_bus_unregister_dev(struct kvm *kvm, - struct kvm_io_bus *bus, - struct kvm_io_device *dev) +/* Caller must have write lock on slots_lock. */ +int kvm_io_bus_unregister_dev(struct kvm *kvm, enum kvm_bus bus_idx, + struct kvm_io_device *dev) { - down_write(&kvm->slots_lock); - __kvm_io_bus_unregister_dev(bus, dev); - up_write(&kvm->slots_lock); -} + int i, r; + struct kvm_io_bus *new_bus, *bus; -/* An unlocked version. Caller must have write lock on slots_lock. */ -void __kvm_io_bus_unregister_dev(struct kvm_io_bus *bus, - struct kvm_io_device *dev) -{ - int i; + new_bus = kzalloc(sizeof(struct kvm_io_bus), GFP_KERNEL); + if (!new_bus) + return -ENOMEM; - for (i = 0; i < bus->dev_count; i++) - if (bus->devs[i] == dev) { - bus->devs[i] = bus->devs[--bus->dev_count]; + bus = kvm->buses[bus_idx]; + memcpy(new_bus, bus, sizeof(struct kvm_io_bus)); + + r = -ENOENT; + for (i = 0; i < new_bus->dev_count; i++) + if (new_bus->devs[i] == dev) { + r = 0; + new_bus->devs[i] = new_bus->devs[--new_bus->dev_count]; break; } + + if (r) { + kfree(new_bus); + return r; + } + + rcu_assign_pointer(kvm->buses[bus_idx], new_bus); + synchronize_srcu_expedited(&kvm->srcu); + kfree(bus); + return r; } static struct notifier_block kvm_cpu_notifier = { -- cgit v1.2.3 From 79fac95ecfa3969aab8119d37ccd7226165f933a Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 23 Dec 2009 14:35:26 -0200 Subject: KVM: convert slots_lock to a mutex Signed-off-by: Marcelo Tosatti --- virt/kvm/coalesced_mmio.c | 14 +++++++------- virt/kvm/eventfd.c | 10 +++++----- virt/kvm/ioapic.c | 4 ++-- virt/kvm/kvm_main.c | 10 +++++----- 4 files changed, 19 insertions(+), 19 deletions(-) (limited to 'virt') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index a736a93ca7b7..5de6594260cb 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -110,9 +110,9 @@ int kvm_coalesced_mmio_init(struct kvm *kvm) dev->kvm = kvm; kvm->coalesced_mmio_dev = dev; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, &dev->dev); - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); if (ret < 0) goto out_free_dev; @@ -140,16 +140,16 @@ int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, if (dev == NULL) return -EINVAL; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); if (dev->nb_zones >= KVM_COALESCED_MMIO_ZONE_MAX) { - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return -ENOBUFS; } dev->zone[dev->nb_zones] = *zone; dev->nb_zones++; - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return 0; } @@ -163,7 +163,7 @@ int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, if (dev == NULL) return -EINVAL; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); i = dev->nb_zones; while(i) { @@ -181,7 +181,7 @@ int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, i--; } - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return 0; } diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c index 315a586ec4d5..486c604365d9 100644 --- a/virt/kvm/eventfd.c +++ b/virt/kvm/eventfd.c @@ -508,7 +508,7 @@ kvm_assign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) else p->wildcard = true; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); /* Verify that there isnt a match already */ if (ioeventfd_check_collision(kvm, p)) { @@ -524,12 +524,12 @@ kvm_assign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) list_add_tail(&p->list, &kvm->ioeventfds); - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return 0; unlock_fail: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); fail: kfree(p); @@ -551,7 +551,7 @@ kvm_deassign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) if (IS_ERR(eventfd)) return PTR_ERR(eventfd); - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); list_for_each_entry_safe(p, tmp, &kvm->ioeventfds, list) { bool wildcard = !(args->flags & KVM_IOEVENTFD_FLAG_DATAMATCH); @@ -571,7 +571,7 @@ kvm_deassign_ioeventfd(struct kvm *kvm, struct kvm_ioeventfd *args) break; } - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); eventfd_ctx_put(eventfd); diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index f326a6f301cc..f01392f51e86 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -372,9 +372,9 @@ int kvm_ioapic_init(struct kvm *kvm) kvm_ioapic_reset(ioapic); kvm_iodevice_init(&ioapic->dev, &ioapic_mmio_ops); ioapic->kvm = kvm; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, &ioapic->dev); - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); if (ret < 0) kfree(ioapic); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 659bc12ad16a..2b7cd6c0d9ca 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -429,7 +429,7 @@ static struct kvm *kvm_create_vm(void) kvm_eventfd_init(kvm); mutex_init(&kvm->lock); mutex_init(&kvm->irq_lock); - init_rwsem(&kvm->slots_lock); + mutex_init(&kvm->slots_lock); atomic_set(&kvm->users_count, 1); spin_lock(&kvm_lock); list_add(&kvm->vm_list, &vm_list); @@ -763,9 +763,9 @@ int kvm_set_memory_region(struct kvm *kvm, { int r; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); r = __kvm_set_memory_region(kvm, mem, user_alloc); - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return r; } EXPORT_SYMBOL_GPL(kvm_set_memory_region); @@ -1997,7 +1997,7 @@ int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, return -EOPNOTSUPP; } -/* Caller must have write lock on slots_lock. */ +/* Caller must hold slots_lock. */ int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, struct kvm_io_device *dev) { @@ -2019,7 +2019,7 @@ int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, return 0; } -/* Caller must have write lock on slots_lock. */ +/* Caller must hold slots_lock. */ int kvm_io_bus_unregister_dev(struct kvm *kvm, enum kvm_bus bus_idx, struct kvm_io_device *dev) { -- cgit v1.2.3 From 46a929bc15fcd48e1e0e770a44040a6949cae133 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 28 Dec 2009 14:08:30 +0200 Subject: KVM: avoid taking ioapic mutex for non-ioapic EOIs When the guest acknowledges an interrupt, it sends an EOI message to the local apic, which broadcasts it to the ioapic. To handle the EOI, we need to take the ioapic mutex. On large guests, this causes a lot of contention on this mutex. Since large guests usually don't route interrupts via the ioapic (they use msi instead), this is completely unnecessary. Avoid taking the mutex by introducing a handled_vectors bitmap. Before taking the mutex, check if the ioapic was actually responsible for the acked vector. If not, we can return early. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- virt/kvm/ioapic.c | 19 +++++++++++++++++++ virt/kvm/ioapic.h | 1 + 2 files changed, 20 insertions(+) (limited to 'virt') diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index f01392f51e86..a2edfd177faf 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -100,6 +100,19 @@ static int ioapic_service(struct kvm_ioapic *ioapic, unsigned int idx) return injected; } +static void update_handled_vectors(struct kvm_ioapic *ioapic) +{ + DECLARE_BITMAP(handled_vectors, 256); + int i; + + memset(handled_vectors, 0, sizeof(handled_vectors)); + for (i = 0; i < IOAPIC_NUM_PINS; ++i) + __set_bit(ioapic->redirtbl[i].fields.vector, handled_vectors); + memcpy(ioapic->handled_vectors, handled_vectors, + sizeof(handled_vectors)); + smp_wmb(); +} + static void ioapic_write_indirect(struct kvm_ioapic *ioapic, u32 val) { unsigned index; @@ -134,6 +147,7 @@ static void ioapic_write_indirect(struct kvm_ioapic *ioapic, u32 val) e->bits |= (u32) val; e->fields.remote_irr = 0; } + update_handled_vectors(ioapic); mask_after = e->fields.mask; if (mask_before != mask_after) kvm_fire_mask_notifiers(ioapic->kvm, index, mask_after); @@ -241,6 +255,9 @@ void kvm_ioapic_update_eoi(struct kvm *kvm, int vector, int trigger_mode) { struct kvm_ioapic *ioapic = kvm->arch.vioapic; + smp_rmb(); + if (!test_bit(vector, ioapic->handled_vectors)) + return; mutex_lock(&ioapic->lock); __kvm_ioapic_update_eoi(ioapic, vector, trigger_mode); mutex_unlock(&ioapic->lock); @@ -352,6 +369,7 @@ void kvm_ioapic_reset(struct kvm_ioapic *ioapic) ioapic->ioregsel = 0; ioapic->irr = 0; ioapic->id = 0; + update_handled_vectors(ioapic); } static const struct kvm_io_device_ops ioapic_mmio_ops = { @@ -401,6 +419,7 @@ int kvm_set_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state) mutex_lock(&ioapic->lock); memcpy(ioapic, state, sizeof(struct kvm_ioapic_state)); + update_handled_vectors(ioapic); mutex_unlock(&ioapic->lock); return 0; } diff --git a/virt/kvm/ioapic.h b/virt/kvm/ioapic.h index 419c43b667ab..a505ce9054f3 100644 --- a/virt/kvm/ioapic.h +++ b/virt/kvm/ioapic.h @@ -46,6 +46,7 @@ struct kvm_ioapic { struct kvm *kvm; void (*ack_notifier)(void *opaque, int irq); struct mutex lock; + DECLARE_BITMAP(handled_vectors, 256); }; #ifdef DEBUG -- cgit v1.2.3 From 647492047763c3ee8fe51ecf9a04f39040aa495b Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Tue, 19 Jan 2010 12:45:23 -0200 Subject: KVM: fix cleanup_srcu_struct on vm destruction cleanup_srcu_struct on VM destruction remains broken: BUG: unable to handle kernel paging request at ffffffffffffffff IP: [] srcu_read_lock+0x16/0x21 RIP: 0010:[] [] srcu_read_lock+0x16/0x21 Call Trace: [] kvm_arch_vcpu_uninit+0x1b/0x48 [kvm] [] kvm_vcpu_uninit+0x9/0x15 [kvm] [] vmx_free_vcpu+0x7f/0x8f [kvm_intel] [] kvm_arch_destroy_vm+0x78/0x111 [kvm] [] kvm_put_kvm+0xd4/0xfe [kvm] Move it to kvm_arch_destroy_vm. Signed-off-by: Marcelo Tosatti Reported-by: Jan Kiszka --- virt/kvm/kvm_main.c | 1 - 1 file changed, 1 deletion(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 2b7cd6c0d9ca..7c5c873812ff 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -506,7 +506,6 @@ static void kvm_destroy_vm(struct kvm *kvm) #else kvm_arch_flush_shadow(kvm); #endif - cleanup_srcu_struct(&kvm->srcu); kvm_arch_destroy_vm(kvm); hardware_disable_all(); mmdrop(mm); -- cgit v1.2.3 From f0f4b930900ffa8daddb5262522c3e5c67ee1835 Mon Sep 17 00:00:00 2001 From: Sheng Yang Date: Thu, 21 Jan 2010 16:20:04 +0800 Subject: KVM: Fix kvm_coalesced_mmio_ring duplicate allocation The commit 0953ca73 "KVM: Simplify coalesced mmio initialization" allocate kvm_coalesced_mmio_ring in the kvm_coalesced_mmio_init(), but didn't discard the original allocation... Signed-off-by: Sheng Yang Signed-off-by: Marcelo Tosatti --- virt/kvm/kvm_main.c | 17 ----------------- 1 file changed, 17 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 7c5c873812ff..2b0974a14835 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -371,9 +371,6 @@ static struct kvm *kvm_create_vm(void) { int r = 0, i; struct kvm *kvm = kvm_arch_create_vm(); -#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET - struct page *page; -#endif if (IS_ERR(kvm)) goto out; @@ -402,23 +399,9 @@ static struct kvm *kvm_create_vm(void) } } -#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET - page = alloc_page(GFP_KERNEL | __GFP_ZERO); - if (!page) { - cleanup_srcu_struct(&kvm->srcu); - goto out_err; - } - - kvm->coalesced_mmio_ring = - (struct kvm_coalesced_mmio_ring *)page_address(page); -#endif - r = kvm_init_mmu_notifier(kvm); if (r) { cleanup_srcu_struct(&kvm->srcu); -#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET - put_page(page); -#endif goto out_err; } -- cgit v1.2.3 From ab9f4ecbb6d39a18e300a0d10a4968c37404aa76 Mon Sep 17 00:00:00 2001 From: "Zhai, Edwin" Date: Fri, 29 Jan 2010 14:38:44 +0800 Subject: KVM: enable PCI multiple-segments for pass-through device Enable optional parameter (default 0) - PCI segment (or domain) besides BDF, when assigning PCI device to guest. Signed-off-by: Zhai Edwin Acked-by: Chris Wright Signed-off-by: Marcelo Tosatti --- virt/kvm/assigned-dev.c | 4 +++- virt/kvm/iommu.c | 9 ++++++--- 2 files changed, 9 insertions(+), 4 deletions(-) (limited to 'virt') diff --git a/virt/kvm/assigned-dev.c b/virt/kvm/assigned-dev.c index f51e684dd238..057e2cca6af5 100644 --- a/virt/kvm/assigned-dev.c +++ b/virt/kvm/assigned-dev.c @@ -526,7 +526,8 @@ static int kvm_vm_ioctl_assign_device(struct kvm *kvm, r = -ENOMEM; goto out; } - dev = pci_get_bus_and_slot(assigned_dev->busnr, + dev = pci_get_domain_bus_and_slot(assigned_dev->segnr, + assigned_dev->busnr, assigned_dev->devfn); if (!dev) { printk(KERN_INFO "%s: host device not found\n", __func__); @@ -548,6 +549,7 @@ static int kvm_vm_ioctl_assign_device(struct kvm *kvm, pci_reset_function(dev); match->assigned_dev_id = assigned_dev->assigned_dev_id; + match->host_segnr = assigned_dev->segnr; match->host_busnr = assigned_dev->busnr; match->host_devfn = assigned_dev->devfn; match->flags = assigned_dev->flags; diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index 65a51432c8e5..80fd3ad3b2de 100644 --- a/virt/kvm/iommu.c +++ b/virt/kvm/iommu.c @@ -106,7 +106,8 @@ int kvm_assign_device(struct kvm *kvm, r = iommu_attach_device(domain, &pdev->dev); if (r) { - printk(KERN_ERR "assign device %x:%x.%x failed", + printk(KERN_ERR "assign device %x:%x:%x.%x failed", + pci_domain_nr(pdev->bus), pdev->bus->number, PCI_SLOT(pdev->devfn), PCI_FUNC(pdev->devfn)); @@ -127,7 +128,8 @@ int kvm_assign_device(struct kvm *kvm, goto out_unmap; } - printk(KERN_DEBUG "assign device: host bdf = %x:%x:%x\n", + printk(KERN_DEBUG "assign device %x:%x:%x.%x\n", + assigned_dev->host_segnr, assigned_dev->host_busnr, PCI_SLOT(assigned_dev->host_devfn), PCI_FUNC(assigned_dev->host_devfn)); @@ -154,7 +156,8 @@ int kvm_deassign_device(struct kvm *kvm, iommu_detach_device(domain, &pdev->dev); - printk(KERN_DEBUG "deassign device: host bdf = %x:%x:%x\n", + printk(KERN_DEBUG "deassign device %x:%x:%x.