diff options
Diffstat (limited to 'arch/x86')
33 files changed, 672 insertions, 410 deletions
diff --git a/arch/x86/include/asm/compat.h b/arch/x86/include/asm/compat.h index 24118c0b4640..5343c19814b3 100644 --- a/arch/x86/include/asm/compat.h +++ b/arch/x86/include/asm/compat.h @@ -116,7 +116,6 @@ struct compat_statfs { int f_spare[4]; }; -#define COMPAT_RLIM_OLD_INFINITY 0x7fffffff #define COMPAT_RLIM_INFINITY 0xffffffff typedef u32 compat_old_sigset_t; /* at least 32 bits */ diff --git a/arch/x86/include/asm/dma-mapping.h b/arch/x86/include/asm/dma-mapping.h index 08a0838b83fb..398c79889f5c 100644 --- a/arch/x86/include/asm/dma-mapping.h +++ b/arch/x86/include/asm/dma-mapping.h @@ -19,8 +19,6 @@ # define ISA_DMA_BIT_MASK DMA_BIT_MASK(32) #endif -#define DMA_ERROR_CODE 0 - extern int iommu_merge; extern struct device x86_dma_fallback_dev; extern int panic_on_overflow; @@ -35,9 +33,6 @@ static inline const struct dma_map_ops *get_arch_dma_ops(struct bus_type *bus) bool arch_dma_alloc_attrs(struct device **dev, gfp_t *gfp); #define arch_dma_alloc_attrs arch_dma_alloc_attrs -#define HAVE_ARCH_DMA_SUPPORTED 1 -extern int dma_supported(struct device *hwdev, u64 mask); - extern void *dma_generic_alloc_coherent(struct device *dev, size_t size, dma_addr_t *dma_addr, gfp_t flag, unsigned long attrs); diff --git a/arch/x86/include/asm/iommu.h b/arch/x86/include/asm/iommu.h index 793869879464..fca144a104e4 100644 --- a/arch/x86/include/asm/iommu.h +++ b/arch/x86/include/asm/iommu.h @@ -6,6 +6,8 @@ extern int force_iommu, no_iommu; extern int iommu_detected; extern int iommu_pass_through; +int x86_dma_supported(struct device *dev, u64 mask); + /* 10 seconds */ #define DMAR_OPERATION_TIMEOUT ((cycles_t) tsc_khz*10*1000) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 695605eb1dfb..1588e9e3dc01 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -48,28 +48,31 @@ #define KVM_IRQCHIP_NUM_PINS KVM_IOAPIC_NUM_PINS /* x86-specific vcpu->requests bit members */ -#define KVM_REQ_MIGRATE_TIMER 8 -#define KVM_REQ_REPORT_TPR_ACCESS 9 -#define KVM_REQ_TRIPLE_FAULT 10 -#define KVM_REQ_MMU_SYNC 11 -#define KVM_REQ_CLOCK_UPDATE 12 -#define KVM_REQ_EVENT 14 -#define KVM_REQ_APF_HALT 15 -#define KVM_REQ_STEAL_UPDATE 16 -#define KVM_REQ_NMI 17 -#define KVM_REQ_PMU 18 -#define KVM_REQ_PMI 19 -#define KVM_REQ_SMI 20 -#define KVM_REQ_MASTERCLOCK_UPDATE 21 -#define KVM_REQ_MCLOCK_INPROGRESS (22 | KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP) -#define KVM_REQ_SCAN_IOAPIC (23 | KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP) -#define KVM_REQ_GLOBAL_CLOCK_UPDATE 24 -#define KVM_REQ_APIC_PAGE_RELOAD (25 | KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP) -#define KVM_REQ_HV_CRASH 26 -#define KVM_REQ_IOAPIC_EOI_EXIT 27 -#define KVM_REQ_HV_RESET 28 -#define KVM_REQ_HV_EXIT 29 -#define KVM_REQ_HV_STIMER 30 +#define KVM_REQ_MIGRATE_TIMER KVM_ARCH_REQ(0) +#define KVM_REQ_REPORT_TPR_ACCESS KVM_ARCH_REQ(1) +#define KVM_REQ_TRIPLE_FAULT KVM_ARCH_REQ(2) +#define KVM_REQ_MMU_SYNC KVM_ARCH_REQ(3) +#define KVM_REQ_CLOCK_UPDATE KVM_ARCH_REQ(4) +#define KVM_REQ_EVENT KVM_ARCH_REQ(6) +#define KVM_REQ_APF_HALT KVM_ARCH_REQ(7) +#define KVM_REQ_STEAL_UPDATE KVM_ARCH_REQ(8) +#define KVM_REQ_NMI KVM_ARCH_REQ(9) +#define KVM_REQ_PMU KVM_ARCH_REQ(10) +#define KVM_REQ_PMI KVM_ARCH_REQ(11) +#define KVM_REQ_SMI KVM_ARCH_REQ(12) +#define KVM_REQ_MASTERCLOCK_UPDATE KVM_ARCH_REQ(13) +#define KVM_REQ_MCLOCK_INPROGRESS \ + KVM_ARCH_REQ_FLAGS(14, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP) +#define KVM_REQ_SCAN_IOAPIC \ + KVM_ARCH_REQ_FLAGS(15, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP) +#define KVM_REQ_GLOBAL_CLOCK_UPDATE KVM_ARCH_REQ(16) +#define KVM_REQ_APIC_PAGE_RELOAD \ + KVM_ARCH_REQ_FLAGS(17, KVM_REQUEST_WAIT | KVM_REQUEST_NO_WAKEUP) +#define KVM_REQ_HV_CRASH KVM_ARCH_REQ(18) +#define KVM_REQ_IOAPIC_EOI_EXIT KVM_ARCH_REQ(19) +#define KVM_REQ_HV_RESET KVM_ARCH_REQ(20) +#define KVM_REQ_HV_EXIT KVM_ARCH_REQ(21) +#define KVM_REQ_HV_STIMER KVM_ARCH_REQ(22) #define CR0_RESERVED_BITS \ (~(unsigned long)(X86_CR0_PE | X86_CR0_MP | X86_CR0_EM | X86_CR0_TS \ @@ -254,7 +257,8 @@ union kvm_mmu_page_role { unsigned cr0_wp:1; unsigned smep_andnot_wp:1; unsigned smap_andnot_wp:1; - unsigned :8; + unsigned ad_disabled:1; + unsigned :7; /* * This is left at the top of the word so that diff --git a/arch/x86/include/asm/msr-index.h b/arch/x86/include/asm/msr-index.h index d406894cd9a2..5573c75f8e4c 100644 --- a/arch/x86/include/asm/msr-index.h +++ b/arch/x86/include/asm/msr-index.h @@ -426,6 +426,8 @@ #define MSR_IA32_TSC_ADJUST 0x0000003b #define MSR_IA32_BNDCFGS 0x00000d90 +#define MSR_IA32_BNDCFGS_RSVD 0x00000ffc + #define MSR_IA32_XSS 0x00000da0 #define FEATURE_CONTROL_LOCKED (1<<0) diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h index a059aac9e937..476ea27f490b 100644 --- a/arch/x86/include/asm/uaccess.h +++ b/arch/x86/include/asm/uaccess.h @@ -565,7 +565,6 @@ copy_from_user_nmi(void *to, const void __user *from, unsigned long n); extern __must_check long strncpy_from_user(char *dst, const char __user *src, long count); -extern __must_check long strlen_user(const char __user *str); extern __must_check long strnlen_user(const char __user *str, long n); unsigned long __must_check clear_user(void __user *mem, unsigned long len); diff --git a/arch/x86/include/asm/xen/hypercall.h b/arch/x86/include/asm/xen/hypercall.h index f6d20f6cca12..11071fcd630e 100644 --- a/arch/x86/include/asm/xen/hypercall.h +++ b/arch/x86/include/asm/xen/hypercall.h @@ -43,6 +43,7 @@ #include <asm/page.h> #include <asm/pgtable.h> +#include <asm/smap.h> #include <xen/interface/xen.h> #include <xen/interface/sched.h> @@ -50,6 +51,8 @@ #include <xen/interface/platform.h> #include <xen/interface/xen-mca.h> +struct xen_dm_op_buf; + /* * The hypercall asms have to meet several constraints: * - Work on 32- and 64-bit. @@ -214,10 +217,12 @@ privcmd_call(unsigned call, __HYPERCALL_DECLS; __HYPERCALL_5ARG(a1, a2, a3, a4, a5); + stac(); asm volatile("call *%[call]" : __HYPERCALL_5PARAM : [call] "a" (&hypercall_page[call]) : __HYPERCALL_CLOBBER5); + clac(); return (long)__res; } @@ -474,9 +479,13 @@ HYPERVISOR_xenpmu_op(unsigned int op, void *arg) static inline int HYPERVISOR_dm_op( - domid_t dom, unsigned int nr_bufs, void *bufs) + domid_t dom, unsigned int nr_bufs, struct xen_dm_op_buf *bufs) { - return _hypercall3(int, dm_op, dom, nr_bufs, bufs); + int ret; + stac(); + ret = _hypercall3(int, dm_op, dom, nr_bufs, bufs); + clac(); + return ret; } static inline void diff --git a/arch/x86/kernel/amd_gart_64.c b/arch/x86/kernel/amd_gart_64.c index 815dd63f49d0..cc0e8bc0ea3f 100644 --- a/arch/x86/kernel/amd_gart_64.c +++ b/arch/x86/kernel/amd_gart_64.c @@ -704,6 +704,7 @@ static const struct dma_map_ops gart_dma_ops = { .alloc = gart_alloc_coherent, .free = gart_free_coherent, .mapping_error = gart_mapping_error, + .dma_supported = x86_dma_supported, }; static void gart_iommu_shutdown(void) diff --git a/arch/x86/kernel/pci-calgary_64.c b/arch/x86/kernel/pci-calgary_64.c index fda7867046d0..5286a4a92cf7 100644 --- a/arch/x86/kernel/pci-calgary_64.c +++ b/arch/x86/kernel/pci-calgary_64.c @@ -50,6 +50,8 @@ #include <asm/x86_init.h> #include <asm/iommu_table.h> +#define CALGARY_MAPPING_ERROR 0 + #ifdef CONFIG_CALGARY_IOMMU_ENABLED_BY_DEFAULT int use_calgary __read_mostly = 1; #else @@ -252,7 +254,7 @@ static unsigned long iommu_range_alloc(struct device *dev, if (panic_on_overflow) panic("Calgary: fix the allocator.