From 010c520e20413dfd567d568aba2b7238acd37e33 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 11 Oct 2010 14:23:39 +0200 Subject: KVM: Don't reset mmu context unnecessarily when updating EFER The only bit of EFER that affects the mmu is NX, and this is already accounted for (LME only takes effect when changing cr0). Based on a patch by Hillf Danton. Signed-off-by: Avi Kivity --- arch/x86/kvm/x86.c | 1 - 1 file changed, 1 deletion(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index b989e1f1e5d..c05d4770129 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -830,7 +830,6 @@ static int set_efer(struct kvm_vcpu *vcpu, u64 efer) kvm_x86_ops->set_efer(vcpu, efer); vcpu->arch.mmu.base_role.nxe = (efer & EFER_NX) && !tdp_enabled; - kvm_mmu_reset_context(vcpu); /* Update reserved bits */ if ((efer ^ old_efer) & EFER_NX) -- cgit v1.2.3-18-g5258 From af585b921e5d1e919947c4b1164b59507fe7cd7b Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Thu, 14 Oct 2010 11:22:46 +0200 Subject: KVM: Halt vcpu if page it tries to access is swapped out If a guest accesses swapped out memory do not swap it in from vcpu thread context. Schedule work to do swapping and put vcpu into halted state instead. Interrupts will still be delivered to the guest and if interrupt will cause reschedule guest will continue to run another task. [avi: remove call to get_user_pages_noio(), nacked by Linus; this makes everything synchrnous again] Acked-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/Kconfig | 1 + arch/x86/kvm/Makefile | 1 + arch/x86/kvm/mmu.c | 52 ++++++++++++++++++++- arch/x86/kvm/paging_tmpl.h | 4 +- arch/x86/kvm/x86.c | 112 +++++++++++++++++++++++++++++++++++++++++++-- 5 files changed, 165 insertions(+), 5 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/Kconfig b/arch/x86/kvm/Kconfig index ddc131ff438..50f63648ce1 100644 --- a/arch/x86/kvm/Kconfig +++ b/arch/x86/kvm/Kconfig @@ -28,6 +28,7 @@ config KVM select HAVE_KVM_IRQCHIP select HAVE_KVM_EVENTFD select KVM_APIC_ARCHITECTURE + select KVM_ASYNC_PF select USER_RETURN_NOTIFIER select KVM_MMIO ---help--- diff --git a/arch/x86/kvm/Makefile b/arch/x86/kvm/Makefile index 31a7035c4bd..c53bf19b1da 100644 --- a/arch/x86/kvm/Makefile +++ b/arch/x86/kvm/Makefile @@ -9,6 +9,7 @@ kvm-y += $(addprefix ../../../virt/kvm/, kvm_main.o ioapic.o \ coalesced_mmio.o irq_comm.o eventfd.o \ assigned-dev.o) kvm-$(CONFIG_IOMMU_API) += $(addprefix ../../../virt/kvm/, iommu.o) +kvm-$(CONFIG_KVM_ASYNC_PF) += $(addprefix ../../../virt/kvm/, async_pf.o) kvm-y += x86.o mmu.o emulate.o i8259.o irq.o lapic.o \ i8254.o timer.o diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index fbb04aee830..4ab04de5a76 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -18,9 +18,11 @@ * */ +#include "irq.h" #include "mmu.h" #include "x86.h" #include "kvm_cache_regs.h" +#include "x86.h" #include #include @@ -2587,6 +2589,50 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva, error_code & PFERR_WRITE_MASK, gfn); } +int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn) +{ + struct kvm_arch_async_pf arch; + arch.gfn = gfn; + + return kvm_setup_async_pf(vcpu, gva, gfn, &arch); +} + +static bool can_do_async_pf(struct kvm_vcpu *vcpu) +{ + if (unlikely(!irqchip_in_kernel(vcpu->kvm) || + kvm_event_needs_reinjection(vcpu))) + return false; + + return kvm_x86_ops->interrupt_allowed(vcpu); +} + +static bool try_async_pf(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gva, + pfn_t *pfn) +{ + bool async; + + *pfn = gfn_to_pfn_async(vcpu->kvm, gfn, &async); + + if (!async) + return false; /* *pfn has correct page already */ + + put_page(pfn_to_page(*pfn)); + + if (can_do_async_pf(vcpu)) { + trace_kvm_try_async_get_page(async, *pfn); + if (kvm_find_async_pf_gfn(vcpu, gfn)) { + trace_kvm_async_pf_doublefault(gva, gfn); + kvm_make_request(KVM_REQ_APF_HALT, vcpu); + return true; + } else if (kvm_arch_setup_async_pf(vcpu, gva, gfn)) + return true; + } + + *pfn = gfn_to_pfn(vcpu->kvm, gfn); + + return false; +} + static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code) { @@ -2609,7 +2655,11 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); - pfn = gfn_to_pfn(vcpu->kvm, gfn); + + if (try_async_pf(vcpu, gfn, gpa, &pfn)) + return 0; + + /* mmio */ if (is_error_pfn(pfn)) return kvm_handle_bad_page(vcpu->kvm, gfn, pfn); spin_lock(&vcpu->kvm->mmu_lock); diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index cd7a833a3b5..c45376dd041 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -568,7 +568,9 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); - pfn = gfn_to_pfn(vcpu->kvm, walker.gfn); + + if (try_async_pf(vcpu, walker.gfn, addr, &pfn)) + return 0; /* mmio */ if (is_error_pfn(pfn)) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index c05d4770129..3cd4d091c2f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -43,6 +43,7 @@ #include #include #include +#include #include #define CREATE_TRACE_POINTS @@ -155,6 +156,13 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { u64 __read_mostly host_xcr0; +static inline void kvm_async_pf_hash_reset(struct kvm_vcpu *vcpu) +{ + int i; + for (i = 0; i < roundup_pow_of_two(ASYNC_PF_PER_VCPU); i++) + vcpu->arch.apf.gfns[i] = ~0; +} + static void kvm_on_user_return(struct user_return_notifier *urn) { unsigned slot; @@ -5115,6 +5123,12 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) vcpu->fpu_active = 0; kvm_x86_ops->fpu_deactivate(vcpu); } + if (kvm_check_request(KVM_REQ_APF_HALT, vcpu)) { + /* Page is swapped out. Do synthetic halt */ + vcpu->arch.apf.halted = true; + r = 1; + goto out; + } } r = kvm_mmu_reload(vcpu); @@ -5243,7 +5257,8 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) r = 1; while (r > 0) { - if (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE) + if (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE && + !vcpu->arch.apf.halted) r = vcpu_enter_guest(vcpu); else { srcu_read_unlock(&kvm->srcu, vcpu->srcu_idx); @@ -5256,6 +5271,7 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) vcpu->arch.mp_state = KVM_MP_STATE_RUNNABLE; case KVM_MP_STATE_RUNNABLE: + vcpu->arch.apf.halted = false; break; case KVM_MP_STATE_SIPI_RECEIVED: default: @@ -5277,6 +5293,9 @@ static int __vcpu_run(struct kvm_vcpu *vcpu) vcpu->run->exit_reason = KVM_EXIT_INTR; ++vcpu->stat.request_irq_exits; } + + kvm_check_async_pf_completion(vcpu); + if (signal_pending(current)) { r = -EINTR; vcpu->run->exit_reason = KVM_EXIT_INTR; @@ -5792,6 +5811,10 @@ int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu) kvm_make_request(KVM_REQ_EVENT, vcpu); + kvm_clear_async_pf_completion_queue(vcpu); + kvm_async_pf_hash_reset(vcpu); + vcpu->arch.apf.halted = false; + return kvm_x86_ops->vcpu_reset(vcpu); } @@ -5880,6 +5903,8 @@ int kvm_arch_vcpu_init(struct kvm_vcpu *vcpu) if (!zalloc_cpumask_var(&vcpu->arch.wbinvd_dirty_mask, GFP_KERNEL)) goto fail_free_mce_banks; + kvm_async_pf_hash_reset(vcpu); + return 0; fail_free_mce_banks: kfree(vcpu->arch.mce_banks); @@ -5938,8 +5963,10 @@ static void kvm_free_vcpus(struct kvm *kvm) /* * Unpin any mmu pages first. */ - kvm_for_each_vcpu(i, vcpu, kvm) + kvm_for_each_vcpu(i, vcpu, kvm) { + kvm_clear_async_pf_completion_queue(vcpu); kvm_unload_vcpu_mmu(vcpu); + } kvm_for_each_vcpu(i, vcpu, kvm) kvm_arch_vcpu_free(vcpu); @@ -6050,7 +6077,9 @@ void kvm_arch_flush_shadow(struct kvm *kvm) int kvm_arch_vcpu_runnable(struct kvm_vcpu *vcpu) { - return vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE + return (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE && + !vcpu->arch.apf.halted) + || !list_empty_careful(&vcpu->async_pf.done) || vcpu->arch.mp_state == KVM_MP_STATE_SIPI_RECEIVED || vcpu->arch.nmi_pending || (kvm_arch_interrupt_allowed(vcpu) && @@ -6109,6 +6138,83 @@ void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) } EXPORT_SYMBOL_GPL(kvm_set_rflags); +static inline u32 kvm_async_pf_hash_fn(gfn_t gfn) +{ + return hash_32(gfn & 0xffffffff, order_base_2(ASYNC_PF_PER_VCPU)); +} + +static inline u32 kvm_async_pf_next_probe(u32 key) +{ + return (key + 1) & (roundup_pow_of_two(ASYNC_PF_PER_VCPU) - 1); +} + +static void kvm_add_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) +{ + u32 key = kvm_async_pf_hash_fn(gfn); + + while (vcpu->arch.apf.gfns[key] != ~0) + key = kvm_async_pf_next_probe(key); + + vcpu->arch.apf.gfns[key] = gfn; +} + +static u32 kvm_async_pf_gfn_slot(struct kvm_vcpu *vcpu, gfn_t gfn) +{ + int i; + u32 key = kvm_async_pf_hash_fn(gfn); + + for (i = 0; i < roundup_pow_of_two(ASYNC_PF_PER_VCPU) && + (vcpu->arch.apf.gfns[key] != gfn || + vcpu->arch.apf.gfns[key] == ~0); i++) + key = kvm_async_pf_next_probe(key); + + return key; +} + +bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) +{ + return vcpu->arch.apf.gfns[kvm_async_pf_gfn_slot(vcpu, gfn)] == gfn; +} + +static void kvm_del_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) +{ + u32 i, j, k; + + i = j = kvm_async_pf_gfn_slot(vcpu, gfn); + while (true) { + vcpu->arch.apf.gfns[i] = ~0; + do { + j = kvm_async_pf_next_probe(j); + if (vcpu->arch.apf.gfns[j] == ~0) + return; + k = kvm_async_pf_hash_fn(vcpu->arch.apf.gfns[j]); + /* + * k lies cyclically in ]i,j] + * | i.k.j | + * |....j i.k.| or |.k..j i...| + */ + } while ((i <= j) ? (i < k && k <= j) : (i < k || k <= j)); + vcpu->arch.apf.gfns[i] = vcpu->arch.apf.gfns[j]; + i = j; + } +} + +void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work) +{ + trace_kvm_async_pf_not_present(work->gva); + + kvm_make_request(KVM_REQ_APF_HALT, vcpu); + kvm_add_async_pf_gfn(vcpu, work->arch.gfn); +} + +void kvm_arch_async_page_present(struct kvm_vcpu *vcpu, + struct kvm_async_pf *work) +{ + trace_kvm_async_pf_ready(work->gva); + kvm_del_async_pf_gfn(vcpu, work->arch.gfn); +} + EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_exit); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_inj_virq); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_page_fault); -- cgit v1.2.3-18-g5258 From 56028d0861e48f7cc9c573d79f2d8a0a933a2bba Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Sun, 17 Oct 2010 18:13:42 +0200 Subject: KVM: Retry fault before vmentry When page is swapped in it is mapped into guest memory only after guest tries to access it again and generate another fault. To save this fault we can map it immediately since we know that guest is going to access the page. Do it only when tdp is enabled for now. Shadow paging case is more complicated. CR[034] and EFER registers should be switched before doing mapping and then switched back. Acked-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/mmu.c | 16 ++++++++-------- arch/x86/kvm/paging_tmpl.h | 6 +++--- arch/x86/kvm/x86.c | 14 ++++++++++++++ 3 files changed, 25 insertions(+), 11 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 4ab04de5a76..b2c60986a7c 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2570,7 +2570,7 @@ static gpa_t nonpaging_gva_to_gpa_nested(struct kvm_vcpu *vcpu, gva_t vaddr, } static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva, - u32 error_code) + u32 error_code, bool no_apf) { gfn_t gfn; int r; @@ -2606,8 +2606,8 @@ static bool can_do_async_pf(struct kvm_vcpu *vcpu) return kvm_x86_ops->interrupt_allowed(vcpu); } -static bool try_async_pf(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gva, - pfn_t *pfn) +static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn, + gva_t gva, pfn_t *pfn) { bool async; @@ -2618,7 +2618,7 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gva, put_page(pfn_to_page(*pfn)); - if (can_do_async_pf(vcpu)) { + if (!no_apf && can_do_async_pf(vcpu)) { trace_kvm_try_async_get_page(async, *pfn); if (kvm_find_async_pf_gfn(vcpu, gfn)) { trace_kvm_async_pf_doublefault(gva, gfn); @@ -2633,8 +2633,8 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, gfn_t gfn, gva_t gva, return false; } -static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, - u32 error_code) +static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, + bool no_apf) { pfn_t pfn; int r; @@ -2656,7 +2656,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); - if (try_async_pf(vcpu, gfn, gpa, &pfn)) + if (try_async_pf(vcpu, no_apf, gfn, gpa, &pfn)) return 0; /* mmio */ @@ -3319,7 +3319,7 @@ int kvm_mmu_page_fault(struct kvm_vcpu *vcpu, gva_t cr2, u32 error_code) int r; enum emulation_result er; - r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code); + r = vcpu->arch.mmu.page_fault(vcpu, cr2, error_code, false); if (r < 0) goto out; diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index c45376dd041..d6b281e989b 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -527,8 +527,8 @@ out_gpte_changed: * Returns: 1 if we need to emulate the instruction, 0 otherwise, or * a negative value on error. */ -static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, - u32 error_code) +static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, + bool no_apf) { int write_fault = error_code & PFERR_WRITE_MASK; int user_fault = error_code & PFERR_USER_MASK; @@ -569,7 +569,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); - if (try_async_pf(vcpu, walker.gfn, addr, &pfn)) + if (try_async_pf(vcpu, no_apf, walker.gfn, addr, &pfn)) return 0; /* mmio */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 3cd4d091c2f..71beb27597f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6138,6 +6138,20 @@ void kvm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) } EXPORT_SYMBOL_GPL(kvm_set_rflags); +void kvm_arch_async_page_ready(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) +{ + int r; + + if (!vcpu->arch.mmu.direct_map || is_error_page(work->page)) + return; + + r = kvm_mmu_reload(vcpu); + if (unlikely(r)) + return; + + vcpu->arch.mmu.page_fault(vcpu, work->gva, 0, true); +} + static inline u32 kvm_async_pf_hash_fn(gfn_t gfn) { return hash_32(gfn & 0xffffffff, order_base_2(ASYNC_PF_PER_VCPU)); -- cgit v1.2.3-18-g5258 From 49c7754ce57063b819b01eb8a4290841ad0886c4 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Mon, 18 Oct 2010 15:22:23 +0200 Subject: KVM: Add memory slot versioning and use it to provide fast guest write interface Keep track of memslots changes by keeping generation number in memslots structure. Provide kvm_write_guest_cached() function that skips gfn_to_hva() translation if memslots was not changed since previous invocation. Acked-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 71beb27597f..bd254779d1c 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3190,6 +3190,7 