From b3ae2096974b12c3af2ad1a4e7716b084949867f Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Mon, 2 Jul 2012 17:56:33 +0900 Subject: KVM: Introduce kvm_unmap_hva_range() for kvm_mmu_notifier_invalidate_range_start() When we tested KVM under memory pressure, with THP enabled on the host, we noticed that MMU notifier took a long time to invalidate huge pages. Since the invalidation was done with mmu_lock held, it not only wasted the CPU but also made the host harder to respond. This patch mitigates this by using kvm_handle_hva_range(). Signed-off-by: Takuya Yoshikawa Cc: Alexander Graf Cc: Paul Mackerras Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index a3e9409e90b6..d4aab865606c 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -944,6 +944,7 @@ extern bool kvm_rebooting; #define KVM_ARCH_WANT_MMU_NOTIFIER int kvm_unmap_hva(struct kvm *kvm, unsigned long hva); +int kvm_unmap_hva_range(struct kvm *kvm, unsigned long start, unsigned long end); int kvm_age_hva(struct kvm *kvm, unsigned long hva); int kvm_test_age_hva(struct kvm *kvm, unsigned long hva); void kvm_set_spte_hva(struct kvm *kvm, unsigned long hva, pte_t pte); -- cgit v1.2.3 From 77d11309b3a10e1ce112058ec2c9b7b979bcf311 Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Mon, 2 Jul 2012 17:57:17 +0900 Subject: KVM: Separate rmap_pde from kvm_lpage_info->write_count This makes it possible to loop over rmap_pde arrays in the same way as we do over rmap so that we can optimize kvm_handle_hva_range() easily in the following patch. Signed-off-by: Takuya Yoshikawa Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kvm/mmu.c | 6 +++--- arch/x86/kvm/x86.c | 11 +++++++++++ 3 files changed, 15 insertions(+), 4 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index d4aab865606c..4f98da9243fc 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -500,11 +500,11 @@ struct kvm_vcpu_arch { }; struct kvm_lpage_info { - unsigned long rmap_pde; int write_count; }; struct kvm_arch_memory_slot { + unsigned long *rmap_pde[KVM_NR_PAGE_SIZES - 1]; struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1]; }; diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index d2855f895fde..3b3f5ae5da6a 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -960,13 +960,13 @@ static void pte_list_walk(unsigned long *pte_list, pte_list_walk_fn fn) static unsigned long *__gfn_to_rmap(gfn_t gfn, int level, struct kvm_memory_slot *slot) { - struct kvm_lpage_info *linfo; + unsigned long idx; if (likely(level == PT_PAGE_TABLE_LEVEL)) return &slot->rmap[gfn - slot->base_gfn]; - linfo = lpage_info_slot(gfn, slot, level); - return &linfo->rmap_pde; + idx = gfn_to_index(gfn, slot->base_gfn, level); + return &slot->arch.rmap_pde[level - PT_DIRECTORY_LEVEL][idx]; } /* diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 59b59508ff07..829b4e972558 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6314,6 +6314,10 @@ void kvm_arch_free_memslot(struct kvm_memory_slot *free, int i; for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) { + if (!dont || free->arch.rmap_pde[i] != dont->arch.rmap_pde[i]) { + kvm_kvfree(free->arch.rmap_pde[i]); + free->arch.rmap_pde[i] = NULL; + } if (!dont || free->arch.lpage_info[i] != dont->arch.lpage_info[i]) { kvm_kvfree(free->arch.lpage_info[i]); free->arch.lpage_info[i] = NULL; @@ -6333,6 +6337,11 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages) lpages = gfn_to_index(slot->base_gfn + npages - 1, slot->base_gfn, level) + 1; + slot->arch.rmap_pde[i] = + kvm_kvzalloc(lpages * sizeof(*slot->arch.rmap_pde[i])); + if (!slot->arch.rmap_pde[i]) + goto out_free; + slot->arch.lpage_info[i] = kvm_kvzalloc(lpages * sizeof(*slot->arch.lpage_info[i])); if (!slot->arch.lpage_info[i]) @@ -6361,7 +6370,9 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages) out_free: for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) { + kvm_kvfree(slot->arch.rmap_pde[i]); kvm_kvfree(slot->arch.lpage_info[i]); + slot->arch.rmap_pde[i] = NULL; slot->arch.lpage_info[i] = NULL; } return -ENOMEM; -- cgit v1.2.3 From d89cc617b954aff4030fce178f7d86f59aaf713d Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Wed, 1 Aug 2012 18:03:28 +0900 Subject: KVM: Push rmap into kvm_arch_memory_slot Two reasons: - x86 can integrate rmap and rmap_pde and remove heuristics in __gfn_to_rmap(). - Some architectures do not need rmap. Since rmap is one of the most memory consuming stuff in KVM, ppc'd better restrict the allocation to Book3S HV. Signed-off-by: Takuya Yoshikawa Acked-by: Paul Mackerras Signed-off-by: Avi Kivity --- arch/powerpc/include/asm/kvm_host.h | 1 + arch/powerpc/kvm/book3s_64_mmu_hv.c | 6 ++-- arch/powerpc/kvm/book3s_hv_rm_mmu.c | 4 +-- arch/powerpc/kvm/powerpc.c | 8 ++++++ arch/x86/include/asm/kvm_host.h | 2 +- arch/x86/kvm/mmu.c | 5 +--- arch/x86/kvm/x86.c | 55 +++++++++++++++++++++---------------- include/linux/kvm_host.h | 1 - virt/kvm/kvm_main.c | 11 +------- 9 files changed, 49 insertions(+), 44 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/powerpc/include/asm/kvm_host.h b/arch/powerpc/include/asm/kvm_host.h index 572ad0141268..a29e0918172a 100644 --- a/arch/powerpc/include/asm/kvm_host.h +++ b/arch/powerpc/include/asm/kvm_host.h @@ -221,6 +221,7 @@ struct revmap_entry { #define KVMPPC_GOT_PAGE 0x80 struct kvm_arch_memory_slot { + unsigned long *rmap; }; struct kvm_arch { diff --git a/arch/powerpc/kvm/book3s_64_mmu_hv.c b/arch/powerpc/kvm/book3s_64_mmu_hv.c index 3c635c0616b0..d95d11322a15 100644 --- a/arch/powerpc/kvm/book3s_64_mmu_hv.c +++ b/arch/powerpc/kvm/book3s_64_mmu_hv.c @@ -705,7 +705,7 @@ int kvmppc_book3s_hv_page_fault(struct kvm_run *run, struct kvm_vcpu *vcpu, goto out_unlock; hpte[0] = (hpte[0] & ~HPTE_V_ABSENT) | HPTE_V_VALID; - rmap = &memslot->rmap[gfn - memslot->base_gfn]; + rmap = &memslot->arch.rmap[gfn - memslot->base_gfn]; lock_rmap(rmap); /* Check if we might have been invalidated; let the guest retry if so */ @@ -788,7 +788,7 @@ static int kvm_handle_hva_range(struct kvm *kvm, for (; gfn < gfn_end; ++gfn) { gfn_t gfn_offset = gfn - memslot->base_gfn; - ret = handler(kvm, &memslot->rmap[gfn_offset], gfn); + ret = handler(kvm, &memslot->arch.rmap[gfn_offset], gfn); retval |= ret; } } @@ -1036,7 +1036,7 @@ long kvmppc_hv_get_dirty_log(struct kvm *kvm, struct