\n"); else - return DMA_ERROR_CODE; + return CALGARY_MAPPING_ERROR; } } @@ -272,10 +274,10 @@ static dma_addr_t iommu_alloc(struct device *dev, struct iommu_table *tbl, entry = iommu_range_alloc(dev, tbl, npages); - if (unlikely(entry == DMA_ERROR_CODE)) { + if (unlikely(entry == CALGARY_MAPPING_ERROR)) { pr_warn("failed to allocate %u pages in iommu %p\n", npages, tbl); - return DMA_ERROR_CODE; + return CALGARY_MAPPING_ERROR; } /* set the return dma address */ @@ -295,7 +297,7 @@ static void iommu_free(struct iommu_table *tbl, dma_addr_t dma_addr, unsigned long flags; /* were we called with bad_dma_address? */ - badend = DMA_ERROR_CODE + (EMERGENCY_PAGES * PAGE_SIZE); + badend = CALGARY_MAPPING_ERROR + (EMERGENCY_PAGES * PAGE_SIZE); if (unlikely(dma_addr < badend)) { WARN(1, KERN_ERR "Calgary: driver tried unmapping bad DMA " "address 0x%Lx\n", dma_addr); @@ -380,7 +382,7 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg, npages = iommu_num_pages(vaddr, s->length, PAGE_SIZE); entry = iommu_range_alloc(dev, tbl, npages); - if (entry == DMA_ERROR_CODE) { + if (entry == CALGARY_MAPPING_ERROR) { /* makes sure unmap knows to stop */ s->dma_length = 0; goto error; @@ -398,7 +400,7 @@ static int calgary_map_sg(struct device *dev, struct scatterlist *sg, error: calgary_unmap_sg(dev, sg, nelems, dir, 0); for_each_sg(sg, s, nelems, i) { - sg->dma_address = DMA_ERROR_CODE; + sg->dma_address = CALGARY_MAPPING_ERROR; sg->dma_length = 0; } return 0; @@ -453,7 +455,7 @@ static void* calgary_alloc_coherent(struct device *dev, size_t size, /* set up tces to cover the allocated range */ mapping = iommu_alloc(dev, tbl, ret, npages, DMA_BIDIRECTIONAL); - if (mapping == DMA_ERROR_CODE) + if (mapping == CALGARY_MAPPING_ERROR) goto free; *dma_handle = mapping; return ret; @@ -478,6 +480,11 @@ static void calgary_free_coherent(struct device *dev, size_t size, free_pages((unsigned long)vaddr, get_order(size)); } +static int calgary_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + return dma_addr == CALGARY_MAPPING_ERROR; +} + static const struct dma_map_ops calgary_dma_ops = { .alloc = calgary_alloc_coherent, .free = calgary_free_coherent, @@ -485,6 +492,8 @@ static const struct dma_map_ops calgary_dma_ops = { .unmap_sg = calgary_unmap_sg, .map_page = calgary_map_page, .unmap_page = calgary_unmap_page, + .mapping_error = calgary_mapping_error, + .dma_supported = x86_dma_supported, }; static inline void __iomem * busno_to_bbar(unsigned char num) @@ -732,7 +741,7 @@ static void __init calgary_reserve_regions(struct pci_dev *dev) struct iommu_table *tbl = pci_iommu(dev->bus); /* reserve EMERGENCY_PAGES from bad_dma_address and up */ - iommu_range_reserve(tbl, DMA_ERROR_CODE, EMERGENCY_PAGES); + iommu_range_reserve(tbl, CALGARY_MAPPING_ERROR, EMERGENCY_PAGES); /* avoid the BIOS/VGA first 640KB-1MB region */ /* for CalIOC2 - avoid the entire first MB */ diff --git a/arch/x86/kernel/pci-dma.c b/arch/x86/kernel/pci-dma.c index 3a216ec869cd..5e16d3f29594 100644 --- a/arch/x86/kernel/pci-dma.c +++ b/arch/x86/kernel/pci-dma.c @@ -213,10 +213,8 @@ static __init int iommu_setup(char *p) } early_param("iommu", iommu_setup); -int dma_supported(struct device *dev, u64 mask) +int x86_dma_supported(struct device *dev, u64 mask) { - const struct dma_map_ops *ops = get_dma_ops(dev); - #ifdef CONFIG_PCI if (mask > 0xffffffff && forbid_dac > 0) { dev_info(dev, "PCI: Disallowing DAC for device\n"); @@ -224,9 +222,6 @@ int dma_supported(struct device *dev, u64 mask) } #endif - if (ops->dma_supported) - return ops->dma_supported(dev, mask); - /* Copied from i386. Doesn't make much sense, because it will only work for pci_alloc_coherent. The caller just has to use GFP_DMA in this case. */ @@ -252,7 +247,6 @@ int dma_supported(struct device *dev, u64 mask) return 