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/kvm') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index a944be392d6..92f5c98b6c4 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/coalesced_mmio.h b/virt/kvm/coalesced_mmio.h index 4b49f27fa31..e7033469733 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 92f5c98b6c4..c10d117c431 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 v1.2.3-70-g09d2 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 --- arch/ia64/kvm/Kconfig | 1 + arch/powerpc/kvm/Kconfig | 1 + arch/x86/kvm/Kconfig | 1 + virt/kvm/Kconfig | 3 +++ 4 files changed, 6 insertions(+) (limited to 'virt/kvm') diff --git a/arch/ia64/kvm/Kconfig b/arch/ia64/kvm/Kconfig index ef3e7be29ca..bf82e47c98b 100644 --- a/arch/ia64/kvm/Kconfig +++ b/arch/ia64/kvm/Kconfig @@ -26,6 +26,7 @@ config KVM select ANON_INODES select HAVE_KVM_IRQCHIP select KVM_APIC_ARCHITECTURE + select KVM_MMIO ---help--- Support hosting fully virtualized guest machines using hardware virtualization extensions. You will need a fairly recent diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig index 6fb6e8aa389..be28968c794 100644 --- a/arch/powerpc/kvm/Kconfig +++ b/arch/powerpc/kvm/Kconfig @@ -20,6 +20,7 @@ config KVM bool select PREEMPT_NOTIFIERS select ANON_INODES + select KVM_MMIO config KVM_BOOK3S_64_HANDLER bool diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index 4cd49833246..06871111bf5 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -29,6 +29,7 @@ config KVM select HAVE_KVM_EVENTFD select KVM_APIC_ARCHITECTURE select USER_RETURN_NOTIFIER + select KVM_MMIO ---help--- Support hosting fully virtualized guest machines using hardware virtualization extensions. You will need a fairly recent diff --git a/virt/kvm/Kconfig b/virt/kvm/Kconfig index daece36c0a5..7f1178f6b83 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index 04d69cd7049..d68e6c68e0f 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 e7033469733..8a5959e3535 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 c10d117c431..bc23b8e0609 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 v1.2.3-70-g09d2 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 --- arch/ia64/kvm/kvm-ia64.c | 10 ++++++---- arch/powerpc/kvm/book3s.c | 2 +- arch/x86/kvm/mmu.c | 11 ++++++----- arch/x86/kvm/vmx.c | 4 ++-- arch/x86/kvm/x86.c | 4 ++-- include/linux/kvm_host.h | 12 ++++++++---- virt/kvm/iommu.c | 18 ++++++++++++------ virt/kvm/kvm_main.c | 36 +++++++++++++++++++++++------------- 8 files changed, 60 insertions(+), 37 deletions(-) (limited to 'virt/kvm') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 5fdeec5fddc..1ca1dbf4811 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1377,12 +1377,14 @@ static void free_kvm(struct kvm *kvm) static void kvm_release_vm_pages(struct kvm *kvm) { + struct kvm_memslots *slots; struct kvm_memory_slot *memslot; int i, j; unsigned long base_gfn; - for (i = 0; i < kvm->nmemslots; i++) { - memslot = &kvm->memslots[i]; + slots = kvm->memslots; + for (i = 0; i < slots->nmemslots; i++) { + memslot = &slots->memslots[i]; base_gfn = memslot->base_gfn; for (j = 0; j < memslot->npages; j++) { @@ -1802,7 +1804,7 @@ static int kvm_ia64_sync_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; @@ -1840,7 +1842,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, /* If nothing is dirty, don't bother messing with page tables. */ if (is_dirty) { kvm_flush_remote_tlbs(kvm); - memslot = &kvm->memslots[log->slot]; + memslot = &kvm->memslots->memslots[log->slot]; n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; memset(memslot->dirty_bitmap, 0, n); } diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index 803505d3e45..bb8873dcb20 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -865,7 +865,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, /* If nothing is dirty, don't bother messing with page tables. */ if (is_dirty) { - memslot = &kvm->memslots[log->slot]; + memslot = &kvm->memslots->memslots[log->slot]; ga = memslot->base_gfn << PAGE_SHIFT; ga_end = ga + (memslot->npages << PAGE_SHIFT); diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 4f499d7f710..81f84d326a8 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -807,13 +807,14 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, { int i, j; int retval = 0; + struct kvm_memslots *slots = kvm->memslots; /* * If mmap_sem isn't taken, we can look the memslots with only * the mmu_lock by skipping over the slots with userspace_addr == 0. */ - 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]; unsigned long start = memslot->userspace_addr; unsigned long end; @@ -3021,8 +3022,8 @@ unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm) unsigned int nr_mmu_pages; unsigned int nr_pages = 0; - for (i = 0; i < kvm->nmemslots; i++) - nr_pages += kvm->memslots[i].npages; + for (i = 0; i < kvm->memslots->nmemslots; i++) + nr_pages += kvm->memslots->memslots[i].npages; nr_mmu_pages = nr_pages * KVM_PERMILLE_MMU_PAGES / 1000; nr_mmu_pages = max(nr_mmu_pages, @@ -3295,7 +3296,7 @@ static int count_rmaps(struct kvm_vcpu *vcpu) int i, j, k; for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { - struct kvm_memory_slot *m = &vcpu->kvm->memslots[i]; + struct kvm_memory_slot *m = &vcpu->kvm->memslots->memslots[i]; struct kvm_rmap_desc *d; for (j = 0; j < m->npages; ++j) { diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 74a66f0c00b..18698799e36 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -1503,8 +1503,8 @@ static void enter_pmode(struct kvm_vcpu *vcpu) static gva_t rmode_tss_base(struct kvm *kvm) { if (!kvm->arch.tss_addr) { - gfn_t base_gfn = kvm->memslots[0].base_gfn + - kvm->memslots[0].npages - 3; + gfn_t base_gfn = kvm->memslots->memslots[0].base_gfn + + kvm->memslots->memslots[0].npages - 3; return base_gfn << PAGE_SHIFT; } return kvm->arch.tss_addr; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 8798504ace1..3b81cb9da8b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2427,7 +2427,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, spin_lock(&kvm->mmu_lock); kvm_mmu_slot_remove_write_access(kvm, log->slot); spin_unlock(&kvm->mmu_lock); - memslot = &kvm->memslots[log->slot]; + memslot = &kvm->memslots->memslots[log->slot]; n = ALIGN(memslot->npages, BITS_PER_LONG) / 8; memset(memslot->dirty_bitmap, 0, n); } @@ -5223,7 +5223,7 @@ int kvm_arch_set_memory_region(struct kvm *kvm, int user_alloc) { int npages = mem->memory_size >> PAGE_SHIFT; - struct kvm_memory_slot *memslot = &kvm->memslots[mem->slot]; + struct kvm_memory_slot *memslot = &kvm->memslots->memslots[mem->slot]; /*To keep backward compatibility with older userspace, *x86 needs to hanlde !user_alloc case. diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index bd5a616d937..782bfb185f8 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -150,14 +150,18 @@ struct kvm_irq_routing_table {}; #endif +struct kvm_memslots { + int nmemslots; + struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS + + KVM_PRIVATE_MEM_SLOTS]; +}; + struct kvm { spinlock_t mmu_lock; spinlock_t requests_lock; struct rw_semaphore slots_lock; struct mm_struct *mm; /* userspace tied to this vm */ - int nmemslots; - struct kvm_memory_slot memslots[KVM_MEMORY_SLOTS + - KVM_PRIVATE_MEM_SLOTS]; + struct kvm_memslots *memslots; #ifdef CONFIG_KVM_APIC_ARCHITECTURE u32 bsp_vcpu_id; struct kvm_vcpu *bsp_vcpu; @@ -482,7 +486,7 @@ static inline void kvm_guest_exit(void) static inline int memslot_id(struct kvm *kvm, struct kvm_memory_slot *slot) { - return slot - kvm->memslots; + return slot - kvm->memslots->memslots; } static inline gpa_t gfn_to_gpa(gfn_t gfn) diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index 15147583abd..bc697a66a88 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 bc23b8e0609..86dd8f3d29c 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 v1.2.3-70-g09d2 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 --- arch/ia64/kvm/kvm-ia64.c | 16 +++++++++++---- arch/powerpc/kvm/powerpc.c | 18 ++++++++++++---- arch/s390/kvm/kvm-s390.c | 25 +++++++++++++++-------- arch/x86/kvm/x86.c | 51 ++++++++++++++++++++++++++-------------------- include/linux/kvm_host.h | 7 ++++++- virt/kvm/kvm_main.c | 12 +++++------ 6 files changed, 82 insertions(+), 47 deletions(-) (limited to 'virt/kvm') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 1ca1dbf4811..0757c702798 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1578,15 +1578,15 @@ out: return r; } -int kvm_arch_set_memory_region(struct kvm *kvm, - struct kvm_userspace_memory_region *mem, +int kvm_arch_prepare_memory_region(struct kvm *kvm, + struct kvm_memory_slot *memslot, struct kvm_memory_slot old, + struct kvm_userspace_memory_region *mem, int user_alloc) { unsigned long i; unsigned long pfn; - int npages = mem->memory_size >> PAGE_SHIFT; - struct kvm_memory_slot *memslot = &kvm->memslots[mem->slot]; + int npages = memslot->npages; unsigned long base_gfn = memslot->base_gfn; if (base_gfn + npages > (KVM_MAX_MEM_SIZE >> PAGE_SHIFT)) @@ -1610,6 +1610,14 @@ int kvm_arch_set_memory_region(struct kvm *kvm, return 0; } +void kvm_arch_commit_memory_region(struct kvm *kvm, + struct kvm_userspace_memory_region *mem, + struct kvm_memory_slot old, + int user_alloc) +{ + return; +} + void kvm_arch_flush_shadow(struct kvm *kvm) { kvm_flush_remote_tlbs(kvm); diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index f06cf93b178..4633e7850dd 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -165,14 +165,24 @@ long kvm_arch_dev_ioctl(struct file *filp, return -EINVAL; } -int kvm_arch_set_memory_region(struct kvm *kvm, - struct kvm_userspace_memory_region *mem, - struct kvm_memory_slot old, - int user_alloc) +int kvm_arch_prepare_memory_region(struct kvm *kvm, + struct kvm_memory_slot *memslot, + struct kvm_memory_slot old, + struct kvm_userspace_memory_region *mem, + int user_alloc) { return 0; } +void kvm_arch_commit_memory_region(struct kvm *kvm, + struct kvm_userspace_memory_region *mem, + struct kvm_memory_slot old, + int user_alloc) +{ + return; +} + + void kvm_arch_flush_shadow(struct kvm *kvm) { } diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index 3fa0a10e466..c8002193d9d 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -690,14 +690,12 @@ long kvm_arch_vcpu_ioctl(struct file *filp, } /* Section: memory related */ -int kvm_arch_set_memory_region(struct kvm *kvm, - struct kvm_userspace_memory_region *mem, - struct kvm_memory_slot old, - int user_alloc) +int kvm_arch_prepare_memory_region(struct kvm *kvm, + struct kvm_memory_slot *memslot, + struct kvm_memory_slot old, + struct kvm_userspace_memory_region *mem, + int user_alloc) { - int i; - struct kvm_vcpu *vcpu; - /* A few sanity checks. We can have exactly one memory slot which has to start at guest virtual zero and which has to be located at a page boundary in userland and which has to end at a page boundary. @@ -720,14 +718,23 @@ int kvm_arch_set_memory_region(struct kvm *kvm, if (!user_alloc) return -EINVAL; + return 0; +} + +void kvm_arch_commit_memory_region(struct kvm *kvm, + struct kvm_userspace_memory_region *mem, + struct kvm_memory_slot old, + int user_alloc) +{ + int i; + struct kvm_vcpu *vcpu; + /* request update of sie control block for all available vcpus */ kvm_for_each_vcpu(i, vcpu, kvm) { if (test_and_set_bit(KVM_REQ_MMU_RELOAD, &vcpu->requests)) continue; kvm_s390_inject_sigp_stop(vcpu, ACTION_RELOADVCPU_ON_STOP); } - - return 0; } void kvm_arch_flush_shadow(struct kvm *kvm) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 1ce83319143..43da65feed4 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -5228,13 +5228,13 @@ void kvm_arch_destroy_vm(struct kvm *kvm) kfree(kvm); } -int kvm_arch_set_memory_region(struct kvm *kvm, - struct kvm_userspace_memory_region *mem, +int kvm_arch_prepare_memory_region(struct kvm *kvm, + struct kvm_memory_slot *memslot, struct kvm_memory_slot old, + struct kvm_userspace_memory_region *mem, int user_alloc) { - int npages = mem->memory_size >> PAGE_SHIFT; - struct kvm_memory_slot *memslot = &kvm->memslots->memslots[mem->slot]; + int npages = memslot->npages; /*To keep backward compatibility with older userspace, *x86 needs to hanlde !user_alloc case. @@ -5254,26 +5254,35 @@ int kvm_arch_set_memory_region(struct kvm *kvm, if (IS_ERR((void *)userspace_addr)) return PTR_ERR((void *)userspace_addr); - /* set userspace_addr atomically for kvm_hva_to_rmapp */ - spin_lock(&kvm->mmu_lock); memslot->userspace_addr = userspace_addr; - spin_unlock(&kvm->mmu_lock); - } else { - if (!old.user_alloc && old.rmap) { - int ret; - - down_write(¤t->mm->mmap_sem); - ret = do_munmap(current->mm, old.userspace_addr, - old.npages * PAGE_SIZE); - up_write(¤t->mm->mmap_sem); - if (ret < 0) - printk(KERN_WARNING - "kvm_vm_ioctl_set_memory_region: " - "failed to munmap memory\n"); - } } } + + return 0; +} + +void kvm_arch_commit_memory_region(struct kvm *kvm, + struct kvm_userspace_memory_region *mem, + struct kvm_memory_slot old, + int user_alloc) +{ + + int npages = mem->memory_size >> PAGE_SHIFT; + + if (!user_alloc && !old.user_alloc && old.rmap && !npages) { + int ret; + + down_write(¤t->mm->mmap_sem); + ret = do_munmap(current->mm, old.userspace_addr, + old.npages * PAGE_SIZE); + up_write(¤t->mm->mmap_sem); + if (ret < 0) + printk(KERN_WARNING + "kvm_vm_ioctl_set_memory_region: " + "failed to munmap memory\n"); + } + spin_lock(&kvm->mmu_lock); if (!kvm->arch.n_requested_mmu_pages) { unsigned int nr_mmu_pages = kvm_mmu_calculate_mmu_pages(kvm); @@ -5282,8 +5291,6 @@ int kvm_arch_set_memory_region(struct kvm *kvm, kvm_mmu_slot_remove_write_access(kvm, mem->slot); spin_unlock(&kvm->mmu_lock); - - return 0; } void kvm_arch_flush_shadow(struct kvm *kvm) diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 782bfb185f8..3c44687b342 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -253,7 +253,12 @@ int kvm_set_memory_region(struct kvm *kvm, int __kvm_set_memory_region(struct kvm *kvm, struct kvm_userspace_memory_region *mem, int user_alloc); -int kvm_arch_set_memory_region(struct kvm *kvm, +int kvm_arch_prepare_memory_region(struct kvm *kvm, + struct kvm_memory_slot *memslot, + struct kvm_memory_slot old, + struct kvm_userspace_memory_region *mem, + int user_alloc); +void kvm_arch_commit_memory_region(struct kvm *kvm, struct kvm_userspace_memory_region *mem, struct kvm_memory_slot old, int user_alloc); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 86dd8f3d29c..c9f6cfe8312 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 v1.2.3-70-g09d2 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 --- include/linux/kvm_host.h | 2 ++ virt/kvm/kvm_main.c | 33 +++++++++++++++++++++++++-------- 2 files changed, 27 insertions(+), 8 deletions(-) (limited to 'virt/kvm') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 3c44687b342..f1f78deece1 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -273,6 +273,8 @@ void kvm_set_page_dirty(struct page *page); void kvm_set_page_accessed(struct page *page); pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn); +pfn_t gfn_to_pfn_memslot(struct kvm *kvm, + struct kvm_memory_slot *slot, gfn_t gfn); void kvm_release_pfn_dirty(pfn_t); void kvm_release_pfn_clean(pfn_t pfn); void kvm_set_pfn_dirty(pfn_t pfn); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index c9f6cfe8312..4e2321c733f 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 v1.2.3-70-g09d2 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 --- include/linux/kvm_host.h | 3 +-- virt/kvm/iommu.c | 13 ++++++------- virt/kvm/kvm_main.c | 2 +- 3 files changed, 8 insertions(+), 10 deletions(-) (limited to 'virt/kvm') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index f1f78deece1..9af240387fe 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -440,8 +440,7 @@ void kvm_free_irq_source_id(struct kvm *kvm, int irq_source_id); #define KVM_IOMMU_CACHE_COHERENCY 0x1 #ifdef CONFIG_IOMMU_API -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); int kvm_iommu_map_guest(struct kvm *kvm); int kvm_iommu_unmap_guest(struct kvm *kvm); int kvm_assign_device(struct kvm *kvm, diff --git a/virt/kvm/iommu.c b/virt/kvm/iommu.c index bc697a66a88..cf567d8033d 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 4e2321c733f..87d296d8b27 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 v1.2.3-70-g09d2 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 --- arch/ia64/kvm/kvm-ia64.c | 4 +- arch/x86/kvm/mmu.c | 28 +++++----- arch/x86/kvm/vmx.c | 6 +- include/linux/kvm.h | 2 +- include/linux/kvm_host.h | 7 +-- virt/kvm/assigned-dev.c | 8 +-- virt/kvm/iommu.c | 4 +- virt/kvm/kvm_main.c | 141 +++++++++++++++++++++++++++++++++++------------ 8 files changed, 136 insertions(+), 64 deletions(-) (limited to 'virt/kvm') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index 0757c702798..b2e4d16dd39 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1382,7 +1382,7 @@ static void kvm_release_vm_pages(struct kvm *kvm) int i, j; unsigned long base_gfn; - slots = kvm->memslots; + slots = rcu_dereference(kvm->memslots); for (i = 0; i < slots->nmemslots; i++) { memslot = &slots->memslots[i]; base_gfn = memslot->base_gfn; @@ -1837,6 +1837,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot; int is_dirty = 0; + down_write(&kvm->slots_lock); spin_lock(&kvm->arch.dirty_log_lock); r = kvm_ia64_sync_dirty_log(kvm, log); @@ -1856,6 +1857,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, } r = 0; out: + up_write(&kvm->slots_lock); spin_unlock(&kvm->arch.dirty_log_lock); return r; } diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 81f84d326a8..f8bf42a2599 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -29,6 +29,7 @@ #include #include #include +#include #include #include @@ -807,21 +808,15 @@ static int kvm_handle_hva(struct kvm *kvm, unsigned long hva, { int i, j; int retval = 0; - struct kvm_memslots *slots = kvm->memslots; + struct kvm_memslots *slots; + + slots = rcu_dereference(kvm->memslots); - /* - * If mmap_sem isn't