kvm_memory_slot *memslot) unsigned long *rmapp, *map; preempt_disable(); - rmapp = memslot->rmap; + rmapp = memslot->arch.rmap; map = memslot->dirty_bitmap; for (i = 0; i < memslot->npages; ++i) { if (kvm_test_clear_dirty(kvm, rmapp)) diff --git a/arch/powerpc/kvm/book3s_hv_rm_mmu.c b/arch/powerpc/kvm/book3s_hv_rm_mmu.c index 5c70d19494f9..56ac1a5d9912 100644 --- a/arch/powerpc/kvm/book3s_hv_rm_mmu.c +++ b/arch/powerpc/kvm/book3s_hv_rm_mmu.c @@ -84,7 +84,7 @@ static void remove_revmap_chain(struct kvm *kvm, long pte_index, if (!memslot || (memslot->flags & KVM_MEMSLOT_INVALID)) return; - rmap = real_vmalloc_addr(&memslot->rmap[gfn - memslot->base_gfn]); + rmap = real_vmalloc_addr(&memslot->arch.rmap[gfn - memslot->base_gfn]); lock_rmap(rmap); head = *rmap & KVMPPC_RMAP_INDEX; @@ -180,7 +180,7 @@ long kvmppc_h_enter(struct kvm_vcpu *vcpu, unsigned long flags, if (!slot_is_aligned(memslot, psize)) return H_PARAMETER; slot_fn = gfn - memslot->base_gfn; - rmap = &memslot->rmap[slot_fn]; + rmap = &memslot->arch.rmap[slot_fn]; if (!kvm->arch.using_mmu_notifiers) { physp = kvm->arch.slot_phys[memslot->id]; diff --git a/arch/powerpc/kvm/powerpc.c b/arch/powerpc/kvm/powerpc.c index 87f4dc886076..879b14a61403 100644 --- a/arch/powerpc/kvm/powerpc.c +++ b/arch/powerpc/kvm/powerpc.c @@ -302,10 +302,18 @@ long kvm_arch_dev_ioctl(struct file *filp, void kvm_arch_free_memslot(struct kvm_memory_slot *free, struct kvm_memory_slot *dont) { + if (!dont || free->arch.rmap != dont->arch.rmap) { + vfree(free->arch.rmap); + free->arch.rmap = NULL; + } } int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages) { + slot->arch.rmap = vzalloc(npages * sizeof(*slot->arch.rmap)); + if (!slot->arch.rmap) + return -ENOMEM; + return 0; } diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 48e713188469..1309e69b57fa 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -504,7 +504,7 @@ struct kvm_lpage_info { }; struct kvm_arch_memory_slot { - unsigned long *rmap_pde[KVM_NR_PAGE_SIZES - 1]; + unsigned long *rmap[KVM_NR_PAGE_SIZES]; struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1]; }; diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index ee768bb2367f..aa9a987ddefb 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -970,11 +970,8 @@ static unsigned long *__gfn_to_rmap(gfn_t gfn, int level, { unsigned long idx; - if (likely(level == PT_PAGE_TABLE_LEVEL)) - return &slot->rmap[gfn - slot->base_gfn]; - idx = gfn_to_index(gfn, slot->base_gfn, level); - return &slot->arch.rmap_pde[level - PT_DIRECTORY_LEVEL][idx]; + return &slot->arch.rmap[level - PT_PAGE_TABLE_LEVEL][idx]; } /* diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index abc039d78428..ebf2109318e0 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6303,14 +6303,18 @@ void kvm_arch_free_memslot(struct kvm_memory_slot *free, { int i; - for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) { - if (!dont || free->arch.rmap_pde[i] != dont->arch.rmap_pde[i]) { - kvm_kvfree(free->arch.rmap_pde[i]); - free->arch.rmap_pde[i] = NULL; + for (i = 0; i < KVM_NR_PAGE_SIZES; ++i) { + if (!dont || free->arch.rmap[i] != dont->arch.rmap[i]) { + kvm_kvfree(free->arch.rmap[i]); + free->arch.rmap[i] = NULL; } - if (!dont || free->arch.lpage_info[i] != dont->arch.lpage_info[i]) { - kvm_kvfree(free->arch.lpage_info[i]); - free->arch.lpage_info[i] = NULL; + if (i == 0) + continue; + + if (!dont || free->arch.lpage_info[i - 1] != + dont->arch.lpage_info[i - 1]) { + kvm_kvfree(free->arch.lpage_info[i - 1]); + free->arch.lpage_info[i - 1] = NULL; } } } @@ -6319,28 +6323,30 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages) { int i; - for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) { + for (i = 0; i < KVM_NR_PAGE_SIZES; ++i) { unsigned long ugfn; int lpages; - int level = i + 2; + int level = i + 1; lpages = gfn_to_index(slot->base_gfn + npages - 1, slot->base_gfn, level) + 1; - slot->arch.rmap_pde[i] = - kvm_kvzalloc(lpages * sizeof(*slot->arch.rmap_pde[i])); - if (!slot->arch.rmap_pde[i]) + slot->arch.rmap[i] = + kvm_kvzalloc(lpages * sizeof(*slot->arch.rmap[i])); + if (!slot->arch.rmap[i]) goto out_free; + if (i == 0) + continue; - slot->arch.lpage_info[i] = - kvm_kvzalloc(lpages * sizeof(*slot->arch.lpage_info[i])); - if (!slot->arch.lpage_info[i]) + slot->arch.lpage_info[i - 1] = kvm_kvzalloc(lpages * + sizeof(*slot->arch.lpage_info[i - 1])); + if (!slot->arch.lpage_info[i - 1]) goto out_free; if (slot->base_gfn & (KVM_PAGES_PER_HPAGE(level) - 1)) - slot->arch.lpage_info[i][0].write_count = 1; + slot->arch.lpage_info[i - 1][0].write_count = 1; if ((slot->base_gfn + npages) & (KVM_PAGES_PER_HPAGE(level) - 1)) - slot->arch.lpage_info[i][lpages - 1].write_count = 1; + slot->arch.lpage_info[i - 1][lpages - 1].write_count = 1; ugfn = slot->userspace_addr >> PAGE_SHIFT; /* * If the gfn and userspace address are not aligned wrt each @@ -6352,18 +6358,21 @@ int kvm_arch_create_memslot(struct kvm_memory_slot *slot, unsigned long npages) unsigned long j; for (j = 0; j < lpages; ++j) - slot->arch.lpage_info[i][j].write_count = 1; + slot->arch.lpage_info[i - 1][j].write_count = 1; } } return 0; out_free: - for (i = 0; i < KVM_NR_PAGE_SIZES - 1; ++i) { - kvm_kvfree(slot->arch.rmap_pde[i]); - kvm_kvfree(slot->arch.lpage_info[i]); - slot->arch.rmap_pde[i] = NULL; - slot->arch.lpage_info[i] = NULL; + for (i = 0; i < KVM_NR_PAGE_SIZES; ++i) { + kvm_kvfree(slot->arch.rmap[i]); + slot->arch.rmap[i] = NULL; + if (i == 0) + continue; + + kvm_kvfree(slot->arch.lpage_info[i - 1]); + slot->arch.lpage_info[i - 1] = NULL; } return -ENOMEM; } diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index dbc65f9d6a2b..3c16f0f1fe35 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -213,7 +213,6 @@ struct kvm_memory_slot { gfn_t base_gfn; unsigned long npages; unsigned long flags; - unsigned long *rmap; unsigned long *dirty_bitmap; struct kvm_arch_memory_slot arch; unsigned long userspace_addr; diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index bcf973ec98ff..14ec567816ab 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -550,16 +550,12 @@ static void kvm_destroy_dirty_bitmap(struct kvm_memory_slot *memslot) static void kvm_free_physmem_slot(struct kvm_memory_slot *free, struct kvm_memory_slot *dont) { - if (!dont || free->rmap != dont->rmap) - vfree(free->rmap); - if (!dont || free->dirty_bitmap != dont->dirty_bitmap) kvm_destroy_dirty_bitmap(free); kvm_arch_free_memslot(free, dont); free->npages = 0; - free->rmap = NULL; } void kvm_free_physmem(struct kvm *kvm) @@ -768,11 +764,7 @@ int __kvm_set_memory_region(struct kvm *kvm, if (npages && !old.npages) { new.user_alloc = user_alloc; new.userspace_addr = mem->userspace_addr; -#ifndef CONFIG_S390 - new.rmap = vzalloc(npages * sizeof(*new.rmap)); - if (!new.rmap) - goto out_free; -#endif /* not defined CONFIG_S390 */ + if (kvm_arch_create_memslot(&new, npages)) goto out_free; } @@ -831,7 +823,6 @@ int __kvm_set_memory_region(struct kvm *kvm, /* actual memory is freed via old in kvm_free_physmem_slot below */ if (!npages) { - new.rmap = NULL; new.dirty_bitmap = NULL; memset(&new.arch, 0, sizeof(new.arch)); } -- cgit v1.2.3 From 51d59c6b422f3f95940ae4e5b42f165595906aee Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 3 Aug 2012 15:57:49 -0300 Subject: KVM: x86: fix pvclock guest stopped flag reporting kvm_guest_time_update unconditionally clears hv_clock.flags field, so the notification never reaches the guest. Fix it by allowing PVCLOCK_GUEST_STOPPED to passthrough. Reviewed-by: Eric B Munson Reviewed-by: Amit Shah Signed-off-by: Marcelo Tosatti --- arch/x86/include/asm/kvm_host.h | 2 ++ arch/x86/kvm/x86.c | 13 ++++++++++--- 2 files changed, 12 insertions(+), 3 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 1309e69b57fa..fc0e752e7564 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -420,6 +420,8 @@ struct kvm_vcpu_arch { unsigned int hw_tsc_khz; unsigned int time_offset; struct page *time_page; + /* set guest stopped flag in pvclock flags field */ + bool pvclock_set_guest_stopped_request; struct { u64 msr_val; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 91a595827deb..fb0d93788bfb 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1134,6 +1134,7 @@ static int kvm_guest_time_update(struct kvm_vcpu *v) unsigned long this_tsc_khz; s64 kernel_ns, max_kernel_ns; u64 tsc_timestamp; + u8 pvclock_flags; /* Keep irq disabled to prevent changes to the clock */ local_irq_save(flags); @@ -1215,7 +1216,14 @@ static int kvm_guest_time_update(struct kvm_vcpu *v) vcpu->hv_clock.system_time = kernel_ns + v->kvm->arch.kvmclock_offset; vcpu->last_kernel_ns = kernel_ns; vcpu->last_guest_tsc = tsc_timestamp; - vcpu->hv_clock.flags = 0; + + pvclock_flags = 0; + if (vcpu->pvclock_set_guest_stopped_request) { + pvclock_flags |= PVCLOCK_GUEST_STOPPED; + vcpu->pvclock_set_guest_stopped_request = false; + } + + vcpu->hv_clock.flags = pvclock_flags; /* * The interface expects us to write an even number signaling that the @@ -2624,10 +2632,9 @@ static int kvm_vcpu_ioctl_x86_set_xcrs(struct kvm_vcpu *vcpu, */ static int kvm_set_guest_paused(struct kvm_vcpu *vcpu) { - struct pvclock_vcpu_time_info *src = &vcpu->arch.hv_clock; if (!vcpu->arch.time_page) return -EINVAL; - src->flags |= PVCLOCK_GUEST_STOPPED; + vcpu->arch.pvclock_set_guest_stopped_request = true; kvm_make_request(KVM_REQ_CLOCK_UPDATE, vcpu); return 0; } -- cgit v1.2.3 From 716d51abff06f48425cef15d78ca6f36093f6dbf Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Mon, 3 Sep 2012 15:24:26 +0300 Subject: KVM: Provide userspace IO exit completion callback Current code assumes that IO exit was due to instruction emulation and handles execution back to emulator directly. This patch adds new userspace IO exit completion callback that can be set by any other code that caused IO exit to userspace. Signed-off-by: Gleb Natapov Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/x86.c | 93 +++++++++++++++++++++++++---------------- 2 files changed, 57 insertions(+), 37 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index fc0e752e7564..64adb6117e19 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -414,6 +414,7 @@ struct kvm_vcpu_arch { struct x86_emulate_ctxt emulate_ctxt; bool emulate_regs_need_sync_to_vcpu; bool emulate_regs_need_sync_from_vcpu; + int (*complete_userspace_io)(struct kvm_vcpu *vcpu); gpa_t time; struct pvclock_vcpu_time_info hv_clock; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 6f6812ec8419..f91e2c9d7cb1 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4544,6 +4544,9 @@ static bool retry_instruction(struct x86_emulate_ctxt *ctxt, return true; } +static int complete_emulated_mmio(struct kvm_vcpu *vcpu); +static int complete_emulated_pio(struct kvm_vcpu *vcpu); + int x86_emulate_instruction(struct kvm_vcpu *vcpu, unsigned long cr2, int emulation_type, @@ -4614,13 +4617,16 @@ restart: } else if (vcpu->arch.pio.count) { if (!vcpu->arch.pio.in) vcpu->arch.pio.count = 0; - else + else { writeback = false; + vcpu->arch.complete_userspace_io = complete_emulated_pio; + } r = EMULATE_DO_MMIO; } else if (vcpu->mmio_needed) { if (!vcpu->mmio_is_write) writeback = false; r = EMULATE_DO_MMIO; + vcpu->arch.complete_userspace_io = complete_emulated_mmio; } else if (r == EMULATION_RESTART) goto restart; else @@ -5476,6 +5482,24 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) return r; } +static inline int complete_emulated_io(struct kvm_vcpu *vcpu) +{ + int r; + vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); + r = emulate_instruction(vcpu, EMULTYPE_NO_DECODE); + srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); + if (r != EMULATE_DONE) + return 0; + return 1; +} + +static int complete_emulated_pio(struct kvm_vcpu *vcpu) +{ + BUG_ON(!vcpu->arch.pio.count); + + return complete_emulated_io(vcpu); +} + /* * Implements the following, as a state machine: * @@ -5492,47 +5516,37 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) * copy data * exit */ -static int complete_mmio(struct kvm_vcpu *vcpu) +static int complete_emulated_mmio(struct kvm_vcpu *vcpu) { struct kvm_run *run = vcpu->run; struct kvm_mmio_fragment *frag; - int r; - if (!