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, } memcpy(slots, kvm->memslots, sizeof(struct kvm_memslots)); slots->memslots[log->slot].dirty_bitmap = dirty_bitmap; + slots->generation++; old_slots = kvm->memslots; rcu_assign_pointer(kvm->memslots, slots); -- cgit v1.2.3-18-g5258 From 344d9588a9df06182684168be4f1408b55c7da3e Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Thu, 14 Oct 2010 11:22:50 +0200 Subject: KVM: Add PV MSR to enable asynchronous page faults delivery. Guest enables async PF vcpu functionality using this MSR. Reviewed-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 38 ++++++++++++++++++++++++++++++++++++-- 1 file changed, 36 insertions(+), 2 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index bd254779d1c..063c0729676 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -783,12 +783,12 @@ EXPORT_SYMBOL_GPL(kvm_get_dr); * kvm-specific. Those are put in the beginning of the list. */ -#define KVM_SAVE_MSRS_BEGIN 7 +#define KVM_SAVE_MSRS_BEGIN 8 static u32 msrs_to_save[] = { MSR_KVM_SYSTEM_TIME, MSR_KVM_WALL_CLOCK, MSR_KVM_SYSTEM_TIME_NEW, MSR_KVM_WALL_CLOCK_NEW, HV_X64_MSR_GUEST_OS_ID, HV_X64_MSR_HYPERCALL, - HV_X64_MSR_APIC_ASSIST_PAGE, + HV_X64_MSR_APIC_ASSIST_PAGE, MSR_KVM_ASYNC_PF_EN, MSR_IA32_SYSENTER_CS, MSR_IA32_SYSENTER_ESP, MSR_IA32_SYSENTER_EIP, MSR_STAR, #ifdef CONFIG_X86_64 @@ -1425,6 +1425,29 @@ static int set_msr_hyperv(struct kvm_vcpu *vcpu, u32 msr, u64 data) return 0; } +static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data) +{ + gpa_t gpa = data & ~0x3f; + + /* Bits 1:5 are resrved, Should be zero */ + if (data & 0x3e) + return 1; + + vcpu->arch.apf.msr_val = data; + + if (!(data & KVM_ASYNC_PF_ENABLED)) { + kvm_clear_async_pf_completion_queue(vcpu); + kvm_async_pf_hash_reset(vcpu); + return 0; + } + + if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.apf.data, gpa)) + return 1; + + kvm_async_pf_wakeup_all(vcpu); + return 0; +} + int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) { switch (msr) { @@ -1506,6 +1529,10 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 data) } break; } + case MSR_KVM_ASYNC_PF_EN: + if (kvm_pv_enable_async_pf(vcpu, data)) + return 1; + break; case MSR_IA32_MCG_CTL: case MSR_IA32_MCG_STATUS: case MSR_IA32_MC0_CTL ... MSR_IA32_MC0_CTL + 4 * KVM_MAX_MCE_BANKS - 1: @@ -1782,6 +1809,9 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, u32 msr, u64 *pdata) case MSR_KVM_SYSTEM_TIME_NEW: data = vcpu->arch.time; break; + case MSR_KVM_ASYNC_PF_EN: + data = vcpu->arch.apf.msr_val; + break; case MSR_IA32_P5_MC_ADDR: case MSR_IA32_P5_MC_TYPE: case MSR_IA32_MCG_CAP: @@ -1929,6 +1959,7 @@ int kvm_dev_ioctl_check_extension(long ext) case KVM_CAP_DEBUGREGS: case KVM_CAP_X86_ROBUST_SINGLESTEP: case KVM_CAP_XSAVE: + case KVM_CAP_ASYNC_PF: r = 1; break; case KVM_CAP_COALESCED_MMIO: @@ -5792,6 +5823,8 @@ free_vcpu: void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) { + vcpu->arch.apf.msr_val = 0; + vcpu_load(vcpu); kvm_mmu_unload(vcpu); vcpu_put(vcpu); @@ -5811,6 +5844,7 @@ int kvm_arch_vcpu_reset(struct kvm_vcpu *vcpu) vcpu->arch.dr7 = DR7_FIXED_1; kvm_make_request(KVM_REQ_EVENT, vcpu); + vcpu->arch.apf.msr_val = 0; kvm_clear_async_pf_completion_queue(vcpu); kvm_async_pf_hash_reset(vcpu); -- cgit v1.2.3-18-g5258 From 631bc4878220932fe67fc46fc7cf7cccdb1ec597 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Thu, 14 Oct 2010 11:22:52 +0200 Subject: KVM: Handle async PF in a guest. When async PF capability is detected hook up special page fault handler that will handle async page fault events and bypass other page faults to regular page fault handler. Also add async PF handling to nested SVM emulation. Async PF always generates exit to L1 where vcpu thread will be scheduled out until page is available. Acked-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 45 ++++++++++++++++++++++++++++++++++++--------- 1 file changed, 36 insertions(+), 9 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index b81a9b7c2ca..93e8120b802 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -31,6 +31,7 @@ #include #include +#include #include #include "trace.h" @@ -133,6 +134,7 @@ struct vcpu_svm { unsigned int3_injected; unsigned long int3_rip; + u32 apf_reason; }; #define MSR_INVALID 0xffffffffU @@ -1383,16 +1385,33 @@ static void svm_set_dr7(struct kvm_vcpu *vcpu, unsigned long value) static int pf_interception(struct vcpu_svm *svm) { - u64 fault_address; + u64 fault_address = svm->vmcb->control.exit_info_2; u32 error_code; + int r = 1; - fault_address = svm->vmcb->control.exit_info_2; - error_code = svm->vmcb->control.exit_info_1; + switch (svm->apf_reason) { + default: + error_code = svm->vmcb->control.exit_info_1; - trace_kvm_page_fault(fault_address, error_code); - if (!npt_enabled && kvm_event_needs_reinjection(&svm->vcpu)) - kvm_mmu_unprotect_page_virt(&svm->vcpu, fault_address); - return kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code); + trace_kvm_page_fault(fault_address, error_code); + if (!npt_enabled && kvm_event_needs_reinjection(&svm->vcpu)) + kvm_mmu_unprotect_page_virt(&svm->vcpu, fault_address); + r = kvm_mmu_page_fault(&svm->vcpu, fault_address, error_code); + break; + case KVM_PV_REASON_PAGE_NOT_PRESENT: + svm->apf_reason = 0; + local_irq_disable(); + kvm_async_pf_task_wait(fault_address); + local_irq_enable(); + break; + case KVM_PV_REASON_PAGE_READY: + svm->apf_reason = 0; + local_irq_disable(); + kvm_async_pf_task_wake(fault_address); + local_irq_enable(); + break; + } + return r; } static int db_interception(struct vcpu_svm *svm) @@ -1836,8 +1855,8 @@ static int nested_svm_exit_special(struct vcpu_svm *svm) return NESTED_EXIT_HOST; break; case SVM_EXIT_EXCP_BASE + PF_VECTOR: - /* When we're shadowing, trap PFs */ - if (!npt_enabled) + /* When we're shadowing, trap PFs, but not async PF */ + if (!npt_enabled && svm->apf_reason == 0) return NESTED_EXIT_HOST; break; case SVM_EXIT_EXCP_BASE + NM_VECTOR: @@ -1893,6 +1912,10 @@ static int nested_svm_intercept(struct vcpu_svm *svm) u32 excp_bits = 1 << (exit_code - SVM_EXIT_EXCP_BASE); if (svm->nested.intercept_exceptions & excp_bits) vmexit = NESTED_EXIT_DONE; + /* async page fault always cause vmexit */ + else if ((exit_code == SVM_EXIT_EXCP_BASE + PF_VECTOR) && + svm->apf_reason != 0) + vmexit = NESTED_EXIT_DONE; break; } case SVM_EXIT_ERR: { @@ -3414,6 +3437,10 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) svm->next_rip = 0; + /* if exit due to PF check for async PF */ + if (svm->vmcb->control.exit_code == SVM_EXIT_EXCP_BASE + PF_VECTOR) + svm->apf_reason = kvm_read_and_reset_pf_reason(); + if (npt_enabled) { vcpu->arch.regs_avail &= ~(1 << VCPU_EXREG_PDPTR); vcpu->arch.regs_dirty &= ~(1 << VCPU_EXREG_PDPTR); -- cgit v1.2.3-18-g5258 From 7c90705bf2a373aa238661bdb6446f27299ef489 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Thu, 14 Oct 2010 11:22:53 +0200 Subject: KVM: Inject asynchronous page fault into a PV guest if page is swapped out. Send async page fault to a PV guest if it accesses swapped out memory. Guest will choose another task to run upon receiving the fault. Allow async page fault injection only when guest is in user mode since otherwise guest may be in non-sleepable context and will not be able