1; } -EXPORT_SYMBOL(dma_supported); static int __init pci_iommu_init(void) { diff --git a/arch/x86/kernel/pci-nommu.c b/arch/x86/kernel/pci-nommu.c index a88952ef371c..a6d404087fe3 100644 --- a/arch/x86/kernel/pci-nommu.c +++ b/arch/x86/kernel/pci-nommu.c @@ -11,6 +11,8 @@ #include <asm/iommu.h> #include <asm/dma.h> +#define NOMMU_MAPPING_ERROR 0 + static int check_addr(char *name, struct device *hwdev, dma_addr_t bus, size_t size) { @@ -33,7 +35,7 @@ static dma_addr_t nommu_map_page(struct device *dev, struct page *page, dma_addr_t bus = page_to_phys(page) + offset; WARN_ON(size == 0); if (!check_addr("map_single", dev, bus, size)) - return DMA_ERROR_CODE; + return NOMMU_MAPPING_ERROR; flush_write_buffers(); return bus; } @@ -88,6 +90,11 @@ static void nommu_sync_sg_for_device(struct device *dev, flush_write_buffers(); } +static int nommu_mapping_error(struct device *dev, dma_addr_t dma_addr) +{ + return dma_addr == NOMMU_MAPPING_ERROR; +} + const struct dma_map_ops nommu_dma_ops = { .alloc = dma_generic_alloc_coherent, .free = dma_generic_free_coherent, @@ -96,4 +103,6 @@ const struct dma_map_ops nommu_dma_ops = { .sync_single_for_device = nommu_sync_single_for_device, .sync_sg_for_device = nommu_sync_sg_for_device, .is_phys = 1, + .mapping_error = nommu_mapping_error, + .dma_supported = x86_dma_supported, }; diff --git a/arch/x86/kvm/cpuid.h b/arch/x86/kvm/cpuid.h index a6fd40aade7c..da6728383052 100644 --- a/arch/x86/kvm/cpuid.h +++ b/arch/x86/kvm/cpuid.h @@ -144,6 +144,14 @@ static inline bool guest_cpuid_has_rtm(struct kvm_vcpu *vcpu) return best && (best->ebx & bit(X86_FEATURE_RTM)); } +static inline bool guest_cpuid_has_mpx(struct kvm_vcpu *vcpu) +{ + struct kvm_cpuid_entry2 *best; + + best = kvm_find_cpuid_entry(vcpu, 7, 0); + return best && (best->ebx & bit(X86_FEATURE_MPX)); +} + static inline bool guest_cpuid_has_rdtscp(struct kvm_vcpu *vcpu) { struct kvm_cpuid_entry2 *best; diff --git a/arch/x86/kvm/emulate.c b/arch/x86/kvm/emulate.c index 80890dee66ce..fb0055953fbc 100644 --- a/arch/x86/kvm/emulate.c +++ b/arch/x86/kvm/emulate.c @@ -900,7 +900,7 @@ static __always_inline int do_insn_fetch_bytes(struct x86_emulate_ctxt *ctxt, if (rc != X86EMUL_CONTINUE) \ goto done; \ ctxt->_eip += sizeof(_type); \ - _x = *(_type __aligned(1) *) ctxt->fetch.ptr; \ + memcpy(&_x, ctxt->fetch.ptr, sizeof(_type)); \ ctxt->fetch.ptr += sizeof(_type); \ _x; \ }) @@ -3942,6 +3942,25 @@ static int check_fxsr(struct x86_emulate_ctxt *ctxt) } /* + * Hardware doesn't save and restore XMM 0-7 without CR4.OSFXSR, but does save + * and restore MXCSR. + */ +static size_t __fxstate_size(int nregs) +{ + return offsetof(struct fxregs_state, xmm_space[0]) + nregs * 16; +} + +static inline size_t fxstate_size(struct x86_emulate_ctxt *ctxt) +{ + bool cr4_osfxsr; + if (ctxt->mode == X86EMUL_MODE_PROT64) + return __fxstate_size(16); + + cr4_osfxsr = ctxt->ops->get_cr(ctxt, 4) & X86_CR4_OSFXSR; + return __fxstate_size(cr4_osfxsr ? 8 : 0); +} + +/* * FXSAVE and FXRSTOR have 4 different formats depending on execution mode, * 1) 16 bit mode * 2) 32 bit mode @@ -3962,7 +3981,6 @@ static int check_fxsr(struct x86_emulate_ctxt *ctxt) static int em_fxsave(struct x86_emulate_ctxt *ctxt) { struct fxregs_state fx_state; - size_t size; int rc; rc = check_fxsr(ctxt); @@ -3978,68 +3996,42 @@ static int em_fxsave(struct x86_emulate_ctxt *ctxt) if (rc != X86EMUL_CONTINUE) return rc; - if (ctxt->ops->get_cr(ctxt, 4) & X86_CR4_OSFXSR) - size = offsetof(struct fxregs_state, xmm_space[8 * 16/4]); - else - size = offsetof(struct fxregs_state, xmm_space[0]); - - return segmented_write_std(ctxt, ctxt->memop.addr.mem, &fx_state, size); -} - -static int fxrstor_fixup(struct x86_emulate_ctxt *ctxt, - struct fxregs_state *new) -{ - int rc = X86EMUL_CONTINUE; - struct fxregs_state old; - - rc = asm_safe("fxsave %[fx]", , [fx] "+m"(old)); - if (rc != X86EMUL_CONTINUE) - return rc; - - /* - * 64 bit host will restore XMM 8-15, which is not correct on non-64 - * bit guests. Load the current values in order to preserve 64 bit - * XMMs after fxrstor. - */ -#ifdef CONFIG_X86_64 - /* XXX: accessing XMM 8-15 very awkwardly */ - memcpy(&new->xmm_space[8 * 16/4], &old.xmm_space[8 * 16/4], 8 * 16); -#endif - - /* - * Hardware doesn't save and restore XMM 0-7 without CR4.OSFXSR, but - * does save and restore MXCSR. - */ - if (!