(vcpu->arch.pio.count || vcpu->mmio_needed)) - return 1; + BUG_ON(!vcpu->mmio_needed); - if (vcpu->mmio_needed) { - /* Complete previous fragment */ - frag = &vcpu->mmio_fragments[vcpu->mmio_cur_fragment++]; - if (!vcpu->mmio_is_write) - memcpy(frag->data, run->mmio.data, frag->len); - if (vcpu->mmio_cur_fragment == vcpu->mmio_nr_fragments) { - vcpu->mmio_needed = 0; - if (vcpu->mmio_is_write) - return 1; - vcpu->mmio_read_completed = 1; - goto done; - } - /* Initiate next fragment */ - ++frag; - run->exit_reason = KVM_EXIT_MMIO; - run->mmio.phys_addr = frag->gpa; + /* Complete previous fragment */ + frag = &vcpu->mmio_fragments[vcpu->mmio_cur_fragment++]; + if (!vcpu->mmio_is_write) + memcpy(frag->data, run->mmio.data, frag->len); + if (vcpu->mmio_cur_fragment == vcpu->mmio_nr_fragments) { + vcpu->mmio_needed = 0; if (vcpu->mmio_is_write) - memcpy(run->mmio.data, frag->data, frag->len); - run->mmio.len = frag->len; - run->mmio.is_write = vcpu->mmio_is_write; - return 0; - - } -done: - vcpu->srcu_idx = srcu_read_lock(&vcpu->kvm->srcu); - r = emulate_instruction(vcpu, EMULTYPE_NO_DECODE); - srcu_read_unlock(&vcpu->kvm->srcu, vcpu->srcu_idx); - if (r != EMULATE_DONE) - return 0; - return 1; + return 1; + vcpu->mmio_read_completed = 1; + return complete_emulated_io(vcpu); + } + /* Initiate next fragment */ + ++frag; + run->exit_reason = KVM_EXIT_MMIO; + run->mmio.phys_addr = frag->gpa; + if (vcpu->mmio_is_write) + memcpy(run->mmio.data, frag->data, frag->len); + run->mmio.len = frag->len; + run->mmio.is_write = vcpu->mmio_is_write; + vcpu->arch.complete_userspace_io = complete_emulated_mmio; + return 0; } + int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) { int r; @@ -5559,9 +5573,14 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) } } - r = complete_mmio(vcpu); - if (r <= 0) - goto out; + if (unlikely(vcpu->arch.complete_userspace_io)) { + int (*cui)(struct kvm_vcpu *) = vcpu->arch.complete_userspace_io; + vcpu->arch.complete_userspace_io = NULL; + r = cui(vcpu); + if (r <= 0) + goto out; + } else + WARN_ON(vcpu->arch.pio.count || vcpu->mmio_needed); r = __vcpu_run(vcpu); -- cgit v1.2.3 From 97d64b788114be1c4dc4bfe7a8ba2bf9643fe6af Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Wed, 12 Sep 2012 14:52:00 +0300 Subject: KVM: MMU: Optimize pte permission checks walk_addr_generic() permission checks are a maze of branchy code, which is performed four times per lookup. It depends on the type of access, efer.nxe, cr0.wp, cr4.smep, and in the near future, cr4.smap. Optimize this away by precalculating all variants and storing them in a bitmap. The bitmap is recalculated when rarely-changing variables change (cr0, cr4) and is indexed by the often-changing variables (page fault error code, pte access permissions). The permission check is moved to the end of the loop, otherwise an SMEP fault could be reported as a false positive, when PDE.U=1 but PTE.U=0. Noted by Xiao Guangrong. The result is short, branch-free code. Reviewed-by: Xiao Guangrong Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 7 +++++++ arch/x86/kvm/mmu.c | 38 ++++++++++++++++++++++++++++++++++++++ arch/x86/kvm/mmu.h | 19 ++++++++----------- arch/x86/kvm/paging_tmpl.h | 22 ++++------------------ arch/x86/kvm/x86.c | 11 ++++------- 5 files changed, 61 insertions(+), 36 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 64adb6117e19..3318bde206a5 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -287,6 +287,13 @@ struct kvm_mmu { union kvm_mmu_page_role base_role; bool direct_map; + /* + * Bitmap; bit set = permission fault + * Byte index: page fault error code [4:1] + * Bit index: pte permissions in ACC_* format + */ + u8 permissions[16]; + u64 *pae_root; u64 *lm_root; u64 rsvd_bits_mask[2][4]; diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index f297a2ccf4f6..9c6188931f87 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -3516,6 +3516,38 @@ static void reset_rsvds_bits_mask(struct kvm_vcpu *vcpu, } } +static void update_permission_bitmask(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu) +{ + unsigned bit, byte, pfec; + u8 map; + bool fault, x, w, u, wf, uf, ff, smep; + + smep = kvm_read_cr4_bits(vcpu, X86_CR4_SMEP); + for (byte = 0; byte < ARRAY_SIZE(mmu->permissions); ++byte) { + pfec = byte << 1; + map = 0; + wf = pfec & PFERR_WRITE_MASK; + uf = pfec & PFERR_USER_MASK; + ff = pfec & PFERR_FETCH_MASK; + for (bit = 0; bit < 8; ++bit) { + x = bit & ACC_EXEC_MASK; + w = bit & ACC_WRITE_MASK; + u = bit & ACC_USER_MASK; + + /* Not really needed: !nx will cause pte.nx to fault */ + x |= !mmu->nx; + /* Allow supervisor writes if !cr0.wp */ + w |= !is_write_protection(vcpu) && !uf; + /* Disallow supervisor fetches of user code if cr4.smep */ + x &= !(smep && u && !uf); + + fault = (ff && !x) || (uf && !u) || (wf && !w); + map |= fault << bit; + } + mmu->permissions[byte] = map; + } +} + static int paging64_init_context_common(struct kvm_vcpu *vcpu, struct kvm_mmu *context, int level) @@ -3524,6 +3556,7 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu, context->root_level = level; reset_rsvds_bits_mask(vcpu, context); + update_permission_bitmask(vcpu, context); ASSERT(is_pae(vcpu)); context->new_cr3 = paging_new_cr3; @@ -3552,6 +3585,7 @@ static int paging32_init_context(struct kvm_vcpu *vcpu, context->root_level = PT32_ROOT_LEVEL; reset_rsvds_bits_mask(vcpu, context); + update_permission_bitmask(vcpu, context); context->new_cr3 = paging_new_cr3; context->page_fault = paging32_page_fault; @@ -3612,6 +3646,8 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->gva_to_gpa = paging32_gva_to_gpa; } + update_permission_bitmask(vcpu, context); + return 0; } @@ -3687,6 +3723,8 @@ static int init_kvm_nested_mmu(struct kvm_vcpu *vcpu) g_context->gva_to_gpa = paging32_gva_to_gpa_nested; } + update_permission_bitmask(vcpu, g_context); + return 0; } diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index 2832081e9b2e..584660775d08 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -89,17 +89,14 @@ static inline bool is_write_protection(struct kvm_vcpu *vcpu) return kvm_read_cr0_bits(vcpu, X86_CR0_WP); } -static inline bool check_write_user_access(struct kvm_vcpu *vcpu, - bool write_fault, bool user_fault, - unsigned long pte) +/* + * Will a fault with a given page-fault error code (pfec) cause a permission + * fault with the given access (in ACC_* format)? + */ +static inline bool permission_fault(struct kvm_mmu *mmu, unsigned pte_access, + unsigned pfec) { - if (unlikely(write_fault && !is_writable_pte(pte) - && (user_fault || is_write_protection(vcpu)))) - return false; - - if (unlikely(user_fault && !