%x\n", + assigned_dev->host_segnr, assigned_dev->host_busnr, PCI_SLOT(assigned_dev->host_devfn), PCI_FUNC(assigned_dev->host_devfn)); -- cgit v1.2.3 From 8f0b1ab6fb045a1324d9435ba00c2940783b0041 Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Thu, 28 Jan 2010 12:37:56 +0100 Subject: KVM: Introduce kvm_host_page_size This patch introduces a generic function to find out the host page size for a given gfn. This function is needed by the kvm iommu code. This patch also simplifies the x86 host_mapping_level function. Signed-off-by: Joerg Roedel Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 25 +++++++++++++++++++++++++ 1 file changed, 25 insertions(+) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 2b0974a14835..0a360c26cc34 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -45,6 +45,7 @@ #include #include #include +#include #include #include @@ -867,6 +868,30 @@ int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn) } EXPORT_SYMBOL_GPL(kvm_is_visible_gfn); +unsigned long kvm_host_page_size(struct kvm *kvm, gfn_t gfn) +{ + struct vm_area_struct *vma; + unsigned long addr, size; + + size = PAGE_SIZE; + + addr = gfn_to_hva(kvm, gfn); + if (kvm_is_error_hva(addr)) + return PAGE_SIZE; + + down_read(¤t->mm->mmap_sem); + vma = find_vma(current->mm, addr); + if (!vma) + goto out; + + size = vma_kernel_pagesize(vma); + +out: + up_read(¤t->mm->mmap_sem); + + return size; +} + int memslot_id(struct kvm *kvm, gfn_t gfn) { int i; -- cgit v1.2.3 From 43db669730743fcc2a74b641ff4d77781eb15bfa Mon Sep 17 00:00:00 2001 From: Jochen Maes Date: Mon, 8 Feb 2010 11:29:33 +0100 Subject: KVM: Fix Codestyle in virt/kvm/coalesced_mmio.c Fixed 2 codestyle issues in virt/kvm/coalesced_mmio.c Signed-off-by: Jochen Maes Signed-off-by: Avi Kivity --- virt/kvm/coalesced_mmio.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'virt') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index 5de6594260cb..5169736377a3 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -133,7 +133,7 @@ void kvm_coalesced_mmio_free(struct kvm *kvm) } int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, - struct kvm_coalesced_mmio_zone *zone) + struct kvm_coalesced_mmio_zone *zone) { struct kvm_coalesced_mmio_dev *dev = kvm->coalesced_mmio_dev; @@ -166,7 +166,7 @@ int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, mutex_lock(&kvm->slots_lock); i = dev->nb_zones; - while(i) { + while (i) { z = &dev->zone[i - 1]; /* unregister all zones -- cgit v1.2.3 From 1ae77badc2504d157800f10a81a58cc9c941e7cb Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 9 Feb 2010 10:31:09 +0800 Subject: KVM: kvm->arch.vioapic should be NULL if kvm_ioapic_init() failure kvm->arch.vioapic should be NULL in case of kvm_ioapic_init() failure due to cannot register io dev. Signed-off-by: Wei Yongjun Signed-off-by: Avi Kivity --- virt/kvm/ioapic.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'virt') diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index a2edfd177faf..f3d06934ae6d 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -393,8 +393,10 @@ int kvm_ioapic_init(struct kvm *kvm) mutex_lock(&kvm->slots_lock); ret = kvm_io_bus_register_dev(kvm, KVM_MMIO_BUS, &ioapic->dev); mutex_unlock(&kvm->slots_lock); - if (ret < 0) + if (ret < 0) { + kvm->arch.vioapic = NULL; kfree(ioapic); + } return ret; } -- cgit v1.2.3 From 72bb2fcd23afe8db53b47e8f9edd736c517ba532 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Tue, 9 Feb 2010 10:33:03 +0800 Subject: KVM: cleanup the failure path of KVM_CREATE_IRQCHIP ioctrl If we fail to init ioapic device or the fail to setup the default irq routing, the device register by kvm_create_pic() and kvm_ioapic_init() remain unregister. This patch fixed to do this. Signed-off-by: Wei Yongjun Signed-off-by: Avi Kivity --- virt/kvm/ioapic.c | 11 +++++++++++ virt/kvm/ioapic.h | 1 + 2 files changed, 12 insertions(+) (limited to 'virt') diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index f3d06934ae6d..3db15a807f80 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -401,6 +401,17 @@ int kvm_ioapic_init(struct kvm *kvm) return ret; } +void kvm_ioapic_destroy(struct kvm *kvm) +{ + struct kvm_ioapic *ioapic = kvm->arch.vioapic; + + if (ioapic) { + kvm_io_bus_unregister_dev(kvm, KVM_MMIO_BUS, &ioapic->dev); + kvm->arch.vioapic = NULL; + kfree(ioapic); + } +} + int kvm_get_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state) { struct kvm_ioapic *ioapic = ioapic_irqchip(kvm); diff --git a/virt/kvm/ioapic.h b/virt/kvm/ioapic.h index a505ce9054f3..8a751b78a430 100644 --- a/virt/kvm/ioapic.h +++ b/virt/kvm/ioapic.h @@ -72,6 +72,7 @@ int kvm_apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source, int kvm_apic_compare_prio(struct kvm_vcpu *vcpu1, struct kvm_vcpu *vcpu2); void kvm_ioapic_update_eoi(struct kvm *kvm, int vector, int trigger_mode); int kvm_ioapic_init(struct kvm *kvm); +void kvm_ioapic_destroy(struct kvm *kvm); int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level); void kvm_ioapic_reset(struct kvm_ioapic *ioapic); int kvm_irq_delivery_to_apic(struct kvm *kvm, struct kvm_lapic *src, -- cgit v1.2.3 From 8b97fb0fcba0e119d539ef6126ddd1d7ba9c007f Mon Sep 17 00:00:00 2001 From: "Michael S. Tsirkin" Date: Wed, 13 Jan 2010 19:12:39 +0200 Subject: KVM: do not store wqh in irqfd wqh is unused, so we do not need to store it in irqfd anymore Signed-off-by: Michael S. Tsirkin Signed-off-by: Avi Kivity --- virt/kvm/eventfd.c | 3 --- 1 file changed, 3 deletions(-) (limited to 'virt') diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c index 486c604365d9..7016319b1ec0 100644 --- a/virt/kvm/eventfd.c +++ b/virt/kvm/eventfd.c @@ -47,7 +47,6 @@ struct _irqfd { int gsi; struct list_head list; poll_table pt; - wait_queue_head_t *wqh; wait_queue_t wait; struct work_struct inject; struct work_struct shutdown; @@ -159,8 +158,6 @@ irqfd_ptable_queue_proc(struct file *file, wait_queue_head_t *wqh, poll_table *pt) { struct _irqfd *irqfd = container_of(pt, struct _irqfd, pt); - - irqfd->wqh = wqh; add_wait_queue(wqh, &irqfd->wait); } -- cgit v1.2.3 From 70e335e16882df5b5d6971022e63c3603a1e8c23 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 18 Feb 2010 11:25:22 +0200 Subject: KVM: Convert kvm->requests_lock to raw_spinlock_t The code relies on kvm->requests_lock inhibiting preemption. Noted by Jan Kiszka. Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 0a360c26cc34..548f9253c195 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -138,7 +138,7 @@ static bool make_all_cpus_request(struct kvm *kvm, unsigned int req) zalloc_cpumask_var(&cpus, GFP_ATOMIC); - spin_lock(&kvm->requests_lock); + raw_spin_lock(&kvm->requests_lock); me = smp_processor_id(); kvm_for_each_vcpu(i, vcpu, kvm) { if (test_and_set_bit(req, &vcpu->requests)) @@ -153,7 +153,7 @@ static bool make_all_cpus_request(struct kvm *kvm, unsigned int req) smp_call_function_many(cpus, ack_flush, NULL, 1); else called = false; - spin_unlock(&kvm->requests_lock); + raw_spin_unlock(&kvm->requests_lock); free_cpumask_var(cpus); return called; } @@ -409,7 +409,7 @@ static struct kvm *kvm_create_vm(void) kvm->mm = current->mm; atomic_inc(&kvm->mm->mm_count); spin_lock_init(&kvm->mmu_lock); - spin_lock_init(&kvm->requests_lock); + raw_spin_lock_init(&kvm->requests_lock); kvm_eventfd_init(kvm); mutex_init(&kvm->lock); mutex_init(&kvm->irq_lock); -- cgit v1.2.3 From fcd95807fb61e67d602610e7ff7129ed769e9fee Mon Sep 17 00:00:00 2001 From: Joerg Roedel Date: Mon, 11 Jan 2010 16:38:18 +0100 Subject: kvm: Change kvm_iommu_map_pages to map large pages This patch changes the implementation of of kvm_iommu_map_pages to map the pages with the host page size into the io virtual address space. Signed-off-by: Joerg Roedel Acked-By: Avi Kivity --- virt/kvm/iommu.c | 113 ++++++++++++++++++++++++++++++++++++++++++++----------- 1 file changed, 91 insertions(+), 22 deletions(-) (limited to 'virt') diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index 