to reschedule. Vcpu will be halted if guest will fault on the same page again or if vcpu executes kernel code. Acked-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/mmu.c | 1 + arch/x86/kvm/x86.c | 43 ++++++++++++++++++++++++++++++++++++++----- 2 files changed, 39 insertions(+), 5 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index b2c60986a7c..64f90db369f 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2592,6 +2592,7 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva, int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn) { struct kvm_arch_async_pf arch; + arch.token = (vcpu->arch.apf.id++ << 12) | vcpu->vcpu_id; arch.gfn = gfn; return kvm_setup_async_pf(vcpu, gva, gfn, &arch); diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 063c0729676..ac4c368afd4 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6248,20 +6248,53 @@ static void kvm_del_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn) } } +static int apf_put_user(struct kvm_vcpu *vcpu, u32 val) +{ + + return kvm_write_guest_cached(vcpu->kvm, &vcpu->arch.apf.data, &val, + sizeof(val)); +} + void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) { - trace_kvm_async_pf_not_present(work->gva); - - kvm_make_request(KVM_REQ_APF_HALT, vcpu); + trace_kvm_async_pf_not_present(work->arch.token, work->gva); kvm_add_async_pf_gfn(vcpu, work->arch.gfn); + + if (!(vcpu->arch.apf.msr_val & KVM_ASYNC_PF_ENABLED) || + kvm_x86_ops->get_cpl(vcpu) == 0) + kvm_make_request(KVM_REQ_APF_HALT, vcpu); + else if (!apf_put_user(vcpu, KVM_PV_REASON_PAGE_NOT_PRESENT)) { + vcpu->arch.fault.error_code = 0; + vcpu->arch.fault.address = work->arch.token; + kvm_inject_page_fault(vcpu); + } } void kvm_arch_async_page_present(struct kvm_vcpu *vcpu, struct kvm_async_pf *work) { - trace_kvm_async_pf_ready(work->gva); - kvm_del_async_pf_gfn(vcpu, work->arch.gfn); + trace_kvm_async_pf_ready(work->arch.token, work->gva); + if (is_error_page(work->page)) + work->arch.token = ~0; /* broadcast wakeup */ + else + kvm_del_async_pf_gfn(vcpu, work->arch.gfn); + + if ((vcpu->arch.apf.msr_val & KVM_ASYNC_PF_ENABLED) && + !apf_put_user(vcpu, KVM_PV_REASON_PAGE_READY)) { + vcpu->arch.fault.error_code = 0; + vcpu->arch.fault.address = work->arch.token; + kvm_inject_page_fault(vcpu); + } +} + +bool kvm_arch_can_inject_async_page_present(struct kvm_vcpu *vcpu) +{ + if (!(vcpu->arch.apf.msr_val & KVM_ASYNC_PF_ENABLED)) + return true; + else + return !kvm_event_needs_reinjection(vcpu) && + kvm_x86_ops->interrupt_allowed(vcpu); } EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_exit); -- cgit v1.2.3-18-g5258 From 6adba527420651b6cacaf392541c09fb108711a2 Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Thu, 14 Oct 2010 11:22:55 +0200 Subject: KVM: Let host know whether the guest can handle async PF in non-userspace context. If guest can detect that it runs in non-preemptable context it can handle async PFs at any time, so let host know that it can send async PF even if guest cpu is not in userspace. Acked-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ac4c368afd4..fff70b50725 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1429,8 +1429,8 @@ static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data) { gpa_t gpa = data & ~0x3f; - /* Bits 1:5 are resrved, Should be zero */ - if (data & 0x3e) + /* Bits 2:5 are resrved, Should be zero */ + if (data & 0x3c) return 1; vcpu->arch.apf.msr_val = data; @@ -1444,6 +1444,7 @@ static int kvm_pv_enable_async_pf(struct kvm_vcpu *vcpu, u64 data) if (kvm_gfn_to_hva_cache_init(vcpu->kvm, &vcpu->arch.apf.data, gpa)) return 1; + vcpu->arch.apf.send_user_only = !(data & KVM_ASYNC_PF_SEND_ALWAYS); kvm_async_pf_wakeup_all(vcpu); return 0; } -- cgit v1.2.3-18-g5258 From fc5f06fac6fb8ce469ea173370f2cd398f1d9f9a Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Thu, 14 Oct 2010 11:22:56 +0200 Subject: KVM: Send async PF when guest is not in userspace too. If guest indicates that it can handle async pf in kernel mode too send it, but only if interrupts are enabled. Acked-by: Rik van Riel Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index fff70b50725..c0bd2a2b3c0 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6263,7 +6263,8 @@ void kvm_arch_async_page_not_present(struct kvm_vcpu *vcpu, kvm_add_async_pf_gfn(vcpu, work->arch.gfn); if (!(vcpu->arch.apf.msr_val & KVM_ASYNC_PF_ENABLED) || - kvm_x86_ops->get_cpl(vcpu) == 0) + (vcpu->arch.apf.send_user_only && + kvm_x86_ops->get_cpl(vcpu) == 0)) kvm_make_request(KVM_REQ_APF_HALT, vcpu); else if (!apf_put_user(vcpu, KVM_PV_REASON_PAGE_NOT_PRESENT)) { vcpu->arch.fault.error_code = 0; -- cgit v1.2.3-18-g5258 From 7e1fbeac6f4b52e12ccaa20b151bd76783069dee Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Wed, 20 Oct 2010 15:18:02 +0200 Subject: KVM: x86: Mark kvm_arch_setup_async_pf static It has no user outside mmu.c and also no prototype. Signed-off-by: Jan Kiszka Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 64f90db369f..0aac41fc8ff 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2589,7 +2589,7 @@ static int nonpaging_page_fault(struct kvm_vcpu *vcpu, gva_t gva, error_code & PFERR_WRITE_MASK, gfn); } -int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn) +static int kvm_arch_setup_async_pf(struct kvm_vcpu *vcpu, gva_t gva, gfn_t gfn) { struct kvm_arch_async_pf arch; arch.token = (vcpu->arch.apf.id++ << 12) | vcpu->vcpu_id; -- cgit v1.2.3-18-g5258 From f56f5369561ccafe0d4a2396179f814454a1571c Mon Sep 17 00:00:00 2001 From: Andi Kleen Date: Wed, 20 Oct 2010 17:56:17 +0200 Subject: KVM: Move KVM context switch into own function gcc 4.5 with some special options is able to duplicate the VMX context switch asm in vmx_vcpu_run(). This results in a compile error because the inline asm sequence uses an on local label. The non local label is needed because other code wants to set up the return address. This patch moves the asm code into an own function and marks that explicitely noinline to avoid this problem. Better would be probably to just move it into an .S file. The diff looks worse than the change really is, it's all just code movement and no logic change. Signed-off-by: Andi Kleen Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/vmx.c | 63 ++++++++++++++++++++++++++++++++---------------------- 1 file changed, 38 insertions(+), 25 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 81fcbe9515c..46c89252f82 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3890,33 +3890,17 @@ static void vmx_cancel_injection(struct kvm_vcpu *vcpu) #define Q "l" #endif -static void vmx_vcpu_run(struct kvm_vcpu *vcpu) +/* + * We put this into a separate noinline function to prevent the compiler + * from duplicating the code. This is needed because this code + * uses non local labels that cannot be duplicated. + * Do not put any flow control into this function. + * Better would be to put this whole monstrosity into a .S file. + */ +static void noinline do_vmx_vcpu_run(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); - - /* Record the guest's net vcpu time for enforced NMI injections. */ - if (unlikely(!cpu_has_virtual_nmis() && vmx->soft_vnmi_blocked)) - vmx->entry_time = ktime_get(); - - /* Don't enter VMX if guest state is