(pte & PT_USER_MASK))) - return false; - - return true; + return (mmu->permissions[pfec >> 1] >> pte_access) & 1; } + #endif diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 35a05dd2f69c..8f6c59fadbbe 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -169,7 +169,7 @@ static int FNAME(walk_addr_generic)(struct guest_walker *walker, pt_element_t pte; pt_element_t __user *uninitialized_var(ptep_user); gfn_t table_gfn; - unsigned index, pt_access, uninitialized_var(pte_access); + unsigned index, pt_access, pte_access; gpa_t pte_gpa; bool eperm, last_gpte; int offset; @@ -237,24 +237,9 @@ retry_walk: goto error; } - if (!check_write_user_access(vcpu, write_fault, user_fault, - pte)) - eperm = true; - -#if PTTYPE == 64 - if (unlikely(fetch_fault && (pte & PT64_NX_MASK))) - eperm = true; -#endif + pte_access = pt_access & gpte_access(vcpu, pte); last_gpte = FNAME(is_last_gpte)(walker, vcpu, mmu, pte); - if (last_gpte) { - pte_access = pt_access & gpte_access(vcpu, pte); - /* check if the kernel is fetching from user page */ - if (unlikely(pte_access & PT_USER_MASK) && - kvm_read_cr4_bits(vcpu, X86_CR4_SMEP)) - if (fetch_fault && !user_fault) - eperm = true; - } walker->ptes[walker->level - 1] = pte; @@ -284,10 +269,11 @@ retry_walk: break; } - pt_access &= gpte_access(vcpu, pte); + pt_access &= pte_access; --walker->level; } + eperm |= permission_fault(mmu, pte_access, access); if (unlikely(eperm)) { errcode |= PFERR_PRESENT_MASK; goto error; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 19047eafa38d..497226e49d4b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3672,20 +3672,17 @@ static int vcpu_mmio_gva_to_gpa(struct kvm_vcpu *vcpu, unsigned long gva, gpa_t *gpa, struct x86_exception *exception, bool write) { - u32 access = (kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0; + u32 access = ((kvm_x86_ops->get_cpl(vcpu) == 3) ? PFERR_USER_MASK : 0) + | (write ? PFERR_WRITE_MASK : 0); - if (vcpu_match_mmio_gva(vcpu, gva) && - check_write_user_access(vcpu, write, access, - vcpu->arch.access)) { + if (vcpu_match_mmio_gva(vcpu, gva) + && !permission_fault(vcpu->arch.walk_mmu, vcpu->arch.access, access)) { *gpa = vcpu->arch.mmio_gfn << PAGE_SHIFT | (gva & (PAGE_SIZE - 1)); trace_vcpu_match_mmio(gva, *gpa, write, false); return 1; } - if (write) - access |= PFERR_WRITE_MASK; - *gpa = vcpu->arch.walk_mmu->gva_to_gpa(vcpu, gva, access, exception); if (*gpa == UNMAPPED_GVA) -- cgit v1.2.3 From 6fd01b711bee96ce3356f7b6f370ab708e37504b Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Wed, 12 Sep 2012 20:46:56 +0300 Subject: KVM: MMU: Optimize is_last_gpte() Instead of branchy code depending on level, gpte.ps, and mmu configuration, prepare everything in a bitmap during mode changes and look it up during runtime. Reviewed-by: Xiao Guangrong Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 7 +++++++ arch/x86/kvm/mmu.c | 31 +++++++++++++++++++++++++++++++ arch/x86/kvm/mmu.h | 3 ++- arch/x86/kvm/paging_tmpl.h | 20 +------------------- 4 files changed, 41 insertions(+), 20 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 3318bde206a5..43aeb9422839 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -298,6 +298,13 @@ struct kvm_mmu { u64 *lm_root; u64 rsvd_bits_mask[2][4]; + /* + * Bitmap: bit set = last pte in walk + * index[0:1]: level (zero-based) + * index[2]: pte.ps + */ + u8 last_pte_bitmap; + bool nx; u64 pdptrs[4]; /* pae */ diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 9c6188931f87..d289fee1ffb8 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -3447,6 +3447,15 @@ static inline unsigned gpte_access(struct kvm_vcpu *vcpu, u64 gpte) return access; } +static inline bool is_last_gpte(struct kvm_mmu *mmu, unsigned level, unsigned gpte) +{ + unsigned index; + + index = level - 1; + index |= (gpte & PT_PAGE_SIZE_MASK) >> (PT_PAGE_SIZE_SHIFT - 2); + return mmu->last_pte_bitmap & (1 << index); +} + #define PTTYPE 64 #include "paging_tmpl.h" #undef PTTYPE @@ -3548,6 +3557,24 @@ static void update_permission_bitmask(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu } } +static void update_last_pte_bitmap(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu) +{ + u8 map; + unsigned level, root_level = mmu->root_level; + const unsigned ps_set_index = 1 << 2; /* bit 2 of index: ps */ + + if (root_level == PT32E_ROOT_LEVEL) + --root_level; + /* PT_PAGE_TABLE_LEVEL always terminates */ + map = 1 | (1 << ps_set_index); + for (level = PT_DIRECTORY_LEVEL; level <= root_level; ++level) { + if (level <= PT_PDPE_LEVEL + && (mmu->root_level >= PT32E_ROOT_LEVEL || is_pse(vcpu))) + map |= 1 << (ps_set_index | (level - 1)); + } + mmu->last_pte_bitmap = map; +} + static int paging64_init_context_common(struct kvm_vcpu *vcpu, struct kvm_mmu *context, int level) @@ -3557,6 +3584,7 @@ static int paging64_init_context_common(struct kvm_vcpu *vcpu, reset_rsvds_bits_mask(vcpu, context); update_permission_bitmask(vcpu, context); + update_last_pte_bitmap(vcpu, context); ASSERT(is_pae(vcpu)); context->new_cr3 = paging_new_cr3; @@ -3586,6 +3614,7 @@ static int paging32_init_context(struct kvm_vcpu *vcpu, reset_rsvds_bits_mask(vcpu, context); update_permission_bitmask(vcpu, context); + update_last_pte_bitmap(vcpu, context); context->new_cr3 = paging_new_cr3; context->page_fault = paging32_page_fault; @@ -3647,6 +3676,7 @@ static int init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) } update_permission_bitmask(vcpu, context); + update_last_pte_bitmap(vcpu, context); return 0; } @@ -3724,6 +3754,7 @@ static int init_kvm_nested_mmu(struct kvm_vcpu *vcpu) } update_permission_bitmask(vcpu, g_context); + update_last_pte_bitmap(vcpu, g_context); return 0; } diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index 584660775d08..69871080e866 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -20,7 +20,8 @@ #define PT_ACCESSED_MASK (1ULL << PT_ACCESSED_SHIFT) #define PT_DIRTY_SHIFT 6 #define PT_DIRTY_MASK (1ULL << PT_DIRTY_SHIFT) -#define PT_PAGE_SIZE_MASK (1ULL << 7) +#define PT_PAGE_SIZE_SHIFT 7 +#define PT_PAGE_SIZE_MASK (1ULL << PT_PAGE_SIZE_SHIFT) #define PT_PAT_MASK (1ULL << 7) #define PT_GLOBAL_MASK (1ULL << 8) #define PT64_NX_SHIFT 63 diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index 1b4c14d235a0..134ea7b1c585 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -103,24 +103,6 @@ static int