taken, we can look the memslots with only - * the mmu_lock by skipping over the slots with userspace_addr == 0. - */ for (i = 0; i < slots->nmemslots; i++) { struct kvm_memory_slot *memslot = &slots->memslots[i]; unsigned long start = memslot->userspace_addr; unsigned long end; - /* mmu_lock protects userspace_addr */ - if (!start) - continue; - end = start + (memslot->npages << PAGE_SHIFT); if (hva >= start && hva < end) { gfn_t gfn_offset = (hva - start) >> PAGE_SHIFT; @@ -1617,7 +1612,7 @@ static void mmu_unshadow(struct kvm *kvm, gfn_t gfn) static void page_header_update_slot(struct kvm *kvm, void *pte, gfn_t gfn) { - int slot = memslot_id(kvm, gfn_to_memslot(kvm, gfn)); + int slot = memslot_id(kvm, gfn); struct kvm_mmu_page *sp = page_header(__pa(pte)); __set_bit(slot, sp->slot_bitmap); @@ -3021,9 +3016,11 @@ unsigned int kvm_mmu_calculate_mmu_pages(struct kvm *kvm) int i; unsigned int nr_mmu_pages; unsigned int nr_pages = 0; + struct kvm_memslots *slots; - for (i = 0; i < kvm->memslots->nmemslots; i++) - nr_pages += kvm->memslots->memslots[i].npages; + slots = rcu_dereference(kvm->memslots); + for (i = 0; i < slots->nmemslots; i++) + nr_pages += slots->memslots[i].npages; nr_mmu_pages = nr_pages * KVM_PERMILLE_MMU_PAGES / 1000; nr_mmu_pages = max(nr_mmu_pages, @@ -3293,10 +3290,12 @@ static void audit_mappings(struct kvm_vcpu *vcpu) static int count_rmaps(struct kvm_vcpu *vcpu) { int nmaps = 0; - int i, j, k; + int i, j, k, idx; + idx = srcu_read_lock(&kvm->srcu); + slots = rcu_dereference(kvm->memslots); for (i = 0; i < KVM_MEMORY_SLOTS; ++i) { - struct kvm_memory_slot *m = &vcpu->kvm->memslots->memslots[i]; + struct kvm_memory_slot *m = &slots->memslots[i]; struct kvm_rmap_desc *d; for (j = 0; j < m->npages; ++j) { @@ -3319,6 +3318,7 @@ static int count_rmaps(struct kvm_vcpu *vcpu) } } } + srcu_read_unlock(&kvm->srcu, idx); return nmaps; } diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 18698799e36..f1cae7d6113 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -1503,7 +1503,11 @@ static void enter_pmode(struct kvm_vcpu *vcpu) static gva_t rmode_tss_base(struct kvm *kvm) { if (!kvm->arch.tss_addr) { - gfn_t base_gfn = kvm->memslots->memslots[0].base_gfn + + struct kvm_memslots *slots; + gfn_t base_gfn; + + slots = rcu_dereference(kvm->memslots); + base_gfn = kvm->memslots->memslots[0].base_gfn + kvm->memslots->memslots[0].npages - 3; return base_gfn << PAGE_SHIFT; } diff --git a/include/linux/kvm.h b/include/linux/kvm.h index a24de0b1858..f2feef68ffd 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -103,7 +103,7 @@ struct kvm_userspace_memory_region { /* for kvm_memory_region::flags */ #define KVM_MEM_LOG_DIRTY_PAGES 1UL - +#define KVM_MEMSLOT_INVALID (1UL << 1) /* for KVM_IRQ_LINE */ struct kvm_irq_level { diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 9af240387fe..93bd30701ca 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -162,6 +162,7 @@ struct kvm { struct rw_semaphore slots_lock; struct mm_struct *mm; /* userspace tied to this vm */ struct kvm_memslots *memslots; + struct srcu_struct srcu; #ifdef CONFIG_KVM_APIC_ARCHITECTURE u32 bsp_vcpu_id; struct kvm_vcpu *bsp_vcpu; @@ -275,6 +276,7 @@ void kvm_set_page_accessed(struct page *page); pfn_t gfn_to_pfn(struct kvm *kvm, gfn_t gfn); pfn_t gfn_to_pfn_memslot(struct kvm *kvm, struct kvm_memory_slot *slot, gfn_t gfn); +int memslot_id(struct kvm *kvm, gfn_t gfn); void kvm_release_pfn_dirty(pfn_t); void kvm_release_pfn_clean(pfn_t pfn); void kvm_set_pfn_dirty(pfn_t pfn); @@ -490,11 +492,6 @@ static inline void kvm_guest_exit(void) current->flags &= ~PF_VCPU; } -static inline int memslot_id(struct kvm *kvm, struct kvm_memory_slot *slot) -{ - return slot - kvm->memslots->memslots; -} - static inline gpa_t gfn_to_gpa(gfn_t gfn) { return (gpa_t)gfn << PAGE_SHIFT; diff --git a/virt/kvm/assigned-dev.c b/virt/kvm/assigned-dev.c index f73de631e3e..f51e684dd23 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 cf567d8033d..65a51432c8e 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 87d296d8b27..2bb24a814fd 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 v1.2.3-70-g09d2 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 --- arch/x86/include/asm/kvm_host.h | 4 +++ arch/x86/kvm/x86.c | 60 ++++++++++++++++++++++++++++++++++------- include/linux/kvm_host.h | 6 +++++ virt/kvm/kvm_main.c | 4 +-- 4 files changed, 63 insertions(+), 11 deletions(-) (limited to 'virt/kvm') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 7cdcb3d0f77..6c8c7c578c4 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -368,8 +368,12 @@ struct kvm_mem_alias { gfn_t base_gfn; unsigned long npages; gfn_t target_gfn; +#define KVM_ALIAS_INVALID 1UL + unsigned long flags; }; +#define KVM_ARCH_HAS_UNALIAS_INSTANTIATION + struct kvm_mem_aliases { struct kvm_mem_alias aliases[KVM_ALIAS_SLOTS]; int naliases; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index e7488350ca1..28127c936c3 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -38,6 +38,7 @@ #include #include #include +#include #include #undef TRACE_INCLUDE_FILE #define CREATE_TRACE_POINTS @@ -2223,11 +2224,32 @@ static int kvm_vm_ioctl_get_nr_mmu_pages(struct kvm *kvm) return kvm->arch.n_alloc_mmu_pages; } +gfn_t unalias_gfn_instantiation(struct kvm *kvm, gfn_t gfn) +{ + int i; + struct kvm_mem_alias *alias; + struct kvm_mem_aliases *aliases; + + aliases = rcu_dereference(kvm->arch.aliases); + + for (i = 0; i < aliases->naliases; ++i) { + alias = &aliases->aliases[i]; + if (alias->flags & KVM_ALIAS_INVALID) + continue; + if (gfn >= alias->base_gfn + && gfn < alias->base_gfn + alias->npages) + return alias->target_gfn + gfn - alias->base_gfn; + } + return gfn; +} + gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn) { int i; struct kvm_mem_alias *alias; - struct kvm_mem_aliases *aliases = kvm->arch.aliases; + struct kvm_mem_aliases *aliases; + + aliases = rcu_dereference(kvm->arch.aliases); for (i = 0; i < aliases->naliases; ++i) { alias = &aliases->aliases[i]; @@ -2248,7 +2270,7 @@ static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm, { int r, n; struct kvm_mem_alias *p; - struct kvm_mem_aliases *aliases; + struct kvm_mem_aliases *aliases, *old_aliases; r = -EINVAL; /* General sanity checks */ @@ -2265,28 +2287,48 @@ static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm, < alias->target_phys_addr) goto out; + r = -ENOMEM; + aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL); + if (!aliases) + goto out; + down_write(&kvm->slots_lock); - spin_lock(&kvm->mmu_lock); - aliases = kvm->arch.aliases; + /* invalidate any gfn reference in case of deletion/shrinking */ + memcpy(aliases, kvm->arch.aliases, sizeof(struct kvm_mem_aliases)); + aliases->aliases[alias->slot].flags |= KVM_ALIAS_INVALID; + old_aliases = kvm->arch.aliases; + rcu_assign_pointer(kvm->arch.aliases, aliases); + synchronize_srcu_expedited(&kvm->srcu); + kvm_mmu_zap_all(kvm); + kfree(old_aliases); + + r = -ENOMEM; + aliases = kzalloc(sizeof(struct kvm_mem_aliases), GFP_KERNEL); + if (!aliases) + goto out_unlock; + + memcpy(aliases, kvm->arch.aliases, sizeof(struct kvm_mem_aliases)); p = &aliases->aliases[alias->slot]; p->base_gfn = alias->guest_phys_addr >> PAGE_SHIFT; p->npages = alias->memory_size >> PAGE_SHIFT; p->target_gfn = alias->target_phys_addr >> PAGE_SHIFT; + p->flags &= ~(KVM_ALIAS_INVALID); for (n = KVM_ALIAS_SLOTS; n > 0; --n) if (aliases->aliases[n - 1].npages) break; aliases->naliases = n; - spin_unlock(&kvm->mmu_lock); - kvm_mmu_zap_all(kvm); + old_aliases = kvm->arch.aliases; + rcu_assign_pointer(kvm->arch.aliases, aliases); + synchronize_srcu_expedited(&kvm->srcu); + kfree(old_aliases); + r = 0; +out_unlock: up_write(&kvm->slots_lock); - - return 0; - out: return r; } diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 93bd30701ca..20941c0f404 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -266,6 +266,8 @@ void kvm_arch_commit_memory_region(struct kvm *kvm, void kvm_disable_largepages(void); void kvm_arch_flush_shadow(struct kvm *kvm); gfn_t unalias_gfn(struct kvm *kvm, gfn_t gfn); +gfn_t unalias_gfn_instantiation(struct kvm *kvm, gfn_t gfn); + struct page *gfn_to_page(struct kvm *kvm, gfn_t gfn); unsigned long gfn_to_hva(struct kvm *kvm, gfn_t gfn); void kvm_release_page_clean(struct page *page); @@ -539,6 +541,10 @@ static inline int mmu_notifier_retry(struct kvm_vcpu *vcpu, unsigned long mmu_se } #endif +#ifndef KVM_ARCH_HAS_UNALIAS_INSTANTIATION +#define unalias_gfn_instantiation unalias_gfn +#endif + #ifdef CONFIG_HAVE_KVM_IRQCHIP #define KVM_MAX_IRQ_ROUTES 1024 diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 2bb24a814fd..c680f7b64c6 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 v1.2.3-70-g09d2 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 --- arch/ia64/kvm/kvm-ia64.c | 4 +- arch/x86/kvm/i8254.c | 6 +-- arch/x86/kvm/i8259.c | 4 +- arch/x86/kvm/x86.c | 13 +++--- include/linux/kvm_host.h | 27 ++++++------ virt/kvm/coalesced_mmio.c | 4 +- virt/kvm/eventfd.c | 8 ++-- virt/kvm/ioapic.c | 4 +- virt/kvm/kvm_main.c | 106 +++++++++++++++++++++++++++------------------- 9 files changed, 101 insertions(+), 75 deletions(-) (limited to 'virt/kvm') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index b2e4d16dd39..d0ad538f008 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -241,10 +241,10 @@ static int handle_mmio(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 0; mmio: if (p->dir) - r = kvm_io_bus_read(&vcpu->kvm->mmio_bus, p->addr, + r = kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, p->addr, p->size, &p->data); else - r = kvm_io_bus_write(&vcpu->kvm->mmio_bus, p->addr, + r = kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, p->addr, p->size, &p->data); if (r) printk(KERN_ERR"kvm: No iodevice found! addr:%lx\n", p->addr); diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 15578f180e5..4b433de02e5 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -645,13 +645,13 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm, u32 flags) kvm_register_irq_mask_notifier(kvm, 0, &pit->mask_notifier); kvm_iodevice_init(&pit->dev, &pit_dev_ops); - ret = __kvm_io_bus_register_dev(&kvm->pio_bus, &pit->dev); + ret = kvm_io_bus_register_dev(kvm, KVM_PIO_BUS, &pit->dev); if (ret < 0) goto fail; if (flags & KVM_PIT_SPEAKER_DUMMY) { kvm_iodevice_init(&pit->speaker_dev, &speaker_dev_ops); - ret = __kvm_io_bus_register_dev(&kvm->pio_bus, + ret = kvm_io_bus_register_dev(kvm, KVM_PIO_BUS, &pit->speaker_dev); if (ret < 0) goto fail_unregister; @@ -660,7 +660,7 @@ struct kvm_pit *kvm_create_pit(struct kvm *kvm, u32 flags) return pit; fail_unregister: - __kvm_io_bus_unregister_dev(&kvm->pio_bus, &pit->dev); + kvm_io_bus_unregister_dev(kvm, KVM_PIO_BUS, &pit->dev); fail: if (pit->irq_source_id >= 0) diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index d057c0cbd24..b7d145b2095 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -533,7 +533,9 @@ struct kvm_pic *kvm_create_pic(struct kvm *kvm) * Initialize PIO device */ kvm_iodevice_init(&s->dev, &picdev_ops); - ret = kvm_io_bus_register_dev(kvm, &kvm->pio_bus, &s->dev); + down_write(&kvm->slots_lock); + ret = kvm_io_bus_register_dev(kvm, KVM_PIO_BUS, &s->dev); + up_write(&kvm->slots_lock); if (ret < 0) { kfree(s); return NULL; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 28127c936c3..9b42673df4a 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2851,7 +2851,7 @@ static int vcpu_mmio_write(struct kvm_vcpu *vcpu, gpa_t addr, int len, !kvm_iodevice_write(&vcpu->arch.apic->dev, addr, len, v)) return 0; - return kvm_io_bus_write(&vcpu->kvm->mmio_bus, addr, len, v); + return kvm_io_bus_write(vcpu->kvm, KVM_MMIO_BUS, addr, len, v); } static int vcpu_mmio_read(struct kvm_vcpu *vcpu, gpa_t addr, int len, void *v) @@ -2860,7 +2860,7 @@ static int vcpu_mmio_read(struct kvm_vcpu *vcpu, gpa_t addr, int len, void *v) !kvm_iodevice_read(&vcpu->arch.apic->dev, addr, len, v)) return 0; - return kvm_io_bus_read(&vcpu->kvm->mmio_bus, addr, len, v); + return