(ctxt->ops->get_cr(ctxt, 4) & X86_CR4_OSFXSR)) - memcpy(new->xmm_space, old.xmm_space, 8 * 16); - - return rc; + return segmented_write_std(ctxt, ctxt->memop.addr.mem, &fx_state, + fxstate_size(ctxt)); } static int em_fxrstor(struct x86_emulate_ctxt *ctxt) { struct fxregs_state fx_state; int rc; + size_t size; rc = check_fxsr(ctxt); if (rc != X86EMUL_CONTINUE) return rc; - rc = segmented_read_std(ctxt, ctxt->memop.addr.mem, &fx_state, 512); - if (rc != X86EMUL_CONTINUE) - return rc; + ctxt->ops->get_fpu(ctxt); - if (fx_state.mxcsr >> 16) - return emulate_gp(ctxt, 0); + size = fxstate_size(ctxt); + if (size < __fxstate_size(16)) { + rc = asm_safe("fxsave %[fx]", , [fx] "+m"(fx_state)); + if (rc != X86EMUL_CONTINUE) + goto out; + } - ctxt->ops->get_fpu(ctxt); + rc = segmented_read_std(ctxt, ctxt->memop.addr.mem, &fx_state, size); + if (rc != X86EMUL_CONTINUE) + goto out; - if (ctxt->mode < X86EMUL_MODE_PROT64) - rc = fxrstor_fixup(ctxt, &fx_state); + if (fx_state.mxcsr >> 16) { + rc = emulate_gp(ctxt, 0); + goto out; + } if (rc == X86EMUL_CONTINUE) rc = asm_safe("fxrstor %[fx]", : [fx] "m"(fx_state)); +out: ctxt->ops->put_fpu(ctxt); return rc; diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c index d24c8742d9b0..2819d4c123eb 100644 --- a/arch/x86/kvm/lapic.c +++ b/arch/x86/kvm/lapic.c @@ -1495,6 +1495,7 @@ EXPORT_SYMBOL_GPL(kvm_lapic_hv_timer_in_use); static void cancel_hv_timer(struct kvm_lapic *apic) { + WARN_ON(!apic->lapic_timer.hv_timer_in_use); preempt_disable(); kvm_x86_ops->cancel_hv_timer(apic->vcpu); apic->lapic_timer.hv_timer_in_use = false; @@ -1503,25 +1504,56 @@ static void cancel_hv_timer(struct kvm_lapic *apic) static bool start_hv_timer(struct kvm_lapic *apic) { - u64 tscdeadline = apic->lapic_timer.tscdeadline; + struct kvm_timer *ktimer = &apic->lapic_timer; + int r; - if ((atomic_read(&apic->lapic_timer.pending) && - !apic_lvtt_period(apic)) || - kvm_x86_ops->set_hv_timer(apic->vcpu, tscdeadline)) { - if (apic->lapic_timer.hv_timer_in_use) - cancel_hv_timer(apic); - } else { - apic->lapic_timer.hv_timer_in_use = true; - hrtimer_cancel(&apic->lapic_timer.timer); + if (!kvm_x86_ops->set_hv_timer) + return false; + + if (!apic_lvtt_period(apic) && atomic_read(&ktimer->pending)) + return false; - /* In case the sw timer triggered in the window */ - if (atomic_read(&apic->lapic_timer.pending) && - !apic_lvtt_period(apic)) - cancel_hv_timer(apic); + r = kvm_x86_ops->set_hv_timer(apic->vcpu, ktimer->tscdeadline); + if (r < 0) + return false; + + ktimer->hv_timer_in_use = true; + hrtimer_cancel(&ktimer->timer); + + /* + * Also recheck ktimer->pending, in case the sw timer triggered in + * the window. For periodic timer, leave the hv timer running for + * simplicity, and the deadline will be recomputed on the next vmexit. + */ + if (!apic_lvtt_period(apic) && (r || atomic_read(&ktimer->pending))) { + if (r) + apic_timer_expired(apic); + return false; } - trace_kvm_hv_timer_state(apic->vcpu->vcpu_id, - apic->lapic_timer.hv_timer_in_use); - return apic->lapic_timer.hv_timer_in_use; + + trace_kvm_hv_timer_state(apic->vcpu->vcpu_id, true); + return true; +} + +static void start_sw_timer(struct kvm_lapic *apic) +{ + struct kvm_timer *ktimer = &apic->lapic_timer; + if (apic->lapic_timer.hv_timer_in_use) + cancel_hv_timer(apic); + if (!apic_lvtt_period(apic) && atomic_read(&ktimer->pending)) + return; + + if (apic_lvtt_period(apic) || apic_lvtt_oneshot(apic)) + start_sw_period(apic); + else if (apic_lvtt_tscdeadline(apic)) + start_sw_tscdeadline(apic); + trace_kvm_hv_timer_state(apic->vcpu->vcpu_id, false); +} + +static void restart_apic_timer(struct kvm_lapic *apic) +{ + if (!start_hv_timer(apic)) + start_sw_timer(apic); } void kvm_lapic_expired_hv_timer(struct kvm_vcpu *vcpu) @@ -1535,19 +1567,14 @@ void kvm_lapic_expired_hv_timer(struct kvm_vcpu *vcpu) if (apic_lvtt_period(apic) && apic->lapic_timer.period) { advance_periodic_target_expiration(apic); - if (!start_hv_timer(apic)) - start_sw_period(apic); + restart_apic_timer(apic); } } EXPORT_SYMBOL_GPL(kvm_lapic_expired_hv_timer); void kvm_lapic_switch_to_hv_timer(struct kvm_vcpu *vcpu) { - struct kvm_lapic *apic = vcpu->arch.apic; - - WARN_ON(apic->lapic_timer.hv_timer_in_use); - - start_hv_timer(apic); + restart_apic_timer(vcpu->arch.apic); } EXPORT_SYMBOL_GPL(kvm_lapic_switch_to_hv_timer); @@ -1556,33 +1583,28 @@ void kvm_lapic_switch_to_sw_timer(struct kvm_vcpu *vcpu) struct kvm_lapic *apic = vcpu->arch.apic; /* Possibly the TSC deadline timer is not enabled yet */ - if (!apic->lapic_timer.hv_timer_in_use) - return; - - cancel_hv_timer(apic); + if (apic->lapic_timer.hv_timer_in_use) + start_sw_timer(apic); +} +EXPORT_SYMBOL_GPL(kvm_lapic_switch_to_sw_timer); - if (atomic_read(&apic->lapic_timer.pending)) - return; +void kvm_lapic_restart_hv_timer(struct kvm_vcpu *vcpu) +{ + struct kvm_lapic *apic = vcpu->arch.apic; - if (apic_lvtt_period(apic) || apic_lvtt_oneshot(apic)) - start_sw_period(apic); - else if (apic_lvtt_tscdeadline(apic)) - start_sw_tscdeadline(apic); + WARN_ON(!apic->lapic_timer.hv_timer_in_use); + restart_apic_timer(apic); } -EXPORT_SYMBOL_GPL(kvm_lapic_switch_to_sw_timer); static void start_apic_timer(struct kvm_lapic *apic) { atomic_set(&apic->lapic_timer.pending, 0); - if (apic_lvtt_period(apic) || apic_lvtt_oneshot(apic)) { - if (set_target_expiration(apic) && - !(kvm_x86_ops->set_hv_timer && start_hv_timer(apic))) - start_sw_period(apic); - } else if (apic_lvtt_tscdeadline(apic)) { - if (!(kvm_x86_ops->set_hv_timer && start_hv_timer(apic))) - start_sw_tscdeadline(apic); - } + if ((apic_lvtt_period(apic) || apic_lvtt_oneshot(apic)) + && !set_target_expiration(apic)) + return; + + restart_apic_timer(apic); } static void apic_manage_nmi_watchdog(struct kvm_lapic *apic, u32 lvt0_val) @@ -1813,16 +1835,6 @@ void kvm_free_lapic(struct kvm_vcpu *vcpu) * LAPIC interface *---------------------------------------------------------------------- */ -u64 kvm_get_lapic_target_expiration_tsc(struct kvm_vcpu *vcpu) -{ - struct kvm_lapic *apic = vcpu->arch.apic; - - if (!lapic_in_kernel(vcpu)) - return 0; - - return apic->lapic_timer.tscdeadline; -} - u64 kvm_get_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu) { struct kvm_lapic *apic = vcpu->arch.apic; diff --git a/arch/x86/kvm/lapic.h b/arch/x86/kvm/lapic.h index bcbe811f3b97..29caa2c3dff9 100644 --- a/arch/x86/kvm/lapic.h +++ b/arch/x86/kvm/lapic.h @@ -87,7 +87,6 @@ int kvm_apic_get_state(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s); int kvm_apic_set_state(struct kvm_vcpu *vcpu, struct kvm_lapic_state *s); int kvm_lapic_find_highest_irr(struct kvm_vcpu *vcpu); -u64 kvm_get_lapic_target_expiration_tsc(struct kvm_vcpu *vcpu); u64 kvm_get_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu); void kvm_set_lapic_tscdeadline_msr(struct kvm_vcpu *vcpu, u64 data); @@ -216,4 +215,5 @@ void kvm_lapic_switch_to_sw_timer(struct kvm_vcpu *vcpu); void kvm_lapic_switch_to_hv_timer(struct kvm_vcpu *vcpu); void