FNAME(cmpxchg_gpte)(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, return (ret != orig_pte); } -static bool FNAME(is_last_gpte)(struct guest_walker *walker, - struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, - pt_element_t gpte) -{ - if (walker->level == PT_PAGE_TABLE_LEVEL) - return true; - - if ((walker->level == PT_DIRECTORY_LEVEL) && is_large_pte(gpte) && - (PTTYPE == 64 || is_pse(vcpu))) - return true; - - if ((walker->level == PT_PDPE_LEVEL) && is_large_pte(gpte) && - (mmu->root_level == PT64_ROOT_LEVEL)) - return true; - - return false; -} - static int FNAME(update_accessed_dirty_bits)(struct kvm_vcpu *vcpu, struct kvm_mmu *mmu, struct guest_walker *walker, @@ -247,7 +229,7 @@ retry_walk: pte_access = pt_access & gpte_access(vcpu, pte); walker->ptes[walker->level - 1] = pte; - } while (!FNAME(is_last_gpte)(walker, vcpu, mmu, pte)); + } while (!is_last_gpte(mmu, walker->level, pte)); eperm |= permission_fault(mmu, pte_access, access); if (unlikely(eperm)) { -- cgit v1.2.3 From 1e08ec4a130e2745d96df169e67c58df98a07311 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Thu, 13 Sep 2012 17:19:24 +0300 Subject: KVM: optimize apic interrupt delivery Most interrupt are delivered to only one vcpu. Use pre-build tables to find interrupt destination instead of looping through all vcpus. In case of logical mode loop only through vcpus in a logical cluster irq is sent to. Signed-off-by: Gleb Natapov Acked-by: Michael S. Tsirkin Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 12 +++ arch/x86/kvm/lapic.c | 188 +++++++++++++++++++++++++++++++++++++--- arch/x86/kvm/lapic.h | 3 + arch/x86/kvm/x86.c | 2 + virt/kvm/irq_comm.c | 7 +- 5 files changed, 199 insertions(+), 13 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 43aeb9422839..0b902c98f279 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -525,6 +525,16 @@ struct kvm_arch_memory_slot { struct kvm_lpage_info *lpage_info[KVM_NR_PAGE_SIZES - 1]; }; +struct kvm_apic_map { + struct rcu_head rcu; + u8 ldr_bits; + /* fields bellow are used to decode ldr values in different modes */ + u32 cid_shift, cid_mask, lid_mask; + struct kvm_lapic *phys_map[256]; + /* first index is cluster id second is cpu id in a cluster */ + struct kvm_lapic *logical_map[16][16]; +}; + struct kvm_arch { unsigned int n_used_mmu_pages; unsigned int n_requested_mmu_pages; @@ -542,6 +552,8 @@ struct kvm_arch { struct kvm_ioapic *vioapic; struct kvm_pit *vpit; int vapics_in_nmi_mode; + struct mutex apic_map_lock; + struct kvm_apic_map *apic_map; unsigned int tss_addr; struct page *apic_access_page; diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 6f9fd633c888..c6e6b721b6ee 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -140,11 +140,110 @@ static inline int apic_enabled(struct kvm_lapic *apic) (LVT_MASK | APIC_MODE_MASK | APIC_INPUT_POLARITY | \ APIC_LVT_REMOTE_IRR | APIC_LVT_LEVEL_TRIGGER) +static inline int apic_x2apic_mode(struct kvm_lapic *apic) +{ + return apic->vcpu->arch.apic_base & X2APIC_ENABLE; +} + static inline int kvm_apic_id(struct kvm_lapic *apic) { return (kvm_apic_get_reg(apic, APIC_ID) >> 24) & 0xff; } +static inline u16 apic_cluster_id(struct kvm_apic_map *map, u32 ldr) +{ + u16 cid; + ldr >>= 32 - map->ldr_bits; + cid = (ldr >> map->cid_shift) & map->cid_mask; + + BUG_ON(cid >= ARRAY_SIZE(map->logical_map)); + + return cid; +} + +static inline u16 apic_logical_id(struct kvm_apic_map *map, u32 ldr) +{ + ldr >>= (32 - map->ldr_bits); + return ldr & map->lid_mask; +} + +static void recalculate_apic_map(struct kvm *kvm) +{ + struct kvm_apic_map *new, *old = NULL; + struct kvm_vcpu *vcpu; + int i; + + new = kzalloc(sizeof(struct kvm_apic_map), GFP_KERNEL); + + mutex_lock(&kvm->arch.apic_map_lock); + + if (!new) + goto out; + + new->ldr_bits = 8; + /* flat mode is default */ + new->cid_shift = 8; + new->cid_mask = 0; + new->lid_mask = 0xff; + + kvm_for_each_vcpu(i, vcpu, kvm) { + struct kvm_lapic *apic = vcpu->arch.apic; + u16 cid, lid; + u32 ldr; + + if (!kvm_apic_present(vcpu)) + continue; + + /* + * All APICs have to be configured in the same mode by an OS. + * We take advatage of this while building logical id loockup + * table. After reset APICs are in xapic/flat mode, so if we + * find apic with different setting we assume this is the mode + * OS wants all apics to be in; build lookup table accordingly. + */ + if (apic_x2apic_mode(apic)) { + new->ldr_bits = 32; + new->cid_shift = 16; + new->cid_mask = new->lid_mask = 0xffff; + } else if (kvm_apic_sw_enabled(apic) && + !new->cid_mask /* flat mode */ && + kvm_apic_get_reg(apic, APIC_DFR) == APIC_DFR_CLUSTER) { + new->cid_shift = 4; + new->cid_mask = 0xf; + new->lid_mask = 0xf; + } + + new->phys_map[kvm_apic_id(apic)] = apic; + + ldr = kvm_apic_get_reg(apic, APIC_LDR); + cid = apic_cluster_id(new, ldr); + lid = apic_logical_id(new, ldr); + + if (lid) + new->logical_map[cid][ffs(lid) - 1] = apic; + } +out: + old = rcu_dereference_protected(kvm->arch.apic_map, + lockdep_is_held(&kvm->arch.apic_map_lock)); + rcu_assign_pointer(kvm->arch.apic_map, new); + mutex_unlock(&kvm->arch.apic_map_lock); + + if (old) + kfree_rcu(old, rcu); +} + +static inline void kvm_apic_set_id(struct kvm_lapic *apic, u8 id) +{ + apic_set_reg(apic, APIC_ID, id << 24); + recalculate_apic_map(apic->vcpu->kvm); +} + +static inline void kvm_apic_set_ldr(struct kvm_lapic *apic, u32 id) +{ + apic_set_reg(apic, APIC_LDR, id); + recalculate_apic_map(apic->vcpu->kvm); +} + static inline int apic_lvt_enabled(struct kvm_lapic *apic, int lvt_type) { return !