invalid, let the exit handler - start emulation until we arrive back to a valid state */ - if (vmx->emulation_required && emulate_invalid_guest_state) - return; - - if (test_bit(VCPU_REGS_RSP, (unsigned long *)&vcpu->arch.regs_dirty)) - vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]); - if (test_bit(VCPU_REGS_RIP, (unsigned long *)&vcpu->arch.regs_dirty)) - vmcs_writel(GUEST_RIP, vcpu->arch.regs[VCPU_REGS_RIP]); - - /* When single-stepping over STI and MOV SS, we must clear the - * corresponding interruptibility bits in the guest state. Otherwise - * vmentry fails as it then expects bit 14 (BS) in pending debug - * exceptions being set, but that's not correct for the guest debugging - * case. */ - if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) - vmx_set_interrupt_shadow(vcpu, 0); - - asm( + asm volatile( /* Store host registers */ "push %%"R"dx; push %%"R"bp;" "push %%"R"cx \n\t" @@ -4011,6 +3995,35 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu) , "r8", "r9", "r10", "r11", "r12", "r13", "r14", "r15" #endif ); +} + +static void vmx_vcpu_run(struct kvm_vcpu *vcpu) +{ + struct vcpu_vmx *vmx = to_vmx(vcpu); + + /* Record the guest's net vcpu time for enforced NMI injections. */ + if (unlikely(!cpu_has_virtual_nmis() && vmx->soft_vnmi_blocked)) + vmx->entry_time = ktime_get(); + + /* Don't enter VMX if guest state is invalid, let the exit handler + start emulation until we arrive back to a valid state */ + if (vmx->emulation_required && emulate_invalid_guest_state) + return; + + if (test_bit(VCPU_REGS_RSP, (unsigned long *)&vcpu->arch.regs_dirty)) + vmcs_writel(GUEST_RSP, vcpu->arch.regs[VCPU_REGS_RSP]); + if (test_bit(VCPU_REGS_RIP, (unsigned long *)&vcpu->arch.regs_dirty)) + vmcs_writel(GUEST_RIP, vcpu->arch.regs[VCPU_REGS_RIP]); + + /* When single-stepping over STI and MOV SS, we must clear the + * corresponding interruptibility bits in the guest state. Otherwise + * vmentry fails as it then expects bit 14 (BS) in pending debug + * exceptions being set, but that's not correct for the guest debugging + * case. */ + if (vcpu->guest_debug & KVM_GUESTDBG_SINGLESTEP) + vmx_set_interrupt_shadow(vcpu, 0); + + do_vmx_vcpu_run(vcpu); vcpu->arch.regs_avail = ~((1 << VCPU_REGS_RIP) | (1 << VCPU_REGS_RSP) | (1 << VCPU_EXREG_PDPTR)); -- cgit v1.2.3-18-g5258 From 13c34e073bec75aa4c8d1bb80c10fd7291f3423a Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 21 Oct 2010 12:20:31 +0200 Subject: KVM: SVM: Move guest register save out of interrupts disabled section Saving guest registers is just a memory copy, and does not need to be in the critical section. Move outside the critical section to improve latency a bit. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 10 +++++----- 1 file changed, 5 insertions(+), 5 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 93e8120b802..80354965f7f 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -3412,11 +3412,6 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) #endif ); - vcpu->arch.cr2 = svm->vmcb->save.cr2; - vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; - vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; - vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; - load_host_msrs(vcpu); kvm_load_ldt(ldt_selector); loadsegment(fs, fs_selector); @@ -3433,6 +3428,11 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) stgi(); + vcpu->arch.cr2 = svm->vmcb->save.cr2; + vcpu->arch.regs[VCPU_REGS_RAX] = svm->vmcb->save.rax; + vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp; + vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip; + sync_cr8_to_lapic(vcpu); svm->next_rip = 0; -- cgit v1.2.3-18-g5258 From afe9e66f8233e33e16fcc5b855070e45978f919e Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 21 Oct 2010 12:20:32 +0200 Subject: KVM: SVM: Move svm->host_gs_base into a separate structure More members will join it soon. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 80354965f7f..a217978b370 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -124,7 +124,9 @@ struct vcpu_svm { u64 next_rip; u64 host_user_msrs[NR_HOST_SAVE_USER_MSRS]; - u64 host_gs_base; + struct { + u64 gs_base; + } host; u32 *msrpm; @@ -1353,14 +1355,14 @@ static void svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) static void load_host_msrs(struct kvm_vcpu *vcpu) { #ifdef CONFIG_X86_64 - wrmsrl(MSR_GS_BASE, to_svm(vcpu)->host_gs_base); + wrmsrl(MSR_GS_BASE, to_svm(vcpu)->host.gs_base); #endif } static void save_host_msrs(struct kvm_vcpu *vcpu) { #ifdef CONFIG_X86_64 - rdmsrl(MSR_GS_BASE, to_svm(vcpu)->host_gs_base); + rdmsrl(MSR_GS_BASE, to_svm(vcpu)->host.gs_base); #endif } -- cgit v1.2.3-18-g5258 From dacccfdd6bb386ec0714e1fec250c4b7d0aaccc9 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 21 Oct 2010 12:20:33 +0200 Subject: KVM: SVM: Move fs/gs/ldt save/restore to heavyweight exit path ldt is never used in the kernel context; same goes for fs (x86_64) and gs (i386). So save/restore them in the heavyweight exit path instead of the lightweight path. By itself, this doesn't buy us much, but it paves the way for moving vmload and vmsave to the heavyweight exit path, since they modify the same registers. [jan: fix copy/pase mistake on i386] Signed-off-by: Avi Kivity Signed-off-by: Jan Kiszka Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 35 +++++++++++++++++++++-------------- 1 file changed, 21 insertions(+), 14 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index a217978b370..8ea4a79bc4e 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -125,6 +125,9 @@ struct vcpu_svm { u64 host_user_msrs[NR_HOST_SAVE_USER_MSRS]; struct { + u16 fs; + u16 gs; + u16 ldt; u64 gs_base; } host; @@ -184,6 +187,9 @@ static int nested_svm_vmexit(struct vcpu_svm *svm); static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr, bool has_error_code, u32 error_code); +static void save_host_msrs(struct kvm_vcpu *vcpu); +static void load_host_msrs(struct kvm_vcpu *vcpu); + static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu) { return container_of(vcpu, struct vcpu_svm, vcpu); @@ -996,6 +1002,11 @@ static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) svm->asid_generation = 0; } + save_host_msrs(vcpu); + savesegment(fs, svm->host.fs); + savesegment(gs, svm->host.gs); + svm->host.ldt = kvm_read_ldt(); + for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++) rdmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]); } @@ -1006,6 +1017,14 @@ static void svm_vcpu_put(struct kvm_vcpu *vcpu) int i; ++vcpu->stat.host_state_reload; + kvm_load_ldt(svm->host.ldt); +#ifdef CONFIG_X86_64 + loadsegment(fs, svm->host.fs); + load_gs_index(svm->host.gs); + wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs); +#else + loadsegment(gs, svm->host.gs); +#endif for (i = 0; i < NR_HOST_SAVE_USER_MSRS; i++) wrmsrl(host_save_user_msrs[i], svm->host_user_msrs[i]); } @@ -3314,9 +3333,6 @@ static void svm_cancel_injection(struct kvm_vcpu *vcpu) static void svm_vcpu_run(struct kvm_vcpu *vcpu) { struct vcpu_svm *svm = to_svm(vcpu); - u16 fs_selector; - u16 gs_selector; - u16 ldt_selector; svm->vmcb->save.rax = vcpu->arch.regs[VCPU_REGS_RAX]; svm->vmcb->save.rsp = vcpu->arch.regs[VCPU_REGS_RSP]; @@ -3333,10 +3349,6 