kvm_io_bus_read(vcpu->kvm, KVM_MMIO_BUS, addr, len, v); } static int kvm_read_guest_virt(gva_t addr, void *val, unsigned int bytes, @@ -3345,11 +3345,12 @@ static int kernel_pio(struct kvm_vcpu *vcpu, void *pd) int r; if (vcpu->arch.pio.in) - r = kvm_io_bus_read(&vcpu->kvm->pio_bus, vcpu->arch.pio.port, + r = kvm_io_bus_read(vcpu->kvm, KVM_PIO_BUS, vcpu->arch.pio.port, vcpu->arch.pio.size, pd); else - r = kvm_io_bus_write(&vcpu->kvm->pio_bus, vcpu->arch.pio.port, - vcpu->arch.pio.size, pd); + r = kvm_io_bus_write(vcpu->kvm, KVM_PIO_BUS, + vcpu->arch.pio.port, vcpu->arch.pio.size, + pd); return r; } @@ -3360,7 +3361,7 @@ static int pio_string_write(struct kvm_vcpu *vcpu) int i, r = 0; for (i = 0; i < io->cur_count; i++) { - if (kvm_io_bus_write(&vcpu->kvm->pio_bus, + if (kvm_io_bus_write(vcpu->kvm, KVM_PIO_BUS, io->port, io->size, pd)) { r = -EOPNOTSUPP; break; diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 20941c0f404..5e9cb902550 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -57,20 +57,20 @@ struct kvm_io_bus { struct kvm_io_device *devs[NR_IOBUS_DEVS]; }; -void kvm_io_bus_init(struct kvm_io_bus *bus); -void kvm_io_bus_destroy(struct kvm_io_bus *bus); -int kvm_io_bus_write(struct kvm_io_bus *bus, gpa_t addr, int len, - const void *val); -int kvm_io_bus_read(struct kvm_io_bus *bus, gpa_t addr, int len, +enum kvm_bus { + KVM_MMIO_BUS, + KVM_PIO_BUS, + KVM_NR_BUSES +}; + +int kvm_io_bus_write(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, + int len, const void *val); +int kvm_io_bus_read(struct kvm *kvm, enum kvm_bus bus_idx, gpa_t addr, int len, void *val); -int __kvm_io_bus_register_dev(struct kvm_io_bus *bus, - struct kvm_io_device *dev); -int kvm_io_bus_register_dev(struct kvm *kvm, struct kvm_io_bus *bus, +int kvm_io_bus_register_dev(struct kvm *kvm, enum kvm_bus bus_idx, struct kvm_io_device *dev); -void __kvm_io_bus_unregister_dev(struct kvm_io_bus *bus, - struct kvm_io_device *dev); -void kvm_io_bus_unregister_dev(struct kvm *kvm, struct kvm_io_bus *bus, - struct kvm_io_device *dev); +int kvm_io_bus_unregister_dev(struct kvm *kvm, enum kvm_bus bus_idx, + struct kvm_io_device *dev); struct kvm_vcpu { struct kvm *kvm; @@ -171,8 +171,7 @@ struct kvm { atomic_t online_vcpus; struct list_head vm_list; struct mutex lock; - struct kvm_io_bus mmio_bus; - struct kvm_io_bus pio_bus; + struct kvm_io_bus *buses[KVM_NR_BUSES]; #ifdef CONFIG_HAVE_KVM_EVENTFD struct { spinlock_t lock; diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index d68e6c68e0f..a736a93ca7b 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 a9d3fc6c681..315a586ec4d 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 38a2d20b89d..f326a6f301c 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 c680f7b64c6..659bc12ad16 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 v1.2.3-70-g09d2 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 --- arch/ia64/kvm/kvm-ia64.c | 4 ++-- arch/powerpc/kvm/book3s.c | 4 ++-- arch/x86/kvm/i8254.c | 2 +- arch/x86/kvm/i8259.c | 4 ++-- arch/x86/kvm/vmx.c | 8 ++++---- arch/x86/kvm/x86.c | 16 ++++++++-------- include/linux/kvm_host.h | 2 +- virt/kvm/coalesced_mmio.c | 14 +++++++------- virt/kvm/eventfd.c | 10 +++++----- virt/kvm/ioapic.c | 4 ++-- virt/kvm/kvm_main.c | 10 +++++----- 11 files changed, 39 insertions(+), 39 deletions(-) (limited to 'virt/kvm') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index d5e38464127..e6ac549f8d5 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1834,7 +1834,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot; int is_dirty = 0; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); spin_lock(&kvm->arch.dirty_log_lock); r = kvm_ia64_sync_dirty_log(kvm, log); @@ -1854,7 +1854,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, } r = 0; out: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); spin_unlock(&kvm->arch.dirty_log_lock); return r; } diff --git a/arch/powerpc/kvm/book3s.c b/arch/powerpc/kvm/book3s.c index bb8873dcb20..492dcc198dd 100644 --- a/arch/powerpc/kvm/book3s.c +++ b/arch/powerpc/kvm/book3s.c @@ -857,7 +857,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, int is_dirty = 0; int r, n; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); r = kvm_get_dirty_log(kvm, log, &is_dirty); if (r) @@ -879,7 +879,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, r = 0; out: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return r; } diff --git a/arch/x86/kvm/i8254.c b/arch/x86/kvm/i8254.c index 4b433de02e5..6a74246f80c 100644 --- a/arch/x86/kvm/i8254.c +++ b/arch/x86/kvm/i8254.c @@ -605,7 +605,7 @@ static const struct kvm_io_device_ops speaker_dev_ops = { .write = speaker_ioport_write, }; -/* Caller must have writers lock on slots_lock */ +/* Caller must hold slots_lock */ struct kvm_pit *kvm_create_pit(struct kvm *kvm, u32 flags) { struct kvm_pit *pit; diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index b7d145b2095..d5753a75d58 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -533,9 +533,9 @@ struct kvm_pic *kvm_create_pic(struct kvm *kvm) * Initialize PIO device */ kvm_iodevice_init(&s->dev, &picdev_ops); - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); ret = kvm_io_bus_register_dev(kvm, KVM_PIO_BUS, &s->dev); - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); if (ret < 0) { kfree(s); return NULL; diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 22ab7137d1d..f04e2ff2138 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2223,7 +2223,7 @@ static int alloc_apic_access_page(struct kvm *kvm) struct kvm_userspace_memory_region kvm_userspace_mem; int r = 0; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); if (kvm->arch.apic_access_page) goto out; kvm_userspace_mem.slot = APIC_ACCESS_PAGE_PRIVATE_MEMSLOT; @@ -2236,7 +2236,7 @@ static int alloc_apic_access_page(struct kvm *kvm) kvm->arch.apic_access_page = gfn_to_page(kvm, 0xfee00); out: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return r; } @@ -2245,7 +2245,7 @@ static