(kvm_apic_get_reg(apic, lvt_type) & APIC_LVT_MASKED); @@ -194,11 +293,6 @@ void kvm_apic_set_version(struct kvm_vcpu *vcpu) apic_set_reg(apic, APIC_LVR, v); } -static inline int apic_x2apic_mode(struct kvm_lapic *apic) -{ - return apic->vcpu->arch.apic_base & X2APIC_ENABLE; -} - static const unsigned int apic_lvt_mask[APIC_LVT_NUM] = { LVT_MASK , /* part LVTT mask, timer mode mask added at runtime */ LVT_MASK | APIC_MODE_MASK, /* LVTTHMR */ @@ -483,6 +577,72 @@ int kvm_apic_match_dest(struct kvm_vcpu *vcpu, struct kvm_lapic *source, return result; } +bool kvm_irq_delivery_to_apic_fast(struct kvm *kvm, struct kvm_lapic *src, + struct kvm_lapic_irq *irq, int *r) +{ + struct kvm_apic_map *map; + unsigned long bitmap = 1; + struct kvm_lapic **dst; + int i; + bool ret = false; + + *r = -1; + + if (irq->shorthand == APIC_DEST_SELF) { + *r = kvm_apic_set_irq(src->vcpu, irq); + return true; + } + + if (irq->shorthand) + return false; + + rcu_read_lock(); + map = rcu_dereference(kvm->arch.apic_map); + + if (!map) + goto out; + + if (irq->dest_mode == 0) { /* physical mode */ + if (irq->delivery_mode == APIC_DM_LOWEST || + irq->dest_id == 0xff) + goto out; + dst = &map->phys_map[irq->dest_id & 0xff]; + } else { + u32 mda = irq->dest_id << (32 - map->ldr_bits); + + dst = map->logical_map[apic_cluster_id(map, mda)]; + + bitmap = apic_logical_id(map, mda); + + if (irq->delivery_mode == APIC_DM_LOWEST) { + int l = -1; + for_each_set_bit(i, &bitmap, 16) { + if (!dst[i]) + continue; + if (l < 0) + l = i; + else if (kvm_apic_compare_prio(dst[i]->vcpu, dst[l]->vcpu) < 0) + l = i; + } + + bitmap = (l >= 0) ? 1 << l : 0; + } + } + + for_each_set_bit(i, &bitmap, 16) { + if (!dst[i]) + continue; + if (*r < 0) + *r = 0; + *r += kvm_apic_set_irq(dst[i]->vcpu, irq); + } + + ret = true; +out: + rcu_read_unlock(); + return ret; +} + /* * Add a pending IRQ into lapic. * Return 1 if successfully added and 0 if discarded. @@ -886,7 +1046,7 @@ static int apic_reg_write(struct kvm_lapic *apic, u32 reg, u32 val) switch (reg) { case APIC_ID: /* Local APIC ID */ if (!apic_x2apic_mode(apic)) - apic_set_reg(apic, APIC_ID, val); + kvm_apic_set_id(apic, val >> 24); else ret = 1; break; @@ -902,15 +1062,16 @@ static int apic_reg_write(struct kvm_lapic *apic, u32 reg, u32 val) case APIC_LDR: if (!apic_x2apic_mode(apic)) - apic_set_reg(apic, APIC_LDR, val & APIC_LDR_MASK); + kvm_apic_set_ldr(apic, val & APIC_LDR_MASK); else ret = 1; break; case APIC_DFR: - if (!apic_x2apic_mode(apic)) + if (!apic_x2apic_mode(apic)) { apic_set_reg(apic, APIC_DFR, val | 0x0FFFFFFF); - else + recalculate_apic_map(apic->vcpu->kvm); + } else ret = 1; break; @@ -1141,6 +1302,7 @@ void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value) static_key_slow_dec_deferred(&apic_hw_disabled); else static_key_slow_inc(&apic_hw_disabled.key); + recalculate_apic_map(vcpu->kvm); } if (!kvm_vcpu_is_bsp(apic->vcpu)) @@ -1150,7 +1312,7 @@ void kvm_lapic_set_base(struct kvm_vcpu *vcpu, u64 value) if (apic_x2apic_mode(apic)) { u32 id = kvm_apic_id(apic); u32 ldr = ((id & ~0xf) << 16) | (1 << (id & 0xf)); - apic_set_reg(apic, APIC_LDR, ldr); + kvm_apic_set_ldr(apic, ldr); } apic->base_address = apic->vcpu->arch.apic_base & MSR_IA32_APICBASE_BASE; @@ -1175,7 +1337,7 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu) /* Stop the timer in case it's a reset to an active apic */ hrtimer_cancel(&apic->lapic_timer.timer); - apic_set_reg(apic, APIC_ID, vcpu->vcpu_id << 24); + kvm_apic_set_id(apic, vcpu->vcpu_id); kvm_apic_set_version(apic->vcpu); for (i = 0; i < APIC_LVT_NUM; i++) @@ -1186,7 +1348,7 @@ void kvm_lapic_reset(struct kvm_vcpu *vcpu) apic_set_reg(apic, APIC_DFR, 0xffffffffU); apic_set_spiv(apic, 0xff); apic_set_reg(apic, APIC_TASKPRI, 0); - apic_set_reg(apic, APIC_LDR, 0); + kvm_apic_set_ldr(apic, 0); apic_set_reg(apic, APIC_ESR, 0); apic_set_reg(apic, APIC_ICR, 0); apic_set_reg(apic, APIC_ICR2, 0); @@ -1404,6 +1566,8 @@ void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu, /* set SPIV separately to get count of SW disabled APICs right */ apic_set_spiv(apic, *((u32 *)(s->regs + APIC_SPIV))); memcpy(vcpu->arch.apic->regs, s->regs, sizeof *s); + /* call kvm_apic_set_id() to put apic into apic_map */ + kvm_apic_set_id(apic, kvm_apic_id(apic)); kvm_apic_set_version(vcpu); apic_update_ppr(apic); diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h index 615a8b030168..e5ebf9f3571f 100644 --- a/arch/x86/kvm/lapic.h +++ b/arch/x86/kvm/lapic.h @@ -52,6 +52,9 @@ int kvm_apic_match_logical_addr(struct kvm_lapic *apic, u8 mda); int kvm_apic_set_irq(struct kvm_vcpu *vcpu, struct kvm_lapic_irq *irq); int kvm_apic_local_deliver(struct kvm_lapic *apic, int lvt_type); +bool kvm_irq_delivery_to_apic_fast(struct kvm *kvm, struct kvm_lapic *src, + struct kvm_lapic_irq *irq, int *r); + u64 kvm_get_apic_base(struct kvm_vcpu *vcpu); void kvm_set_apic_base(struct kvm_vcpu *vcpu, u64 data); void kvm_apic_post_state_restore(struct kvm_vcpu *vcpu, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 497226e49d4b..fc2a0a132e4b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6270,6 +6270,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type) set_bit(KVM_USERSPACE_IRQ_SOURCE_ID, &kvm->arch.irq_sources_bitmap); raw_spin_lock_init(&kvm->arch.tsc_write_lock); + mutex_init(&kvm->arch.apic_map_lock); return 0; } @@ -6322,6 +6323,7 @@ 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); + kfree(rcu_dereference_check(kvm->arch.apic_map, 1)); } void kvm_arch_free_memslot(struct kvm_memory_slot *free, diff --git a/virt/kvm/irq_comm.c b/virt/kvm/irq_comm.c index 7118be0f2f2c..3ca89c451d6b 100644 --- a/virt/kvm/irq_comm.c +++ b/virt/kvm/irq_comm.c @@ -68,8 +68,13 @@ int kvm_irq_delivery_to_apic(struct kvm *kvm, struct kvm_lapic *src, struct kvm_vcpu *vcpu, *lowest = NULL; if (irq->dest_mode == 0 && irq->dest_id == 0xff && - kvm_is_dm_lowest_prio(irq)) + kvm_is_dm_lowest_prio(irq)) { printk(KERN_INFO "kvm: apic: phys broadcast and lowest prio\n"); + irq->delivery_mode = APIC_DM_FIXED; + } + + if (kvm_irq_delivery_to_apic_fast(kvm, src, irq, &r)) + return r; kvm_for_each_vcpu(i, vcpu, kvm) { if (!kvm_apic_present(vcpu)) -- cgit v1.2.3 From c863901075a42d50678616d8ee4b96ef13080498 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Fri, 21 Sep 2012 05:42:55 +0200 Subject: KVM: x86: Fix guest debug across vcpu INIT reset If we reset a vcpu on INIT, we so far overwrote dr7 as provided by KVM_SET_GUEST_DEBUG, and we also cleared switch_db_regs unconditionally. Fix this by saving the dr7 used for guest debugging and calculating the effective register value as well as switch_db_regs on any potential change. This will change to focus of the set_guest_debug vendor op to update_dp_bp_intercept. Found while trying to stop on start_secondary. Signed-off-by: Jan Kiszka Signed-off-by: Avi Kivity --- arch/x86/include/asm/kvm_host.h | 4 ++-- arch/x86/kvm/svm.c | 23 ++++------------------- arch/x86/kvm/vmx.c | 14 +------------- arch/x86/kvm/x86.c | 26 +++++++++++++++++--------- 4 files changed, 24 