80fd3ad3b2de..11692b9e8830 100644 --- a/virt/kvm/iommu.c +++ b/virt/kvm/iommu.c @@ -32,12 +32,30 @@ 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); +static pfn_t kvm_pin_pages(struct kvm *kvm, struct kvm_memory_slot *slot, + gfn_t gfn, unsigned long size) +{ + gfn_t end_gfn; + pfn_t pfn; + + pfn = gfn_to_pfn_memslot(kvm, slot, gfn); + end_gfn = gfn + (size >> PAGE_SHIFT); + gfn += 1; + + if (is_error_pfn(pfn)) + return pfn; + + while (gfn < end_gfn) + gfn_to_pfn_memslot(kvm, slot, gfn++); + + return pfn; +} + int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot) { - gfn_t gfn = slot->base_gfn; - unsigned long npages = slot->npages; + gfn_t gfn, end_gfn; pfn_t pfn; - int i, r = 0; + int r = 0; struct iommu_domain *domain = kvm->arch.iommu_domain; int flags; @@ -45,31 +63,62 @@ int kvm_iommu_map_pages(struct kvm *kvm, struct kvm_memory_slot *slot) if (!domain) return 0; + gfn = slot->base_gfn; + end_gfn = gfn + slot->npages; + flags = IOMMU_READ | IOMMU_WRITE; if (kvm->arch.iommu_flags & KVM_IOMMU_CACHE_COHERENCY) flags |= IOMMU_CACHE; - for (i = 0; i < npages; i++) { - /* check if already mapped */ - if (iommu_iova_to_phys(domain, gfn_to_gpa(gfn))) + + while (gfn < end_gfn) { + unsigned long page_size; + + /* Check if already mapped */ + if (iommu_iova_to_phys(domain, gfn_to_gpa(gfn))) { + gfn += 1; + continue; + } + + /* Get the page size we could use to map */ + page_size = kvm_host_page_size(kvm, gfn); + + /* Make sure the page_size does not exceed the memslot */ + while ((gfn + (page_size >> PAGE_SHIFT)) > end_gfn) + page_size >>= 1; + + /* Make sure gfn is aligned to the page size we want to map */ + while ((gfn << PAGE_SHIFT) & (page_size - 1)) + page_size >>= 1; + + /* + * Pin all pages we are about to map in memory. This is + * important because we unmap and unpin in 4kb steps later. + */ + pfn = kvm_pin_pages(kvm, slot, gfn, page_size); + if (is_error_pfn(pfn)) { + gfn += 1; continue; + } - pfn = gfn_to_pfn_memslot(kvm, slot, gfn); - r = iommu_map_range(domain, - gfn_to_gpa(gfn), - pfn_to_hpa(pfn), - PAGE_SIZE, flags); + /* Map into IO address space */ + r = iommu_map(domain, gfn_to_gpa(gfn), pfn_to_hpa(pfn), + get_order(page_size), flags); if (r) { printk(KERN_ERR "kvm_iommu_map_address:" "iommu failed to map pfn=%lx\n", pfn); goto unmap_pages; } - gfn++; + + gfn += page_size >> PAGE_SHIFT; + + } + return 0; unmap_pages: - kvm_iommu_put_pages(kvm, slot->base_gfn, i); + kvm_iommu_put_pages(kvm, slot->base_gfn, gfn); return r; } @@ -189,27 +238,47 @@ out_unmap: return r; } +static void kvm_unpin_pages(struct kvm *kvm, pfn_t pfn, unsigned long npages) +{ + unsigned long i; + + for (i = 0; i < npages; ++i) + kvm_release_pfn_clean(pfn + i); +} + static void kvm_iommu_put_pages(struct kvm *kvm, gfn_t base_gfn, unsigned long npages) { - gfn_t gfn = base_gfn; + struct iommu_domain *domain; + gfn_t end_gfn, gfn; pfn_t pfn; - struct iommu_domain *domain = kvm->arch.iommu_domain; - unsigned long i; u64 phys; + domain = kvm->arch.iommu_domain; + end_gfn = base_gfn + npages; + gfn = base_gfn; + /* check if iommu exists and in use */ if (!domain) return; - for (i = 0; i < npages; i++) { + while (gfn < end_gfn) { + unsigned long unmap_pages; + int order; + + /* Get physical address */ phys = iommu_iova_to_phys(domain, gfn_to_gpa(gfn)); - pfn = phys >> PAGE_SHIFT; - kvm_release_pfn_clean(pfn); - gfn++; - } + pfn = phys >> PAGE_SHIFT; + + /* Unmap address from IO address space */ + order = iommu_unmap(domain, gfn_to_gpa(gfn), PAGE_SIZE); + unmap_pages = 1ULL << order; - iommu_unmap_range(domain, gfn_to_gpa(base_gfn), PAGE_SIZE * npages); + /* Unpin all pages we just unmapped to not leak any memory */ + kvm_unpin_pages(kvm, pfn, unmap_pages); + + gfn += unmap_pages; + } } static int kvm_iommu_unmap_memslots(struct kvm *kvm) -- cgit v1.2.3 From 5a0e3ad6af8660be21ca98a971cd00f331318c05 Mon Sep 17 00:00:00 2001 From: Tejun Heo Date: Wed, 24 Mar 2010 17:04:11 +0900 Subject: include cleanup: Update gfp.h and slab.h includes to prepare for breaking implicit slab.h inclusion from percpu.h percpu.h is included by sched.h and module.h and thus ends up being included when building most .c files. percpu.h includes slab.h which in turn includes gfp.h making everything defined by the two files universally available and complicating inclusion dependencies. percpu.h -> slab.h dependency is about to be removed. Prepare for this change by updating users of gfp and slab facilities include those headers directly instead of assuming availability. As this conversion needs to touch large number of source files, the following script is used as the basis of conversion. http://userweb.kernel.org/~tj/misc/slabh-sweep.py The script does the followings. * Scan files for gfp and slab usages and update includes such that only the necessary includes are there. ie. if only gfp is used, gfp.h, if slab is used, slab.h. * When the script inserts a new include, it looks at the include blocks and try to put the new include such that its order conforms to its surrounding. It's put in the include block which contains core kernel includes, in the same order that the rest are ordered - alphabetical, Christmas tree, rev-Xmas-tree or at the end if there doesn't seem to be any matching order. * If the script can't find a place to put a new include (mostly because the file doesn't have fitting include block), it prints out an error message indicating which .h file needs to be added to the file. The conversion was done in the following steps. 1. The initial automatic conversion of all .c files updated slightly over 4000 files, deleting around 700 includes and adding ~480 gfp.h and ~3000 slab.h inclusions. The script emitted errors for ~400 files. 2. Each error was manually checked. Some didn't need the inclusion, some needed manual addition while adding it to implementation .h or embedding .c file was more appropriate for others. This step added inclusions to around 150 files. 3. The script was run again and the output was compared to the edits from #2 to make sure no file was left behind. 4. Several build tests were done and a couple of problems were fixed. e.g. lib/decompress_*.c used malloc/free() wrappers around slab APIs requiring slab.h to be added manually. 5. The script was run on all .h files but without automatically editing them as sprinkling gfp.h and slab.h inclusions around .h files could easily lead to inclusion dependency hell. Most gfp.h inclusion directives were ignored as stuff from gfp.h was usually wildly available and often used in preprocessor macros. Each slab.h inclusion directive was examined and added manually as necessary. 