int alloc_identity_pagetable(struct kvm *kvm) struct kvm_userspace_memory_region kvm_userspace_mem; int r = 0; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); if (kvm->arch.ept_identity_pagetable) goto out; kvm_userspace_mem.slot = IDENTITY_PAGETABLE_PRIVATE_MEMSLOT; @@ -2260,7 +2260,7 @@ static int alloc_identity_pagetable(struct kvm *kvm) kvm->arch.ept_identity_pagetable = gfn_to_page(kvm, kvm->arch.ept_identity_map_addr >> PAGE_SHIFT); out: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return r; } diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 53bc06a6810..aff3479867a 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2208,14 +2208,14 @@ static int kvm_vm_ioctl_set_nr_mmu_pages(struct kvm *kvm, if (kvm_nr_mmu_pages < KVM_MIN_ALLOC_MMU_PAGES) return -EINVAL; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); spin_lock(&kvm->mmu_lock); kvm_mmu_change_mmu_pages(kvm, kvm_nr_mmu_pages); kvm->arch.n_requested_mmu_pages = kvm_nr_mmu_pages; spin_unlock(&kvm->mmu_lock); - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return 0; } @@ -2292,7 +2292,7 @@ static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm, if (!aliases) goto out; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); /* invalidate any gfn reference in case of deletion/shrinking */ memcpy(aliases, kvm->arch.aliases, sizeof(struct kvm_mem_aliases)); @@ -2328,7 +2328,7 @@ static int kvm_vm_ioctl_set_memory_alias(struct kvm *kvm, r = 0; out_unlock: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); out: return r; } @@ -2462,7 +2462,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, unsigned long is_dirty = 0; unsigned long *dirty_bitmap = NULL; - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); r = -EINVAL; if (log->slot >= KVM_MEMORY_SLOTS) @@ -2512,7 +2512,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, out_free: vfree(dirty_bitmap); out: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); return r; } @@ -2625,7 +2625,7 @@ long kvm_arch_vm_ioctl(struct file *filp, sizeof(struct kvm_pit_config))) goto out; create_pit: - down_write(&kvm->slots_lock); + mutex_lock(&kvm->slots_lock); r = -EEXIST; if (kvm->arch.vpit) goto create_pit_unlock; @@ -2634,7 +2634,7 @@ long kvm_arch_vm_ioctl(struct file *filp, if (kvm->arch.vpit) r = 0; create_pit_unlock: - up_write(&kvm->slots_lock); + mutex_unlock(&kvm->slots_lock); break; case KVM_IRQ_LINE_STATUS: case KVM_IRQ_LINE: { diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 0bb9aa295e6..bb0314ea926 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -161,7 +161,7 @@ struct kvm_memslots { struct kvm { spinlock_t mmu_lock; spinlock_t requests_lock; - struct rw_semaphore slots_lock; + struct mutex slots_lock; struct mm_struct *mm; /* userspace tied to this vm */ struct kvm_memslots *memslots; struct srcu_struct srcu; diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index a736a93ca7b..5de6594260c 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 315a586ec4d..486c604365d 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 f326a6f301c..f01392f51e8 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 659bc12ad16..2b7cd6c0d9c 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index f01392f51e8..a2edfd177fa 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 419c43b667a..a505ce9054f 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 v1.2.3-70-g09d2 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 --- arch/ia64/kvm/kvm-ia64.c | 1 + arch/powerpc/kvm/powerpc.c | 1 + arch/s390/kvm/kvm-s390.c | 1 + arch/x86/kvm/x86.c | 1 + virt/kvm/kvm_main.c | 1 - 5 files changed, 4 insertions(+), 1 deletion(-) (limited to 'virt/kvm') diff --git a/arch/ia64/kvm/kvm-ia64.c b/arch/ia64/kvm/kvm-ia64.c index e6ac549f8d5..06188988ed2 100644 --- a/arch/ia64/kvm/kvm-ia64.c +++ b/arch/ia64/kvm/kvm-ia64.c @@ -1404,6 +1404,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm) kfree(kvm->arch.vioapic); kvm_release_vm_pages(kvm); kvm_free_physmem(kvm); + cleanup_srcu_struct(&kvm->srcu); free_kvm(kvm); } diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 2c291161df8..51aedd7f16b 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -137,6 +137,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm) { kvmppc_free_vcpus(kvm); kvm_free_physmem(kvm); + cleanup_srcu_struct(&kvm->srcu); kfree(kvm); } diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c index c8002193d9d..49292869a5c 100644 --- a/arch/s390/kvm/kvm-s390.c +++ b/arch/s390/kvm/kvm-s390.c @@ -242,6 +242,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm) kvm_free_physmem(kvm); free_page((unsigned long)(kvm->arch.sca)); debug_unregister(kvm->arch.dbf); + cleanup_srcu_struct(&kvm->srcu); kfree(kvm); } diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 9b0758b4d55..322c2c5f9bc 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -5547,6 +5547,7 @@ void kvm_arch_destroy_vm(struct kvm *kvm) put_page(kvm->arch.apic_access_page); if (kvm->arch.ept_identity_pagetable) put_page(kvm->arch.ept_identity_pagetable); + cleanup_srcu_struct(&kvm->srcu); kfree(kvm->arch.aliases); kfree(kvm); } diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 2b7cd6c0d9c..7c5c873812f 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 7c5c873812f..2b0974a1483 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 v1.2.3-70-g09d2 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 --- arch/x86/kvm/x86.c | 1 + include/linux/kvm.h | 4 +++- include/linux/kvm_host.h | 1 + virt/kvm/assigned-dev.c | 4 +++- virt/kvm/iommu.c | 9 ++++++--- 5 files changed, 14 insertions(+), 5 deletions(-) (limited to 'virt/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index d47ceda7a92..0bf3df527af 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1569,6 +1569,7 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_HYPERV: case KVM_CAP_HYPERV_VAPIC: case KVM_CAP_HYPERV_SPIN: + case KVM_CAP_PCI_SEGMENT: r = 1; break; case KVM_CAP_COALESCED_MMIO: diff --git a/include/linux/kvm.h b/include/linux/kvm.h