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) sync_lapic_to_cr8(vcpu); - save_host_msrs(vcpu); - savesegment(fs, fs_selector); - savesegment(gs, gs_selector); - ldt_selector = kvm_read_ldt(); svm->vmcb->save.cr2 = vcpu->arch.cr2; clgi(); @@ -3415,13 +3427,8 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) ); load_host_msrs(vcpu); - kvm_load_ldt(ldt_selector); - loadsegment(fs, fs_selector); -#ifdef CONFIG_X86_64 - load_gs_index(gs_selector); - wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs); -#else - loadsegment(gs, gs_selector); +#ifndef CONFIG_X86_64 + loadsegment(fs, svm->host.fs); #endif reload_tss(vcpu); -- cgit v1.2.3-18-g5258 From 82ca2d108b493f7d916975805668bef2096f1147 Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Thu, 21 Oct 2010 12:20:34 +0200 Subject: KVM: SVM: Fold save_host_msrs() and load_host_msrs() into their callers This abstraction only serves to obfuscate. Remove. Signed-off-by: Avi Kivity Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/svm.c | 26 ++++++-------------------- 1 file changed, 6 insertions(+), 20 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 8ea4a79bc4e..cf47a2fc24e 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -187,9 +187,6 @@ static int nested_svm_vmexit(struct vcpu_svm *svm); static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr, bool has_error_code, u32 error_code); -static void save_host_msrs(struct kvm_vcpu *vcpu); -static void load_host_msrs(struct kvm_vcpu *vcpu); - static inline struct vcpu_svm *to_svm(struct kvm_vcpu *vcpu) { return container_of(vcpu, struct vcpu_svm, vcpu); @@ -1002,7 +999,9 @@ static void svm_vcpu_load(struct kvm_vcpu *vcpu, int cpu) svm->asid_generation = 0; } - save_host_msrs(vcpu); +#ifdef CONFIG_X86_64 + rdmsrl(MSR_GS_BASE, to_svm(vcpu)->host.gs_base); +#endif savesegment(fs, svm->host.fs); savesegment(gs, svm->host.gs); svm->host.ldt = kvm_read_ldt(); @@ -1371,20 +1370,6 @@ static void svm_guest_debug(struct kvm_vcpu *vcpu, struct kvm_guest_debug *dbg) update_db_intercept(vcpu); } -static void load_host_msrs(struct kvm_vcpu *vcpu) -{ -#ifdef CONFIG_X86_64 - wrmsrl(MSR_GS_BASE, to_svm(vcpu)->host.gs_base); -#endif -} - -static void save_host_msrs(struct kvm_vcpu *vcpu) -{ -#ifdef CONFIG_X86_64 - rdmsrl(MSR_GS_BASE, to_svm(vcpu)->host.gs_base); -#endif -} - static void new_asid(struct vcpu_svm *svm, struct svm_cpu_data *sd) { if (sd->next_asid > sd->max_asid) { @@ -3426,8 +3411,9 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) #endif ); - load_host_msrs(vcpu); -#ifndef CONFIG_X86_64 +#ifdef CONFIG_X86_64 + wrmsrl(MSR_GS_BASE, svm->host.gs_base); +#else loadsegment(fs, svm->host.fs); #endif -- cgit v1.2.3-18-g5258 From 83bcacb1a548d4d8df532376c981277761622bce Mon Sep 17 00:00:00 2001 From: Avi Kivity Date: Mon, 25 Oct 2010 15:23:55 +0200 Subject: KVM: Avoid double interrupt injection with vapic After an interrupt injection, the PPR changes, and we have to reflect that into the vapic. This causes a KVM_REQ_EVENT to be set, which causes the whole interrupt injection routine to be run again (harmlessly). Optimize by only setting KVM_REQ_EVENT if the ppr was lowered; otherwise there is no chance that a new injection is needed. Signed-off-by: Avi Kivity --- arch/x86/kvm/lapic.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index 413f8973a85..93cf9d0d365 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -277,7 +277,8 @@ static void apic_update_ppr(struct kvm_lapic *apic) if (old_ppr != ppr) { apic_set_reg(apic, APIC_PROCPRI, ppr); - kvm_make_request(KVM_REQ_EVENT, apic->vcpu); + if (ppr < old_ppr) + kvm_make_request(KVM_REQ_EVENT, apic->vcpu); } } -- cgit v1.2.3-18-g5258 From ff1fcb9ebd53ee3f21ae117e6952204e465f46d8 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 22 Oct 2010 14:18:15 -0200 Subject: KVM: VMX: remove setting of shadow_base_ptes for EPT The EPT present/writable bits use the same position as normal pagetable bits. Since direct_map passes ACC_ALL to mmu_set_spte, thus always setting the writable bit on sptes, use the generic PT_PRESENT shadow_base_pte. Also pass present/writable error code information from EPT violation to generic pagefault handler. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.c | 4 +--- 1 file changed, 1 insertion(+), 3 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 46c89252f82..e42727b305c 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3476,7 +3476,7 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu) gpa = vmcs_read64(GUEST_PHYSICAL_ADDRESS); trace_kvm_page_fault(gpa, exit_qualification); - return kvm_mmu_page_fault(vcpu, gpa & PAGE_MASK, 0); + return kvm_mmu_page_fault(vcpu, gpa, exit_qualification & 0x3); } static u64 ept_rsvd_mask(u64 spte, int level) @@ -4409,8 +4409,6 @@ static int __init vmx_init(void) if (enable_ept) { bypass_guest_pf = 0; - kvm_mmu_set_base_ptes(VMX_EPT_READABLE_MASK | - VMX_EPT_WRITABLE_MASK); kvm_mmu_set_mask_ptes(0ull, 0ull, 0ull, 0ull, VMX_EPT_EXECUTABLE_MASK); kvm_enable_tdp(); -- cgit v1.2.3-18-g5258 From 982c25658c066fbbeb42e44a6a8fcd59ae008837 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 22 Oct 2010 14:18:16 -0200 Subject: KVM: MMU: remove kvm_mmu_set_base_ptes Unused. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 9 +-------- arch/x86/kvm/x86.c | 1 - 2 files changed, 1 insertion(+), 9 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 0aac41fc8ff..11b9102f411 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -196,7 +196,6 @@ static struct percpu_counter kvm_total_used_mmu_pages; static u64 __read_mostly shadow_trap_nonpresent_pte; static u64 __read_mostly shadow_notrap_nonpresent_pte; -static u64 __read_mostly shadow_base_present_pte; static u64 __read_mostly shadow_nx_mask; static u64 __read_mostly shadow_x_mask; /* mutual exclusive with nx_mask */ static u64 __read_mostly shadow_user_mask; @@ -215,12 +214,6 @@ void kvm_mmu_set_nonpresent_ptes(u64 trap_pte, u64 notrap_pte) } EXPORT_SYMBOL_GPL(kvm_mmu_set_nonpresent_ptes); -void kvm_mmu_set_base_ptes(u64 base_pte) -{ - shadow_base_present_pte = base_pte; -} -EXPORT_SYMBOL_GPL(kvm_mmu_set_base_ptes); - void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, u64 dirty_mask, u64 nx_mask, u64 x_mask) { @@ -1975,7 +1968,7 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, * whether the guest actually used the pte (in order to detect * demand paging). */ - spte = shadow_base_present_pte; + spte = PT_PRESENT_MASK; if (!speculative) spte |= shadow_accessed_mask; if (!dirty) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index c0bd2a2b3c0..a2a78547243 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4693,7 +4693,6 @@ int kvm_arch_init(void *opaque) kvm_x86_ops = ops; kvm_mmu_set_nonpresent_ptes(0ull, 0ull); - kvm_mmu_set_base_ptes(PT_PRESENT_MASK); kvm_mmu_set_mask_ptes(PT_USER_MASK, PT_ACCESSED_MASK, PT_DIRTY_MASK, PT64_NX_MASK, 0); -- cgit v1.2.3-18-g5258 From 7905d9a5ad7a83f1c1c00559839857ab90afbdfc Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 22 Oct 2010 14:18:17 -0200 Subject: KVM: MMU: flush TLBs on writable -> read-only spte overwrite This can happen in the following scenario: vcpu0 vcpu1 read fault gup(.write=0) gup(.write=1) reuse swap cache, no COW set writable spte use writable spte set read-only spte Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 10 ++++++++++ 1 file changed, 10 insertions(+) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 11b9102f411..99433943170 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2069,6 +2069,16 @@ static void mmu_set_spte(struct kvm_vcpu *vcpu, u64 *sptep, spte_to_pfn(*sptep), pfn); drop_spte(vcpu->kvm, sptep, shadow_trap_nonpresent_pte); kvm_flush_remote_tlbs(vcpu->kvm); + /* + * If we overwrite a writable spte with a read-only one, + * drop it and flush remote TLBs. Otherwise rmap_write_protect + * will find a read-only spte, even though the writable spte + * might be cached on a CPU's TLB. + */ + } else if (is_writable_pte(*sptep) && + (!