6. percpu.h was updated not to include slab.h. 7. Build test were done on the following configurations and failures were fixed. CONFIG_GCOV_KERNEL was turned off for all tests (as my distributed build env didn't work with gcov compiles) and a few more options had to be turned off depending on archs to make things build (like ipr on powerpc/64 which failed due to missing writeq). * x86 and x86_64 UP and SMP allmodconfig and a custom test config. * powerpc and powerpc64 SMP allmodconfig * sparc and sparc64 SMP allmodconfig * ia64 SMP allmodconfig * s390 SMP allmodconfig * alpha SMP allmodconfig * um on x86_64 SMP allmodconfig 8. percpu.h modifications were reverted so that it could be applied as a separate patch and serve as bisection point. Given the fact that I had only a couple of failures from tests on step 6, I'm fairly confident about the coverage of this conversion patch. If there is a breakage, it's likely to be something in one of the arch headers which should be easily discoverable easily on most builds of the specific arch. Signed-off-by: Tejun Heo Guess-its-ok-by: Christoph Lameter Cc: Ingo Molnar Cc: Lee Schermerhorn --- virt/kvm/assigned-dev.c | 1 + virt/kvm/coalesced_mmio.c | 1 + virt/kvm/eventfd.c | 1 + virt/kvm/ioapic.c | 1 + virt/kvm/irq_comm.c | 1 + virt/kvm/kvm_main.c | 2 +- 6 files changed, 6 insertions(+), 1 deletion(-) (limited to 'virt') diff --git a/virt/kvm/assigned-dev.c b/virt/kvm/assigned-dev.c index 057e2cca6af5..02ff2b19dbe2 100644 --- a/virt/kvm/assigned-dev.c +++ b/virt/kvm/assigned-dev.c @@ -16,6 +16,7 @@ #include #include #include +#include #include "irq.h" static struct kvm_assigned_dev_kernel *kvm_find_assigned_dev(struct list_head *head, diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index 5169736377a3..36e258029649 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -10,6 +10,7 @@ #include "iodev.h" #include +#include #include #include "coalesced_mmio.h" diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c index 7016319b1ec0..b81f0ebbaaad 100644 --- a/virt/kvm/eventfd.c +++ b/virt/kvm/eventfd.c @@ -30,6 +30,7 @@ #include #include #include +#include #include "iodev.h" diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index 3db15a807f80..03a5eb22da2b 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -33,6 +33,7 @@ #include #include #include +#include #include #include #include diff --git a/virt/kvm/irq_comm.c b/virt/kvm/irq_comm.c index 9fd5b3ebc517..a0e88809e45e 100644 --- a/virt/kvm/irq_comm.c +++ b/virt/kvm/irq_comm.c @@ -20,6 +20,7 @@ */ #include +#include #include #include diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 548f9253c195..5a0cd194dce0 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -22,7 +22,6 @@ #include #include #include -#include #include #include #include @@ -46,6 +45,7 @@ #include #include #include +#include #include #include -- cgit v1.2.3 From 87bf6e7de1134f48681fd2ce4b7c1ec45458cb6d Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Mon, 12 Apr 2010 19:35:35 +0900 Subject: KVM: fix the handling of dirty bitmaps to avoid overflows Int is not long enough to store the size of a dirty bitmap. This patch fixes this problem with the introduction of a wrapper function to calculate the sizes of dirty bitmaps. Note: in mark_page_dirty(), we have to consider the fact that __set_bit() takes the offset as int, not long. Signed-off-by: Takuya Yoshikawa Signed-off-by: Marcelo Tosatti --- virt/kvm/kvm_main.c | 13 ++++++++----- 1 file changed, 8 insertions(+), 5 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 5a0cd194dce0..364daacafb58 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -648,7 +648,7 @@ skip_lpage: /* Allocate page dirty bitmap if needed */ if ((new.flags & KVM_MEM_LOG_DIRTY_PAGES) && !new.dirty_bitmap) { - unsigned dirty_bytes = ALIGN(npages, BITS_PER_LONG) / 8; + unsigned long dirty_bytes = kvm_dirty_bitmap_bytes(&new); new.dirty_bitmap = vmalloc(dirty_bytes); if (!new.dirty_bitmap) @@ -768,7 +768,7 @@ int kvm_get_dirty_log(struct kvm *kvm, { struct kvm_memory_slot *memslot; int r, i; - int n; + unsigned long n; unsigned long any = 0; r = -EINVAL; @@ -780,7 +780,7 @@ int kvm_get_dirty_log(struct kvm *kvm, if (!memslot->dirty_bitmap) goto out; - n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; + n = kvm_dirty_bitmap_bytes(memslot); for (i = 0; !any && i < n/sizeof(long); ++i) any = memslot->dirty_bitmap[i]; @@ -1186,10 +1186,13 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn) memslot = gfn_to_memslot_unaliased(kvm, gfn); if (memslot && memslot->dirty_bitmap) { unsigned long rel_gfn = gfn - memslot->base_gfn; + unsigned long *p = memslot->dirty_bitmap + + rel_gfn / BITS_PER_LONG; + int offset = rel_gfn % BITS_PER_LONG; /* avoid RMW */ - if (!generic_test_le_bit(rel_gfn, memslot->dirty_bitmap)) - generic___set_le_bit(rel_gfn, memslot->dirty_bitmap); + if (!generic_test_le_bit(offset, p)) + generic___set_le_bit(offset, p); } } -- cgit v1.2.3 From eda2beda835697878e309b6049aa0a3bad051590 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Tue, 20 Apr 2010 14:29:29 +0800 Subject: KVM: Add missing srcu_read_lock() for kvm_mmu_notifier_release() I got this dmesg due to srcu_read_lock() is missing in kvm_mmu_notifier_release(). =================================================== [ INFO: suspicious rcu_dereference_check() usage. ] --------------------------------------------------- arch/x86/kvm/x86.h:72 invoked rcu_dereference_check() without protection! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 2 locks held by qemu-system-x86/3100: #0: (rcu_read_lock){.+.+..}, at: [] __mmu_notifier_release+0x38/0xdf #1: (&(&kvm->mmu_lock)->rlock){+.+...}, at: [] kvm_mmu_zap_all+0x21/0x5e [kvm] stack backtrace: Pid: 3100, comm: qemu-system-x86 Not tainted 2.6.34-rc3-22949-gbc8a97a-dirty #2 Call Trace: [] lockdep_rcu_dereference+0xaa/0xb3 [] unalias_gfn+0x56/0xab [kvm] [] gfn_to_memslot+0x16/0x25 [kvm] [] gfn_to_rmap+0x17/0x6e [kvm] [] rmap_remove+0xa0/0x19d [kvm] [] kvm_mmu_zap_page+0x109/0x34d [kvm] [] kvm_mmu_zap_all+0x35/0x5e [kvm] [] kvm_arch_flush_shadow+0x16/0x22 [kvm] [] kvm_mmu_notifier_release+0x15/0x17 [kvm] [] __mmu_notifier_release+0x88/0xdf [] ? __mmu_notifier_release+0x38/0xdf [] ? exit_mm+0xe0/0x115 [] exit_mmap+0x2c/0x17e [] mmput+0x2d/0xd4 [] exit_mm+0x108/0x115 [...] Signed-off-by: Lai Jiangshan Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 364daacafb58..c82ae2492634 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -341,7 +341,11 @@ static void kvm_mmu_notifier_release(struct mmu_notifier *mn, struct mm_struct *mm) { struct kvm *kvm = mmu_notifier_to_kvm(mn); + int idx; + + idx = srcu_read_lock(&kvm->srcu); kvm_arch_flush_shadow(kvm); + srcu_read_unlock(&kvm->srcu, idx); } static const struct mmu_notifier_ops kvm_mmu_notifier_ops = { -- cgit v1.2.3 From f5c9803173848864d0c56108b9e102db0bf601de Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Thu, 25 Feb 2010 11:33:19 +0900 Subject: KVM: update gfn_to_hva() to use gfn_to_hva_memslot() Marcelo introduced gfn_to_hva_memslot() when he implemented gfn_to_pfn_memslot(). Let's use this for gfn_to_hva() too. Note: also remove parentheses next to return as checkpatch said to do. Signed-off-by: Takuya Yoshikawa Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index c82ae2492634..5bac6eb0f0a9 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -914,6 +914,11 @@ int memslot_id(struct kvm *kvm, gfn_t gfn) return memslot - slots->memslots; } +static unsigned long gfn_to_hva_memslot(struct kvm_memory_slot *slot, gfn_t gfn) +{ + return slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE; +} + unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn) { struct