index 4c4937e7f65..dfa54be881f 100644 --- a/include/linux/kvm.h +++ b/include/linux/kvm.h @@ -500,6 +500,7 @@ struct kvm_ioeventfd { #define KVM_CAP_HYPERV 44 #define KVM_CAP_HYPERV_VAPIC 45 #define KVM_CAP_HYPERV_SPIN 46 +#define KVM_CAP_PCI_SEGMENT 47 #ifdef KVM_CAP_IRQ_ROUTING @@ -694,8 +695,9 @@ struct kvm_assigned_pci_dev { __u32 busnr; __u32 devfn; __u32 flags; + __u32 segnr; union { - __u32 reserved[12]; + __u32 reserved[11]; }; }; diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index dfde04b0d45..665c37063f3 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -400,6 +400,7 @@ struct kvm_assigned_dev_kernel { struct work_struct interrupt_work; struct list_head list; int assigned_dev_id; + int host_segnr; int host_busnr; int host_devfn; unsigned int entries_nr; diff --git a/virt/kvm/assigned-dev.c b/virt/kvm/assigned-dev.c index f51e684dd23..057e2cca6af 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 65a51432c8e..80fd3ad3b2d 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 v1.2.3-70-g09d2 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 --- arch/x86/kvm/mmu.c | 18 ++---------------- include/linux/kvm_host.h | 1 + virt/kvm/kvm_main.c | 25 +++++++++++++++++++++++++ 3 files changed, 28 insertions(+), 16 deletions(-) (limited to 'virt/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index dc4d954efac..913ef4b7939 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -468,24 +468,10 @@ static int has_wrprotected_page(struct kvm *kvm, static int host_mapping_level(struct kvm *kvm, gfn_t gfn) { - unsigned long page_size = PAGE_SIZE; - struct vm_area_struct *vma; - unsigned long addr; + unsigned long page_size; int i, ret = 0; - addr = gfn_to_hva(kvm, gfn); - if (kvm_is_error_hva(addr)) - return PT_PAGE_TABLE_LEVEL; - - down_read(¤t->mm->mmap_sem); - vma = find_vma(current->mm, addr); - if (!vma) - goto out; - - page_size = vma_kernel_pagesize(vma); - -out: - up_read(¤t->mm->mmap_sem); + page_size = kvm_host_page_size(kvm, gfn); for (i = PT_PAGE_TABLE_LEVEL; i < (PT_PAGE_TABLE_LEVEL + KVM_NR_PAGE_SIZES); ++i) { diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 665c37063f3..3145b281de9 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -300,6 +300,7 @@ int kvm_clear_guest_page(struct kvm *kvm, gfn_t gfn, int offset, int len); int kvm_clear_guest(struct kvm *kvm, gpa_t gpa, unsigned long len); struct kvm_memory_slot *gfn_to_memslot(struct kvm *kvm, gfn_t gfn); int kvm_is_visible_gfn(struct kvm *kvm, gfn_t gfn); +unsigned long kvm_host_page_size(struct kvm *kvm, gfn_t gfn); void mark_page_dirty(struct kvm *kvm, gfn_t gfn); void kvm_vcpu_block(struct kvm_vcpu *vcpu); diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 2b0974a1483..0a360c26cc3 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/coalesced_mmio.c b/virt/kvm/coalesced_mmio.c index 5de6594260c..5169736377a 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index a2edfd177fa..f3d06934ae6 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 v1.2.3-70-g09d2 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 --- arch/x86/kvm/i8259.c | 11 +++++++++++ arch/x86/kvm/irq.h | 1 + arch/x86/kvm/x86.c | 8 ++++---- virt/kvm/ioapic.c | 11 +++++++++++ virt/kvm/ioapic.h | 1 + 5 files changed, 28 insertions(+), 4 deletions(-) (limited to 'virt/kvm') diff --git a/arch/x86/kvm/i8259.c b/arch/x86/kvm/i8259.c index d5753a75d58..a3711f9e580 100644 --- a/arch/x86/kvm/i8259.c +++ b/arch/x86/kvm/i8259.c @@ -543,3 +543,14 @@ struct kvm_pic *kvm_create_pic(struct kvm *kvm) return s; } + +void kvm_destroy_pic(struct kvm *kvm) +{ + struct kvm_pic *vpic = kvm->arch.vpic; + + if (vpic) { + kvm_io_bus_unregister_dev(kvm, KVM_PIO_BUS, &vpic->dev); + kvm->arch.vpic = NULL; + kfree(vpic); + } +} diff --git a/arch/x86/kvm/irq.h b/arch/x86/kvm/irq.h index be399e207d5..0b71d480ebf 100644 --- a/arch/x86/kvm/irq.h +++ b/arch/x86/kvm/irq.h @@ -75,6 +75,7 @@ struct kvm_pic { }; struct kvm_pic *kvm_create_pic(struct kvm *kvm); +void kvm_destroy_pic(struct kvm *kvm); int kvm_pic_read_irq(struct kvm *kvm); void kvm_pic_update_irq(struct kvm_pic *s); void kvm_pic_clear_isr_ack(struct kvm *kvm); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index bd3161c6dae..b2f91b9af00 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2771,6 +2771,8 @@ long kvm_arch_vm_ioctl(struct file *filp, if (vpic) { r = kvm_ioapic_init(kvm); if (r) { + kvm_io_bus_unregister_dev(kvm, KVM_PIO_BUS, + &vpic->dev); kfree(vpic); goto create_irqchip_unlock; } @@ -2782,10 +2784,8 @@ long kvm_arch_vm_ioctl(struct file *filp, r = kvm_setup_default_irq_routing(kvm); if (r) { mutex_lock(&kvm->irq_lock); - kfree(kvm->arch.vpic); - kfree(kvm->arch.vioapic); - kvm->arch.vpic = NULL; - kvm->arch.vioapic = NULL; + kvm_ioapic_destroy(kvm); + kvm_destroy_pic(kvm); mutex_unlock(&kvm->irq_lock); } create_irqchip_unlock: diff --git a/virt/kvm/ioapic.c b/virt/kvm/ioapic.c index f3d06934ae6..3db15a807f8 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 a505ce9054f..8a751b78a43 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 v1.2.3-70-g09d2 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/kvm') diff --git a/virt/kvm/eventfd.c b/virt/kvm/eventfd.c index 486c604365d..7016319b1ec 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 v1.2.3-70-g09d2 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 --- include/linux/kvm_host.h | 2 +- virt/kvm/kvm_main.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) (limited to 'virt/kvm') diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 3145b281de9..a3fd0f91d94 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -161,7 +161,7 @@ struct kvm_memslots { struct kvm { spinlock_t mmu_lock; - spinlock_t requests_lock; + raw_spinlock_t requests_lock; struct mutex slots_lock; struct mm_struct *mm; /* userspace tied to this vm */ struct kvm_memslots *memslots; diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 0a360c26cc3..548f9253c19 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 v1.2.3-70-g09d2