(pte_access & ACC_WRITE_MASK) || !dirty)) { + drop_spte(vcpu->kvm, sptep, shadow_trap_nonpresent_pte); + kvm_flush_remote_tlbs(vcpu->kvm); } else was_rmapped = 1; } -- cgit v1.2.3-18-g5258 From 612819c3c6e67bac8fceaa7cc402f13b1b63f7e4 Mon Sep 17 00:00:00 2001 From: Marcelo Tosatti Date: Fri, 22 Oct 2010 14:18:18 -0200 Subject: KVM: propagate fault r/w information to gup(), allow read-only memory As suggested by Andrea, pass r/w error code to gup(), upgrading read fault to writable if host pte allows it. Signed-off-by: Marcelo Tosatti Signed-off-by: Avi Kivity --- arch/x86/kvm/mmu.c | 27 +++++++++++++++++---------- arch/x86/kvm/paging_tmpl.h | 13 +++++++++---- 2 files changed, 26 insertions(+), 14 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 99433943170..53509f5973d 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2216,7 +2216,7 @@ static void direct_pte_prefetch(struct kvm_vcpu *vcpu, u64 *sptep) } static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, - int level, gfn_t gfn, pfn_t pfn) + int map_writable, int level, gfn_t gfn, pfn_t pfn) { struct kvm_shadow_walk_iterator iterator; struct kvm_mmu_page *sp; @@ -2225,9 +2225,13 @@ static int __direct_map(struct kvm_vcpu *vcpu, gpa_t v, int write, for_each_shadow_entry(vcpu, (u64)gfn << PAGE_SHIFT, iterator) { if (iterator.level == level) { - mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, ACC_ALL, + unsigned pte_access = ACC_ALL; + + if (!map_writable) + pte_access &= ~ACC_WRITE_MASK; + mmu_set_spte(vcpu, iterator.sptep, ACC_ALL, pte_access, 0, write, 1, &pt_write, - level, gfn, pfn, false, true); + level, gfn, pfn, false, map_writable); direct_pte_prefetch(vcpu, iterator.sptep); ++vcpu->stat.pf_fixed; break; @@ -2288,6 +2292,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) int level; pfn_t pfn; unsigned long mmu_seq; + bool map_writable; level = mapping_level(vcpu, gfn); @@ -2302,7 +2307,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); - pfn = gfn_to_pfn(vcpu->kvm, gfn); + pfn = gfn_to_pfn_prot(vcpu->kvm, gfn, write, &map_writable); /* mmio */ if (is_error_pfn(pfn)) @@ -2312,7 +2317,7 @@ static int nonpaging_map(struct kvm_vcpu *vcpu, gva_t v, int write, gfn_t gfn) if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; kvm_mmu_free_some_pages(vcpu); - r = __direct_map(vcpu, v, write, level, gfn, pfn); + r = __direct_map(vcpu, v, write, map_writable, level, gfn, pfn); spin_unlock(&vcpu->kvm->mmu_lock); @@ -2611,11 +2616,11 @@ static bool can_do_async_pf(struct kvm_vcpu *vcpu) } static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn, - gva_t gva, pfn_t *pfn) + gva_t gva, pfn_t *pfn, bool write, bool *writable) { bool async; - *pfn = gfn_to_pfn_async(vcpu->kvm, gfn, &async); + *pfn = gfn_to_pfn_async(vcpu->kvm, gfn, &async, write, writable); if (!async) return false; /* *pfn has correct page already */ @@ -2632,7 +2637,7 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn, return true; } - *pfn = gfn_to_pfn(vcpu->kvm, gfn); + *pfn = gfn_to_pfn_prot(vcpu->kvm, gfn, write, writable); return false; } @@ -2645,6 +2650,8 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, int level; gfn_t gfn = gpa >> PAGE_SHIFT; unsigned long mmu_seq; + int write = error_code & PFERR_WRITE_MASK; + bool map_writable; ASSERT(vcpu); ASSERT(VALID_PAGE(vcpu->arch.mmu.root_hpa)); @@ -2660,7 +2667,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); - if (try_async_pf(vcpu, no_apf, gfn, gpa, &pfn)) + if (try_async_pf(vcpu, no_apf, gfn, gpa, &pfn, write, &map_writable)) return 0; /* mmio */ @@ -2670,7 +2677,7 @@ static int tdp_page_fault(struct kvm_vcpu *vcpu, gva_t gpa, u32 error_code, if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; kvm_mmu_free_some_pages(vcpu); - r = __direct_map(vcpu, gpa, error_code & PFERR_WRITE_MASK, + r = __direct_map(vcpu, gpa, write, map_writable, level, gfn, pfn); spin_unlock(&vcpu->kvm->mmu_lock); diff --git a/arch/x86/kvm/paging_tmpl.h b/arch/x86/kvm/paging_tmpl.h index d6b281e989b..ba00eefa7bc 100644 --- a/arch/x86/kvm/paging_tmpl.h +++ b/arch/x86/kvm/paging_tmpl.h @@ -427,7 +427,7 @@ static void FNAME(pte_prefetch)(struct kvm_vcpu *vcpu, struct guest_walker *gw, static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, struct guest_walker *gw, int user_fault, int write_fault, int hlevel, - int *ptwrite, pfn_t pfn) + int *ptwrite, pfn_t pfn, bool map_writable) { unsigned access = gw->pt_access; struct kvm_mmu_page *sp = NULL; @@ -501,7 +501,7 @@ static u64 *FNAME(fetch)(struct kvm_vcpu *vcpu, gva_t addr, mmu_set_spte(vcpu, it.sptep, access, gw->pte_access & access, user_fault, write_fault, dirty, ptwrite, it.level, - gw->gfn, pfn, false, true); + gw->gfn, pfn, false, map_writable); FNAME(pte_prefetch)(vcpu, gw, it.sptep); return it.sptep; @@ -539,6 +539,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, pfn_t pfn; int level = PT_PAGE_TABLE_LEVEL; unsigned long mmu_seq; + bool map_writable; pgprintk("%s: addr %lx err %x\n", __func__, addr, error_code); @@ -569,13 +570,17 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, mmu_seq = vcpu->kvm->mmu_notifier_seq; smp_rmb(); - if (try_async_pf(vcpu, no_apf, walker.gfn, addr, &pfn)) + if (try_async_pf(vcpu, no_apf, walker.gfn, addr, &pfn, write_fault, + &map_writable)) return 0; /* mmio */ if (is_error_pfn(pfn)) return kvm_handle_bad_page(vcpu->kvm, walker.gfn, pfn); + if (!map_writable) + walker.pte_access &= ~ACC_WRITE_MASK; + spin_lock(&vcpu->kvm->mmu_lock); if (mmu_notifier_retry(vcpu, mmu_seq)) goto out_unlock; @@ -583,7 +588,7 @@ static int FNAME(page_fault)(struct kvm_vcpu *vcpu, gva_t addr, u32 error_code, trace_kvm_mmu_audit(vcpu, AUDIT_PRE_PAGE_FAULT); kvm_mmu_free_some_pages(vcpu); sptep = FNAME(fetch)(vcpu, addr, &walker, user_fault, write_fault, - level, &write_pt, pfn); + level, &write_pt, pfn, map_writable); (void)sptep; pgprintk("%s: shadow pte %p %llx ptwrite %d\n", __func__, sptep, *sptep, write_pt); -- cgit v1.2.3-18-g5258 From 515a01279a187415322a80736800a7d6325876ab Mon Sep 17 00:00:00 2001 From: Takuya Yoshikawa Date: Wed, 27 Oct 2010 18:23:54 +0900 Subject: KVM: pre-allocate one more dirty bitmap to avoid vmalloc() Currently x86's kvm_vm_ioctl_get_dirty_log() needs to allocate a bitmap by vmalloc() which will be used in the next logging and this has been causing bad effect to VGA and live-migration: vmalloc() consumes extra systime, triggers tlb flush, etc. This patch resolves this issue by pre-allocating one more bitmap and switching between two bitmaps during dirty logging. Performance