kvm_memory_slot *slot; @@ -922,7 +927,7 @@ unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn) slot = gfn_to_memslot_unaliased(kvm, gfn); if (!slot || slot->flags & KVM_MEMSLOT_INVALID) return bad_hva(); - return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE); + return gfn_to_hva_memslot(slot, gfn); } EXPORT_SYMBOL_GPL(gfn_to_hva); @@ -972,11 +977,6 @@ pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn) } EXPORT_SYMBOL_GPL(gfn_to_pfn); -static unsigned long gfn_to_hva_memslot(struct kvm_memory_slot *slot, gfn_t gfn) -{ - return (slot->userspace_addr + (gfn - slot->base_gfn) * PAGE_SIZE); -} - pfn_t gfn_to_pfn_memslot(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn) { -- cgit v1.2.3 From 46a47b1ed118cda1a08b7f6077b837a00fbc112b Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 23 Apr 2010 14:03:38 -0300 Subject: KVM: convert ioapic lock to spinlock kvm_set_irq is used from non sleepable contexes, so convert ioapic from mutex to spinlock. KVM-Stable-Tag. Tested-by: Ralf Bonenkamp Signed-off-by: Marcelo Tosatti --- virt/kvm/ioapic.c | 30 +++++++++++++++--------------- virt/kvm/ioapic.h | 2 +- 2 files changed, 16 insertions(+), 16 deletions(-) (limited to 'virt') diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index 03a5eb22da2b..7c79c1d76d0c 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -197,7 +197,7 @@ int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level) union kvm_ioapic_redirect_entry entry; int ret = 1; - mutex_lock(&ioapic->lock); + spin_lock(&ioapic->lock); if (irq >= 0 && irq < IOAPIC_NUM_PINS) { entry = ioapic->redirtbl[irq]; level ^= entry.fields.polarity; @@ -214,7 +214,7 @@ int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level) } trace_kvm_ioapic_set_irq(entry.bits, irq, ret == 0); } - mutex_unlock(&ioapic->lock); + spin_unlock(&ioapic->lock); return ret; } @@ -238,9 +238,9 @@ static void __kvm_ioapic_update_eoi(struct kvm_ioapic *ioapic, int vector, * is dropped it will be put into irr and will be delivered * after ack notifier returns. */ - mutex_unlock(&ioapic->lock); + spin_unlock(&ioapic->lock); kvm_notify_acked_irq(ioapic->kvm, KVM_IRQCHIP_IOAPIC, i); - mutex_lock(&ioapic->lock); + spin_lock(&ioapic->lock); if (trigger_mode != IOAPIC_LEVEL_TRIG) continue; @@ -259,9 +259,9 @@ void kvm_ioapic_update_eoi(struct kvm *kvm, int vector, int trigger_mode) smp_rmb(); if (!test_bit(vector, ioapic->handled_vectors)) return; - mutex_lock(&ioapic->lock); + spin_lock(&ioapic->lock); __kvm_ioapic_update_eoi(ioapic, vector, trigger_mode); - mutex_unlock(&ioapic->lock); + spin_unlock(&ioapic->lock); } static inline struct kvm_ioapic *to_ioapic(struct kvm_io_device *dev) @@ -287,7 +287,7 @@ static int ioapic_mmio_read(struct kvm_io_device *this, gpa_t addr, int len, ASSERT(!(addr & 0xf)); /* check alignment */ addr &= 0xff; - mutex_lock(&ioapic->lock); + spin_lock(&ioapic->lock); switch (addr) { case IOAPIC_REG_SELECT: result = ioapic->ioregsel; @@ -301,7 +301,7 @@ static int ioapic_mmio_read(struct kvm_io_device *this, gpa_t addr, int len, result = 0; break; } - mutex_unlock(&ioapic->lock); + spin_unlock(&ioapic->lock); switch (len) { case 8: @@ -338,7 +338,7 @@ static int ioapic_mmio_write(struct kvm_io_device *this, gpa_t addr, int len, } addr &= 0xff; - mutex_lock(&ioapic->lock); + spin_lock(&ioapic->lock); switch (addr) { case IOAPIC_REG_SELECT: ioapic->ioregsel = data; @@ -356,7 +356,7 @@ static int ioapic_mmio_write(struct kvm_io_device *this, gpa_t addr, int len, default: break; } - mutex_unlock(&ioapic->lock); + spin_unlock(&ioapic->lock); return 0; } @@ -386,7 +386,7 @@ int kvm_ioapic_init(struct kvm *kvm) ioapic = kzalloc(sizeof(struct kvm_ioapic), GFP_KERNEL); if (!ioapic) return -ENOMEM; - mutex_init(&ioapic->lock); + spin_lock_init(&ioapic->lock); kvm->arch.vioapic = ioapic; kvm_ioapic_reset(ioapic); kvm_iodevice_init(&ioapic->dev, &ioapic_mmio_ops); @@ -419,9 +419,9 @@ int kvm_get_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state) if (!ioapic) return -EINVAL; - mutex_lock(&ioapic->lock); + spin_lock(&ioapic->lock); memcpy(state, ioapic, sizeof(struct kvm_ioapic_state)); - mutex_unlock(&ioapic->lock); + spin_unlock(&ioapic->lock); return 0; } @@ -431,9 +431,9 @@ int kvm_set_ioapic(struct kvm *kvm, struct kvm_ioapic_state *state) if (!ioapic) return -EINVAL; - mutex_lock(&ioapic->lock); + spin_lock(&ioapic->lock); memcpy(ioapic, state, sizeof(struct kvm_ioapic_state)); update_handled_vectors(ioapic); - mutex_unlock(&ioapic->lock); + spin_unlock(&ioapic->lock); return 0; } diff --git a/virt/kvm/ioapic.h b/virt/kvm/ioapic.h index 8a751b78a430..0b190c34ccc3 100644 --- a/virt/kvm/ioapic.h +++ b/virt/kvm/ioapic.h @@ -45,7 +45,7 @@ struct kvm_ioapic { struct kvm_io_device dev; struct kvm *kvm; void (*ack_notifier)(void *opaque, int irq); - struct mutex lock; + spinlock_t lock; DECLARE_BITMAP(handled_vectors, 256); }; -- cgit v1.2.3 From 2ed152afc7ed61830b848b32936e1541a1a57799 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Wed, 10 Mar 2010 19:00:43 +0800 Subject: KVM: cleanup kvm trace This patch does: - no need call tracepoint_synchronize_unregister() when kvm module is unloaded since ftrace can handle it - cleanup ftrace's macro Signed-off-by: Xiao Guangrong Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 1 - 1 file changed, 1 deletion(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 5bac6eb0f0a9..b152b23cd095 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -2279,7 +2279,6 @@ EXPORT_SYMBOL_GPL(kvm_init); void kvm_exit(void) { - tracepoint_synchronize_unregister(); kvm_exit_debug(); misc_deregister(&kvm_dev); kmem_cache_destroy(kvm_vcpu_cache); -- cgit v1.2.3 From a87fa35514bcf17aeca97245574e3f735cec9074 Mon Sep 17 00:00:00 2001 From: Wei Yongjun Date: Fri, 12 Mar 2010 12:59:06 +0800 Subject: KVM: fix the errno of ioctl KVM_[UN]REGISTER_COALESCED_MMIO failure This patch change the errno of ioctl KVM_[UN]REGISTER_COALESCED_MMIO from -EINVAL to -ENXIO if no coalesced mmio dev exists. Signed-off-by: Wei Yongjun Signed-off-by: Marcelo Tosatti --- virt/kvm/coalesced_mmio.c | 4 ++-- virt/kvm/kvm_main.c | 2 -- 2 files changed, 2 insertions(+), 4 deletions(-) (limited to 'virt') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index 36e258029649..b66001f8754c 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -139,7 +139,7 @@ int kvm_vm_ioctl_register_coalesced_mmio(struct kvm *kvm, struct kvm_coalesced_mmio_dev *dev = kvm->coalesced_mmio_dev; if (dev == NULL) - return -EINVAL; + return -ENXIO; mutex_lock(&kvm->slots_lock); if (dev->nb_zones >= KVM_COALESCED_MMIO_ZONE_MAX) { @@ -162,7 +162,7 @@ int kvm_vm_ioctl_unregister_coalesced_mmio(struct kvm *kvm, struct kvm_coalesced_mmio_zone *z; if (dev == NULL) - return -EINVAL; + return -ENXIO; mutex_lock(&kvm->slots_lock); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index b152b23cd095..0db6dfcc8420 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1609,7 +1609,6 @@ static long kvm_vm_ioctl(struct file *filp, r = -EFAULT; if (copy_from_user(&zone, argp, sizeof zone)) goto out; - r = -ENXIO; r = kvm_vm_ioctl_register_coalesced_mmio(kvm, &zone); if (r) goto out; @@ -1621,7 +1620,6 @@ static long kvm_vm_ioctl(struct file *filp, r = -EFAULT; if (copy_from_user(&zone, argp, sizeof zone)) goto out; - r = -ENXIO; r = kvm_vm_ioctl_unregister_coalesced_mmio(kvm, &zone); if (r) goto out; -- cgit v1.2.3 From d57e2c0740bbdd768dcbafe58cf62174f31d7c2d Mon Sep 17 00:00:00 2001 From: jing zhang Date: Sat, 13 Mar 2010 15:00:45 +0800 Subject: KVM: fix assigned_device_enable_host_msix error handling Free IRQ's and disable MSIX upon failure. Cc: Avi Kivity Signed-off-by: Jing Zhang Signed-off-by: Marcelo Tosatti --- virt/kvm/assigned-dev.