insertions(+), 43 deletions(-) (limited to 'arch/x86/include/asm/kvm_host.h') diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 0b902c98f279..c9a91368fc5e 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -471,6 +471,7 @@ struct kvm_vcpu_arch { unsigned long dr6; unsigned long dr7; unsigned long eff_db[KVM_NR_DB_REGS]; + unsigned long guest_debug_dr7; u64 mcg_cap; u64 mcg_status; @@ -647,8 +648,7 @@ struct kvm_x86_ops { void (*vcpu_load)(struct kvm_vcpu *vcpu, int cpu); void (*vcpu_put)(struct kvm_vcpu *vcpu); - void (*set_guest_debug)(struct kvm_vcpu *vcpu, - struct kvm_guest_debug *dbg); + void (*update_db_bp_intercept)(struct kvm_vcpu *vcpu); int (*get_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 *pdata); int (*set_msr)(struct kvm_vcpu *vcpu, u32 msr_index, u64 data); u64 (*get_segment_base)(struct kvm_vcpu *vcpu, int seg); diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 818fceb3091e..d017df3899ef 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -1146,7 +1146,6 @@ static void init_vmcb(struct vcpu_svm *svm) svm_set_efer(&svm->vcpu, 0); save->dr6 = 0xffff0ff0; - save->dr7 = 0x400; kvm_set_rflags(&svm->vcpu, 2); save->rip = 0x0000fff0; svm->vcpu.arch.regs[VCPU_REGS_RIP] = save->rip; @@ -1643,7 +1642,7 @@ static void svm_set_segment(struct kvm_vcpu *vcpu, mark_dirty(svm->vmcb, VMCB_SEG); } -static void update_db_intercept(struct kvm_vcpu *vcpu) +static void update_db_bp_intercept(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); @@ -1663,20 +1662,6 @@ static void update_db_intercept(struct kvm_vcpu *vcpu) vcpu->guest_debug = 0; } -static void svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) -{ - struct vcpu_svm *svm = to_svm(vcpu); - - if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) - svm->vmcb->save.dr7 = dbg->arch.debugreg[7]; - else - svm->vmcb->save.dr7 = vcpu->arch.dr7; - - mark_dirty(svm->vmcb, VMCB_DR); - - update_db_intercept(vcpu); -} - static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *sd) { if (sd->next_asid > sd->max_asid) { @@ -1748,7 +1733,7 @@ static int db_interception(struct vcpu_svm *svm) if (!(svm->vcpu.guest_debug & KVM_GUESTDBG_SINGLESTEP)) svm->vmcb->save.rflags &= ~(X86_EFLAGS_TF | X86_EFLAGS_RF); - update_db_intercept(&svm->vcpu); + update_db_bp_intercept(&svm->vcpu); } if (svm->vcpu.guest_debug & @@ -3659,7 +3644,7 @@ static void enable_nmi_window(struct kvm_vcpu *vcpu) */ svm->nmi_singlestep = true; svm->vmcb->save.rflags |= (X86_EFLAGS_TF | X86_EFLAGS_RF); - update_db_intercept(vcpu); + update_db_bp_intercept(vcpu); } static int svm_set_tss_addr(struct kvm *kvm, unsigned int addr) @@ -4253,7 +4238,7 @@ static struct kvm_x86_ops svm_x86_ops = { .vcpu_load = svm_vcpu_load, .vcpu_put = svm_vcpu_put, - .set_guest_debug = svm_guest_debug, + .update_db_bp_intercept = update_db_bp_intercept, .get_msr = svm_get_msr, .set_msr = svm_set_msr, .get_segment_base = svm_get_segment_base, diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 30bcb953afee..5d46c905e06f 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -2288,16 +2288,6 @@ static void vmx_cache_reg(struct kvm_vcpu *vcpu, enum kvm_reg reg) } } -static void set_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) -{ - if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) - vmcs_writel(GUEST_DR7, dbg->arch.debugreg[7]); - else - vmcs_writel(GUEST_DR7, vcpu->arch.dr7); - - update_exception_bitmap(vcpu); -} - static __init int cpu_has_kvm_support(void) { return cpu_has_vmx(); @@ -3960,8 +3950,6 @@ static int vmx_vcpu_reset(struct kvm_vcpu *vcpu) kvm_rip_write(vcpu, 0); kvm_register_write(vcpu, VCPU_REGS_RSP, 0); - vmcs_writel(GUEST_DR7, 0x400); - vmcs_writel(GUEST_GDTR_BASE, 0); vmcs_write32(GUEST_GDTR_LIMIT, 0xffff); @@ -7237,7 +7225,7 @@ static struct kvm_x86_ops vmx_x86_ops = { .vcpu_load = vmx_vcpu_load, .vcpu_put = vmx_vcpu_put, - .set_guest_debug = set_guest_debug, + .update_db_bp_intercept = update_exception_bitmap, .get_msr = vmx_get_msr, .set_msr = vmx_set_msr, .get_segment_base = vmx_get_segment_base, diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 7d44204c6041..b16d4a5bfa41 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -692,6 +692,18 @@ unsigned long kvm_get_cr8(struct kvm_vcpu *vcpu) } EXPORT_SYMBOL_GPL(kvm_get_cr8); +static void kvm_update_dr7(struct kvm_vcpu *vcpu) +{ + unsigned long dr7; + + if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) + dr7 = vcpu->arch.guest_debug_dr7; + else + dr7 = vcpu->arch.dr7; + kvm_x86_ops->set_dr7(vcpu, dr7); + vcpu->arch.switch_db_regs = (dr7 & DR7_BP_EN_MASK); +} + static int __kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val) { switch (dr) { @@ -717,10 +729,7 @@ static int __kvm_set_dr(struct kvm_vcpu *vcpu, int dr, unsigned long val) if (val & 0xffffffff00000000ULL) return -1; /* #GP */ vcpu->arch.dr7 = (val & DR7_VOLATILE) | DR7_FIXED_1; - if (!(vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP)) { - kvm_x86_ops->set_dr7(vcpu, vcpu->arch.dr7); - vcpu->arch.switch_db_regs = (val & DR7_BP_EN_MASK); - } + kvm_update_dr7(vcpu); break; } @@ -5851,13 +5860,12 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, if (vcpu->guest_debug & KVM_GUESTDBG_USE_HW_BP) { for (i = 0; i < KVM_NR_DB_REGS; ++i) vcpu->arch.eff_db[i] = dbg->arch.debugreg[i]; - vcpu->arch.switch_db_regs = - (dbg->arch.debugreg[7] & DR7_BP_EN_MASK); + vcpu->arch.guest_debug_dr7 = dbg->arch.debugreg[7]; } else { for (i = 0; i < KVM_NR_DB_REGS; i++) vcpu->arch.eff_db[i] = vcpu->arch.db[i]; - vcpu->arch.switch_db_regs = (vcpu->arch.dr7 & DR7_BP_EN_MASK); } + kvm_update_dr7(vcpu); if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) vcpu->arch.singlestep_rip = kvm_rip_read(vcpu) + @@ -5869,7 +5877,7 @@ int kvm_arch_vcpu_ioctl_set_guest_debug(struct kvm_vcpu *vcpu, */ kvm_set_rflags(vcpu, rflags); - kvm_x86_ops->set_guest_debug(vcpu, dbg); + kvm_x86_ops->update_db_bp_intercept(vcpu); r = 0; @@ -6045,10 +6053,10 @@ int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu) vcpu->arch.nmi_pending = 0; vcpu->arch.nmi_injected = false; - vcpu->arch.switch_db_regs = 0; memset(vcpu->arch.db, 0, sizeof(vcpu->arch.db)); vcpu->arch.dr6 = DR6_FIXED_1; vcpu->arch.dr7 = DR7_FIXED_1; + kvm_update_dr7(vcpu); kvm_make_request(KVM_REQ_EVENT, vcpu); vcpu->arch.apf.msr_val = 0; -- cgit v1.2.3