improvement: I measured performance for the case of VGA update by trace-cmd. The result was 1.5 times faster than the original one. In the case of live migration, the improvement ratio depends on the workload and the guest memory size. In general, the larger the memory size is the more benefits we get. Note: This does not change other architectures's logic but the allocation size becomes twice. This will increase the actual memory consumption only when the new size changes the number of pages allocated by vmalloc(). Signed-off-by: Takuya Yoshikawa Signed-off-by: Fernando Luis Vazquez Cao Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index a2a78547243..35f82f2c66f 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -3208,18 +3208,15 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, struct kvm_memslots *slots, *old_slots; unsigned long *dirty_bitmap; - r = -ENOMEM; - dirty_bitmap = vmalloc(n); - if (!dirty_bitmap) - goto out; + dirty_bitmap = memslot->dirty_bitmap_head; + if (memslot->dirty_bitmap == dirty_bitmap) + dirty_bitmap += n / sizeof(long); memset(dirty_bitmap, 0, n); r = -ENOMEM; slots = kzalloc(sizeof(struct kvm_memslots), GFP_KERNEL); - if (!slots) { - vfree(dirty_bitmap); + if (!slots) goto out; - } memcpy(slots, kvm->memslots, sizeof(struct kvm_memslots)); slots->memslots[log->slot].dirty_bitmap = dirty_bitmap; slots->generation++; @@ -3235,11 +3232,8 @@ int kvm_vm_ioctl_get_dirty_log(struct kvm *kvm, spin_unlock(&kvm->mmu_lock); r = -EFAULT; - if (copy_to_user(log->dirty_bitmap, dirty_bitmap, n)) { - vfree(dirty_bitmap); + if (copy_to_user(log->dirty_bitmap, dirty_bitmap, n)) goto out; - } - vfree(dirty_bitmap); } else { r = -EFAULT; if (clear_user(log->dirty_bitmap, n)) -- cgit v1.2.3-18-g5258 From 2eec73437487aa690882cafddca6e4d93df46f26 Mon Sep 17 00:00:00 2001 From: Jan Kiszka Date: Mon, 1 Nov 2010 14:01:29 +0100 Subject: KVM: x86: Avoid issuing wbinvd twice Micro optimization to avoid calling wbinvd twice on the CPU that has to emulate it. As we might be preempted between smp_call_function_many and the local wbinvd, the cache might be filled again so that real work could be done uselessly. Signed-off-by: Jan Kiszka Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 35f82f2c66f..c10135bc0f2 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4007,13 +4007,15 @@ int kvm_emulate_wbinvd(struct kvm_vcpu *vcpu) return X86EMUL_CONTINUE; if (kvm_x86_ops->has_wbinvd_exit()) { - preempt_disable(); + int cpu = get_cpu(); + + cpumask_set_cpu(cpu, vcpu->arch.wbinvd_dirty_mask); smp_call_function_many(vcpu->arch.wbinvd_dirty_mask, wbinvd_ipi, NULL, 1); - preempt_enable(); + put_cpu(); cpumask_clear(vcpu->arch.wbinvd_dirty_mask); - } - wbinvd(); + } else + wbinvd(); return X86EMUL_CONTINUE; } EXPORT_SYMBOL_GPL(kvm_emulate_wbinvd); -- cgit v1.2.3-18-g5258 From ec25d5e66ee152e371fd7046f3f8441859579aea Mon Sep 17 00:00:00 2001 From: Gleb Natapov Date: Mon, 1 Nov 2010 15:35:01 +0200 Subject: KVM: handle exit due to INVD in VMX Currently the exit is unhandled, so guest halts with error if it tries to execute INVD instruction. Call into emulator when INVD instruction is executed by a guest instead. This instruction is not needed by ordinary guests, but firmware (like OpenBIOS) use it and fail. Signed-off-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/vmx.c | 6 ++++++ 1 file changed, 6 insertions(+) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index e42727b305c..12c30733e23 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -3349,6 +3349,11 @@ static int handle_vmx_insn(struct kvm_vcpu *vcpu) return 1; } +static int handle_invd(struct kvm_vcpu *vcpu) +{ + return emulate_instruction(vcpu, 0, 0, 0) == EMULATE_DONE; +} + static int handle_invlpg(struct kvm_vcpu *vcpu) { unsigned long exit_qualification = vmcs_readl(EXIT_QUALIFICATION); @@ -3649,6 +3654,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu) = { [EXIT_REASON_MSR_WRITE] = handle_wrmsr, [EXIT_REASON_PENDING_INTERRUPT] = handle_interrupt_window, [EXIT_REASON_HLT] = handle_halt, + [EXIT_REASON_INVD] = handle_invd, [EXIT_REASON_INVLPG] = handle_invlpg, [EXIT_REASON_VMCALL] = handle_vmcall, [EXIT_REASON_VMCLEAR] = handle_vmx_insn, -- cgit v1.2.3-18-g5258 From c9b263d2be9c535b410f6617710534f798bf0ff0 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Mon, 1 Nov 2010 16:58:43 +0800 Subject: KVM: fix tracing kvm_try_async_get_page Tracing 'async' and *pfn is useless, since 'async' is always true, and '*pfn' is always "fault_pfn' We can trace 'gva' and 'gfn' instead, it can help us to see the life-cycle of an async_pf Signed-off-by: Xiao Guangrong Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/mmu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index 53509f5973d..272e3020003 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -2628,7 +2628,7 @@ static bool try_async_pf(struct kvm_vcpu *vcpu, bool no_apf, gfn_t gfn, put_page(pfn_to_page(*pfn)); if (!no_apf && can_do_async_pf(vcpu)) { - trace_kvm_try_async_get_page(async, *pfn); + trace_kvm_try_async_get_page(gva, gfn); if (kvm_find_async_pf_gfn(vcpu, gfn)) { trace_kvm_async_pf_doublefault(gva, gfn); kvm_make_request(KVM_REQ_APF_HALT, vcpu); -- cgit v1.2.3-18-g5258 From c7d28c24042f7969adc29147d6c0a190192a6808 Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Mon, 1 Nov 2010 17:00:30 +0800 Subject: KVM: fix searching async gfn in kvm_async_pf_gfn_slot Don't search later slots if the slot is empty Acked-by: Gleb Natapov Signed-off-by: Xiao Guangrong Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index c10135bc0f2..ab10a6ca154 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6209,8 +6209,8 @@ static u32 kvm_async_pf_gfn_slot(struct kvm_vcpu *vcpu, gfn_t gfn) u32 key = kvm_async_pf_hash_fn(gfn); for (i = 0; i < roundup_pow_of_two(ASYNC_PF_PER_VCPU) && - (vcpu->arch.apf.gfns[key] != gfn || - vcpu->arch.apf.gfns[key] == ~0); i++) + (vcpu->arch.apf.gfns[key] != gfn && + vcpu->arch.apf.gfns[key] != ~0); i++) key = kvm_async_pf_next_probe(key); return key; -- cgit v1.2.3-18-g5258 From e6d53e3b0db7ae3641f01a2b2af1470fda86d10c Mon Sep 17 00:00:00 2001 From: Xiao Guangrong Date: Mon, 1 Nov 2010 17:01:28 +0800 Subject: KVM: avoid unnecessary wait for a async pf In current code, it checks async pf completion out of the wait context, like this: if (vcpu->arch.mp_state == KVM_MP_STATE_RUNNABLE && !vcpu->arch.apf.halted) r = vcpu_enter_guest(vcpu); else { ...... kvm_vcpu_block(vcpu) ^- waiting until 'async_pf.done' is not empty } kvm_check_async_pf_completion(vcpu) ^- delete list from async_pf.done So, if we check aysnc pf completion first, it can be blocked at kvm_vcpu_block Fixed by mark the vcpu is unhalted in kvm_check_async_pf_completion() path Signed-off-by: Xiao Guangrong Acked-by: Gleb Natapov Signed-off-by: Marcelo Tosatti --- arch/x86/kvm/x86.c | 1 + 1 file changed, 1 insertion(+) (limited to 'arch/x86/kvm') diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index ab10a6ca154..5daead83366 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -6283,6 +6283,7 @@ void kvm_arch_async_page_present(struct kvm_vcpu *vcpu, vcpu->arch.fault.address = work->arch.token; kvm_inject_page_fault(vcpu); } + vcpu->arch.apf.halted = false; } bool kvm_arch