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-) (limited to 'virt') diff --git a/virt/kvm/assigned-dev.c b/virt/kvm/assigned-dev.c index 02ff2b19dbe2..4d10b1e047f4 100644 --- a/virt/kvm/assigned-dev.c +++ b/virt/kvm/assigned-dev.c @@ -316,12 +316,16 @@ static int assigned_device_enable_host_msix(struct kvm *kvm, kvm_assigned_dev_intr, 0, "kvm_assigned_msix_device", (void *)dev); - /* FIXME: free requested_irq's on failure */ if (r) - return r; + goto err; } return 0; +err: + for (i -= 1; i >= 0; i--) + free_irq(dev->host_msix_entries[i].vector, (void *)dev); + pci_disable_msix(dev->dev); + return r; } #endif -- cgit v1.2.3 From 6ce5a090a9a0ea4266a2cad058c69e2f27201e11 Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Mon, 15 Mar 2010 22:13:30 +0900 Subject: KVM: coalesced_mmio: fix kvm_coalesced_mmio_init()'s error handling kvm_coalesced_mmio_init() keeps to hold the addresses of a coalesced mmio ring page and dev even after it has freed them. Also, if this function fails, though it might be rare, it seems to be suggesting the system's serious state: so we'd better stop the works following the kvm_creat_vm(). This patch clears these problems. We move the coalesced mmio's initialization out of kvm_create_vm(). This seems to be natural because it includes a registration which can be done only when vm is successfully created. Signed-off-by: Takuya Yoshikawa Signed-off-by: Marcelo Tosatti --- virt/kvm/coalesced_mmio.c | 2 ++ virt/kvm/kvm_main.c | 12 ++++++++---- 2 files changed, 10 insertions(+), 4 deletions(-) (limited to 'virt') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index b66001f8754c..53850177163f 100644 --- a/virt/kvm/coalesced_mmio.c +++ b/virt/kvm/coalesced_mmio.c @@ -120,8 +120,10 @@ int kvm_coalesced_mmio_init(struct kvm *kvm) return ret; out_free_dev: + kvm->coalesced_mmio_dev = NULL; kfree(dev); out_free_page: + kvm->coalesced_mmio_ring = NULL; __free_page(page); out_err: return ret; diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 0db6dfcc8420..55a5d4804499 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -422,9 +422,6 @@ static struct kvm *kvm_create_vm(void) spin_lock(&kvm_lock); list_add(&kvm->vm_list, &vm_list); spin_unlock(&kvm_lock); -#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET - kvm_coalesced_mmio_init(kvm); -#endif out: return kvm; @@ -1753,12 +1750,19 @@ static struct file_operations kvm_vm_fops = { static int kvm_dev_ioctl_create_vm(void) { - int fd; + int fd, r; struct kvm *kvm; kvm = kvm_create_vm(); if (IS_ERR(kvm)) return PTR_ERR(kvm); +#ifdef KVM_COALESCED_MMIO_PAGE_OFFSET + r = kvm_coalesced_mmio_init(kvm); + if (r < 0) { + kvm_put_kvm(kvm); + return r; + } +#endif fd = anon_inode_getfd("kvm-vm", &kvm_vm_fops, kvm, O_RDWR); if (fd < 0) kvm_put_kvm(kvm); -- cgit v1.2.3 From 660c22c425cbe14badfb3b0a0206862577701ab7 Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Tue, 13 Apr 2010 22:47:24 +0900 Subject: KVM: limit the number of pages per memory slot This patch limits the number of pages per memory slot to make us free from extra care about type issues. Signed-off-by: Takuya Yoshikawa Signed-off-by: Marcelo Tosatti --- virt/kvm/kvm_main.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 55a5d4804499..d6351a34b297 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -557,6 +557,10 @@ int __kvm_set_memory_region(struct kvm *kvm, base_gfn = mem->guest_phys_addr >> PAGE_SHIFT; npages = mem->memory_size >> PAGE_SHIFT; + r = -EINVAL; + if (npages > KVM_MEM_MAX_NR_PAGES) + goto out; + if (!npages) mem->flags &= ~KVM_MEM_LOG_DIRTY_PAGES; @@ -1187,13 +1191,10 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn) memslot = gfn_to_memslot_unaliased(kvm, gfn); if (memslot && memslot->dirty_bitmap) { unsigned long rel_gfn = gfn - memslot->base_gfn; - unsigned long *p = memslot->dirty_bitmap + - rel_gfn / BITS_PER_LONG; - int offset = rel_gfn % BITS_PER_LONG; /* avoid RMW */ - if (!generic_test_le_bit(offset, p)) - generic___set_le_bit(offset, p); + if (!generic_test_le_bit(rel_gfn, memslot->dirty_bitmap)) + generic___set_le_bit(rel_gfn, memslot->dirty_bitmap); } } -- cgit v1.2.3 From 90d83dc3d49f5101addae962ccc1b4aff66b68d8 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Mon, 19 Apr 2010 17:41:23 +0800 Subject: KVM: use the correct RCU API for PROVE_RCU=y The RCU/SRCU API have already changed for proving RCU usage. I got the following dmesg when PROVE_RCU=y because we used incorrect API. This patch coverts rcu_deference() to srcu_dereference() or family API. =================================================== [ INFO: suspicious rcu_dereference_check() usage. ] --------------------------------------------------- arch/x86/kvm/mmu.c:3020 invoked rcu_dereference_check() without protection! other info that might help us debug this: rcu_scheduler_active = 1, debug_locks = 0 2 locks held by qemu-system-x86/8550: #0: (&kvm->slots_lock){+.+.+.}, at: [] kvm_set_memory_region+0x29/0x50 [kvm] #1: (&(&kvm->mmu_lock)->rlock){+.+...}, at: [] kvm_arch_commit_memory_region+0xa6/0xe2 [kvm] stack backtrace: Pid: 8550, comm: qemu-system-x86 Not tainted 2.6.34-rc4-tip-01028-g939eab1 #27 Call Trace: [] lockdep_rcu_dereference+0xaa/0xb3 [] kvm_mmu_calculate_mmu_pages+0x44/0x7d [kvm] [] kvm_arch_commit_memory_region+0xb7/0xe2 [kvm] [] __kvm_set_memory_region+0x636/0x6e2 [kvm] [] kvm_set_memory_region+0x37/0x50 [kvm] [] vmx_set_tss_addr+0x46/0x5a [kvm_intel] [] kvm_arch_vm_ioctl+0x17a/0xcf8 [kvm] [] ? unlock_page+0x27/0x2c [] ? __do_fault+0x3a9/0x3e1 [] kvm_vm_ioctl+0x364/0x38d [kvm] [] ? up_read+0x23/0x3d [] vfs_ioctl+0x32/0xa6 [] do_vfs_ioctl+0x495/0x4db [] ? fget_light+0xc2/0x241 [] ? do_sys_open+0x104/0x116 [] ? retint_swapgs+0xe/0x13 [] sys_ioctl+0x47/0x6a [] system_call_fastpath+0x16/0x1b Signed-off-by: Lai Jiangshan Signed-off-by: Avi Kivity --- virt/kvm/iommu.c | 4 ++-- virt/kvm/kvm_main.c | 13 ++++++++----- 2 files changed, 10 insertions(+), 7 deletions(-) (limited to 'virt') diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index 80fd3ad3b2de..37ca71ebdba8 100644 --- a/virt/kvm/iommu.c +++ b/virt/kvm/iommu.c @@ -78,7 +78,7 @@ static int kvm_iommu_map_memslots(struct kvm *kvm) int i, r = 0; struct kvm_memslots *slots; - slots = rcu_dereference(kvm->memslots); + slots = kvm_memslots(kvm); for (i = 0; i < slots->nmemslots; i++) { r = kvm_iommu_map_pages(kvm, &slots->memslots[i]); @@ -217,7 +217,7 @@ static int kvm_iommu_unmap_memslots(struct kvm *kvm) int i; struct kvm_memslots *slots; - slots = rcu_dereference(kvm->memslots); + slots = kvm_memslots(kvm); for (i = 0; i < slots->nmemslots; i++) { kvm_iommu_put_pages(kvm, slots->memslots[i].base_gfn, diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index d6351a34b297..4901ec5061ba 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -834,7 +834,7 @@ EXPORT_SYMBOL_GPL(kvm_is_error_hva); struct kvm_memory_slot *gfn_to_memslot_unaliased(struct kvm *kvm, gfn_t gfn) { int i; - struct kvm_memslots *slots = rcu_dereference(kvm->memslots); + struct kvm_memslots *slots = kvm_memslots(kvm); for (i = 0; i < slots->nmemslots; ++i) { struct kvm_memory_slot *memslot = &slots->memslots[i]; @@ -856,7 +856,7 @@ struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn) int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn) { int i; - struct kvm_memslots *slots = rcu_dereference(kvm->memslots); + struct kvm_memslots *slots = kvm_memslots(kvm); gfn = unalias_gfn_instantiation(kvm, gfn); for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { @@ -900,7 +900,7 @@ out: int memslot_id(struct kvm *kvm, gfn_t gfn) { int i; - struct kvm_memslots *slots = rcu_dereference(kvm->memslots); + struct kvm_memslots *slots = kvm_memslots(kvm); struct kvm_memory_slot *memslot = NULL; gfn = unalias_gfn(kvm, gfn); @@ -1994,7 +1994,9 @@ int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len, const void *val) { int i; - struct kvm_io_bus *bus = rcu_dereference(kvm->buses[bus_idx]); + struct kvm_io_bus *bus; + + bus = srcu_dereference(kvm->buses[bus_idx], &kvm->srcu); for (i = 0; i < bus->dev_count; i++) if (!kvm_iodevice_write(bus->devs[i], addr, len, val)) return 0; @@ -2006,8 +2008,9 @@ int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len, void *val) { int i; - struct kvm_io_bus *bus = rcu_dereference(kvm->buses[bus_idx]); + struct kvm_io_bus *bus; + bus = srcu_dereference(kvm->buses[bus_idx], &kvm->srcu); for (i = 0; i < bus->dev_count; i++) if (!kvm_iodevice_read(bus->devs[i], addr, len, val)) return 0; -- cgit v1.2.3 From 66cbff59a16017234ea6c80ecddfc17b27a83504 Mon Sep 17 00:00:00 2001 From: Lai Jiangshan Date: Sat, 17 Apr 2010 17:00:19 +0800 Subject: KVM: do not call hardware_disable() on CPU_UP_CANCELED When CPU_UP_CANCELED, hardware_enable() has not been called at the CPU which is going up because raw_notifier_call_chain(CPU_ONLINE) has not been called for this cpu. Drop the handling for CPU_UP_CANCELED. Signed-off-by: Lai Jiangshan Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 5 ----- 1 file changed, 5 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 4901ec5061ba..6dc940455e8b 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1931,11 +1931,6 @@ static int kvm_cpu_hotplug(struct notifier_block *notifier, unsigned long val, cpu); hardware_disable(NULL); break; - case CPU_UP_CANCELED: - printk(KERN_INFO "kvm: disabling virtualization on CPU%d\n", - cpu); - smp_call_function_single(cpu, hardware_disable, NULL, 1); - break; case CPU_ONLINE: printk(KERN_INFO "kvm: enabling virtualization on CPU%d\n", cpu); -- cgit v1.2.3 From d14769377a247d4e7b570592a090474c8a059938 Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Fri, 23 Apr 2010 17:48:35 +0900 Subject: KVM: Remove test-before-set optimization for dirty bits As Avi pointed out, testing bit part in mark_page_dirty() was important in the days of shadow paging, but currently EPT and NPT has already become common and the chance of faulting a page more that once per iteration is small. So let's remove the test bit to avoid extra access. Signed-off-by: Takuya Yoshikawa Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 6dc940455e8b..9ab1a77941ef 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -1192,9 +1192,7 @@ void mark_page_dirty(struct kvm *kvm, gfn_t gfn) if (memslot && memslot->dirty_bitmap) { unsigned long rel_gfn = gfn - memslot->base_gfn; - /* avoid RMW */ - if (!generic_test_le_bit(rel_gfn, memslot->dirty_bitmap)) - generic___set_le_bit(rel_gfn, memslot->dirty_bitmap); + generic___set_le_bit(rel_gfn, memslot->dirty_bitmap); } } -- cgit v1.2.3 From 0ee75bead83da4791e5cbf659806c54d8ee40f12 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Wed, 28 Apr 2010 15:39:01 +0300 Subject: KVM: Let vcpu structure alignment be determined at runtime vmx and svm vcpus have different contents and therefore may have different alignmment requirements. Let each specify its required alignment. Signed-off-by: Avi Kivity --- virt/kvm/kvm_main.c | 7 ++++--- 1 file changed, 4 insertions(+), 3 deletions(-) (limited to 'virt') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 9ab1a77941ef..f032806a212f 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -2178,7 +2178,7 @@ static void kvm_sched_out(struct preempt_notifier *pn, kvm_arch_vcpu_put(vcpu); } -int kvm_init(void *opaque, unsigned int vcpu_size, +int kvm_init(void *opaque, unsigned vcpu_size, unsigned vcpu_align, struct module *module) { int r; @@ -2228,8 +2228,9 @@ int kvm_init(void *opaque, unsigned int vcpu_size, goto out_free_4; /* A kmem cache lets us meet the alignment requirements of fx_save. */ - kvm_vcpu_cache = kmem_cache_create("kvm_vcpu", vcpu_size, - __alignof__(struct kvm_vcpu), + if (!vcpu_align) + vcpu_align = __alignof__(struct kvm_vcpu); + kvm_vcpu_cache = kmem_cache_create("kvm_vcpu", vcpu_size, vcpu_align, 0, NULL); if (!kvm_vcpu_cache) { r = -ENOMEM; -- cgit v1.2.3 From 05b782ab951a896d7da41775999821f692dc9e01 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Wed, 26 May 2010 21:36:33 +0200 Subject: KVM: Fix order passed to iommu_unmap This is obviously a left-over from the the old interface taking the size. Apparently a mostly harmless issue with the current iommu_unmap implementation. Signed-off-by: Jan Kiszka Acked-by: Joerg Roedel Signed-off-by: Avi Kivity --- virt/kvm/iommu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'virt') diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index d2f06be63354..96048ee9e39e 100644 --- a/virt/kvm/iommu.c +++ b/virt/kvm/iommu.c @@ -271,7 +271,7 @@ static void kvm_iommu_put_pages(struct kvm *kvm, pfn = phys >> PAGE_SHIFT; /* Unmap address from IO address space */ - order = iommu_unmap(domain, gfn_to_gpa(gfn), PAGE_SIZE); + order = iommu_unmap(domain, gfn_to_gpa(gfn), 0); unmap_pages = 1ULL << order; /* Unpin all pages we just unmapped to not leak any memory */ -- cgit v1.2.3 From 07dc7263b99e4ddad2b4c69765a428ccb7d48938 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Wed, 2 Jun 2010 11:26:26 -0300 Subject: KVM: read apic->irr with ioapic lock held Read ioapic->irr inside ioapic->lock protected section. KVM-Stable-Tag Signed-off-by: Marcelo Tosatti --- virt/kvm/ioapic.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'virt') diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index 7c79c1d76d0c..3500dee9cf2b 100644 --- a/virt/kvm/ioapic.c +++ b/virt/kvm/ioapic.c @@ -192,12 +192,13 @@ static int ioapic_deliver(struct kvm_ioapic *ioapic, int irq) int kvm_ioapic_set_irq(struct kvm_ioapic *ioapic, int irq, int level) { - u32 old_irr = ioapic->irr; + u32 old_irr; u32 mask = 1 << irq; union kvm_ioapic_redirect_entry entry; int ret = 1; spin_lock(&ioapic->lock); + old_irr = ioapic->irr; if (irq >= 0 && irq < IOAPIC_NUM_PINS) { entry = ioapic->redirtbl[irq]; level ^= entry.fields.polarity; -- cgit v1.2.3