kvm_lapic_expired_hv_timer(struct kvm_vcpu *vcpu); bool kvm_lapic_hv_timer_in_use(struct kvm_vcpu *vcpu); +void kvm_lapic_restart_hv_timer(struct kvm_vcpu *vcpu); #endif diff --git a/arch/x86/kvm/mmu.c b/arch/x86/kvm/mmu.c index cb8225969255..aafd399cf8c6 100644 --- a/arch/x86/kvm/mmu.c +++ b/arch/x86/kvm/mmu.c @@ -183,13 +183,13 @@ static u64 __read_mostly shadow_user_mask; static u64 __read_mostly shadow_accessed_mask; static u64 __read_mostly shadow_dirty_mask; static u64 __read_mostly shadow_mmio_mask; +static u64 __read_mostly shadow_mmio_value; static u64 __read_mostly shadow_present_mask; /* - * The mask/value to distinguish a PTE that has been marked not-present for - * access tracking purposes. - * The mask would be either 0 if access tracking is disabled, or - * SPTE_SPECIAL_MASK|VMX_EPT_RWX_MASK if access tracking is enabled. + * SPTEs used by MMUs without A/D bits are marked with shadow_acc_track_value. + * Non-present SPTEs with shadow_acc_track_value set are in place for access + * tracking. */ static u64 __read_mostly shadow_acc_track_mask; static const u64 shadow_acc_track_value = SPTE_SPECIAL_MASK; @@ -207,16 +207,40 @@ static const u64 shadow_acc_track_saved_bits_shift = PT64_SECOND_AVAIL_BITS_SHIF static void mmu_spte_set(u64 *sptep, u64 spte); static void mmu_free_roots(struct kvm_vcpu *vcpu); -void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask) +void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask, u64 mmio_value) { + BUG_ON((mmio_mask & mmio_value) != mmio_value); + shadow_mmio_value = mmio_value | SPTE_SPECIAL_MASK; shadow_mmio_mask = mmio_mask | SPTE_SPECIAL_MASK; } EXPORT_SYMBOL_GPL(kvm_mmu_set_mmio_spte_mask); +static inline bool sp_ad_disabled(struct kvm_mmu_page *sp) +{ + return sp->role.ad_disabled; +} + +static inline bool spte_ad_enabled(u64 spte) +{ + MMU_WARN_ON((spte & shadow_mmio_mask) == shadow_mmio_value); + return !(spte & shadow_acc_track_value); +} + +static inline u64 spte_shadow_accessed_mask(u64 spte) +{ + MMU_WARN_ON((spte & shadow_mmio_mask) == shadow_mmio_value); + return spte_ad_enabled(spte) ? shadow_accessed_mask : 0; +} + +static inline u64 spte_shadow_dirty_mask(u64 spte) +{ + MMU_WARN_ON((spte & shadow_mmio_mask) == shadow_mmio_value); + return spte_ad_enabled(spte) ? shadow_dirty_mask : 0; +} + static inline bool is_access_track_spte(u64 spte) { - /* Always false if shadow_acc_track_mask is zero. */ - return (spte & shadow_acc_track_mask) == shadow_acc_track_value; + return !spte_ad_enabled(spte) && (spte & shadow_acc_track_mask) == 0; } /* @@ -270,7 +294,7 @@ static void mark_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, u64 gfn, u64 mask = generation_mmio_spte_mask(gen); access &= ACC_WRITE_MASK | ACC_USER_MASK; - mask |= shadow_mmio_mask | access | gfn << PAGE_SHIFT; + mask |= shadow_mmio_value | access | gfn << PAGE_SHIFT; trace_mark_mmio_spte(sptep, gfn, access, gen); mmu_spte_set(sptep, mask); @@ -278,7 +302,7 @@ static void mark_mmio_spte(struct kvm_vcpu *vcpu, u64 *sptep, u64 gfn, static bool is_mmio_spte(u64 spte) { - return (spte & shadow_mmio_mask) == shadow_mmio_mask; + return (spte & shadow_mmio_mask) == shadow_mmio_value; } static gfn_t get_mmio_spte_gfn(u64 spte) @@ -315,12 +339,20 @@ static bool check_mmio_spte(struct kvm_vcpu *vcpu, u64 spte) return likely(kvm_gen == spte_gen); } +/* + * Sets the shadow PTE masks used by the MMU. + * + * Assumptions: + * - Setting either @accessed_mask or @dirty_mask requires setting both + * - At least one of @accessed_mask or @acc_track_mask must be set + */ void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, u64 dirty_mask, u64 nx_mask, u64 x_mask, u64 p_mask, u64 acc_track_mask) { - if (acc_track_mask != 0) - acc_track_mask |= SPTE_SPECIAL_MASK; + BUG_ON(!dirty_mask != !accessed_mask); + BUG_ON(!accessed_mask && !acc_track_mask); + BUG_ON(acc_track_mask & shadow_acc_track_value); shadow_user_mask = user_mask; shadow_accessed_mask = accessed_mask; @@ -329,7 +361,6 @@ void kvm_mmu_set_mask_ptes(u64 user_mask, u64 accessed_mask, shadow_x_mask = x_mask; shadow_present_mask = p_mask; shadow_acc_track_mask = acc_track_mask; - WARN_ON(shadow_accessed_mask != 0 && shadow_acc_track_mask != 0); } EXPORT_SYMBOL_GPL(kvm_mmu_set_mask_ptes); @@ -549,7 +580,7 @@ static bool spte_has_volatile_bits(u64 spte) is_access_track_spte(spte)) return true; - if (shadow_accessed_mask) { + if (spte_ad_enabled(spte)) { if ((spte & shadow_accessed_mask) == 0 || (is_writable_pte(spte) && (spte & shadow_dirty_mask) == 0)) return true; @@ -560,14 +591,17 @@ static bool spte_has_volatile_bits(u64 spte) static bool is_accessed_spte(u64 spte) { - return shadow_accessed_mask ? spte & shadow_accessed_mask - : !is_access_track_spte(spte); + u64 accessed_mask = spte_shadow_accessed_mask(spte); + + return accessed_mask ? spte & accessed_mask + : !is_access_track_spte(spte); } static bool is_dirty_spte(u64 spte) { - return shadow_dirty_mask ? spte & shadow_dirty_mask - : spte & PT_WRITABLE_MASK; + u64 dirty_mask = spte_shadow_dirty_mask(spte); + + return dirty_mask ? spte & dirty_mask : spte & PT_WRITABLE_MASK; } /* Rules for using mmu_spte_set: @@ -707,10 +741,10 @@ static u64 mmu_spte_get_lockless(u64 *sptep) static u64 mark_spte_for_access_track(u64 spte) { - if (shadow_accessed_mask != 0) + if (spte_ad_enabled(spte)) return spte & ~shadow_accessed_mask; - if (shadow_acc_track_mask == 0 || is_access_track_spte(spte)) + if (is_access_track_spte(spte)) return spte; /* @@ -729,7 +763,6 @@ static u64 mark_spte_for_access_track(u64 spte) spte |= (spte & shadow_acc_track_saved_bits_mask) << shadow_acc_track_saved_bits_shift; spte &= ~shadow_acc_track_mask; - spte |= shadow_acc_track_value; return spte; } @@ -741,6 +774,7 @@ static u64 restore_acc_track_spte(u64 spte) u64 saved_bits = (spte >> shadow_acc_track_saved_bits_shift) & shadow_acc_track_saved_bits_mask; + WARN_ON_ONCE(spte_ad_enabled(spte)); WARN_ON_ONCE(!is_access_track_spte(spte)); new_spte &= ~shadow_acc_track_mask; @@ -759,7 +793,7 @@ static bool mmu_spte_age(u64 *sptep) if (!is_accessed_spte(spte)) return false; - if (shadow_accessed_mask) { + if (spte_ad_enabled(spte)) { clear_bit((ffs(shadow_accessed_mask) - 1), (unsigned long *)sptep); } else { @@ -1390,6 +1424,22 @@ static bool spte_clear_dirty(u64 *sptep) return mmu_spte_update(sptep, spte); } +static bool wrprot_ad_disabled_spte(u64 *sptep) +{ + bool was_writable = test_and_clear_bit(PT_WRITABLE_SHIFT, + (unsigned long *)sptep); + if (was_writable) + kvm_set_pfn_dirty(spte_to_pfn(*sptep)); + + return was_writable; +} + +/* + * Gets the GFN ready for another round of dirty logging by clearing the + * - D bit on ad-enabled SPTEs, and + * - W bit on ad-disabled SPTEs. + * Returns true iff any D or W bits were cleared. + */ static bool __rmap_clear_dirty(struct kvm *kvm, struct kvm_rmap_head *rmap_head) { u64 *sptep; @@ -1397,7 +1447,10 @@ static bool __rmap_clear_dirty(struct kvm *kvm, struct kvm_rmap_head *rmap_head) bool flush = false; for_each_rmap_spte(rmap_head, &iter, sptep) - flush |= spte_clear_dirty(sptep); + if (spte_ad_enabled(*sptep)) + flush |= spte_clear_dirty(sptep); + else + flush |= wrprot_ad_disabled_spte(sptep); return flush; } @@ -1420,7 +1473,8 @@ static bool __rmap_set_dirty(struct kvm *kvm, struct kvm_rmap_head *rmap_head) bool flush = false; for_each_rmap_spte(rmap_head, &iter, sptep) - flush |= spte_set_dirty(sptep); + if (spte_ad_enabled(*sptep)) + flush |= spte_set_dirty(sptep); return flush; } @@ -1452,7 +1506,8 @@ static void kvm_mmu_write_protect_pt_masked(struct kvm *kvm, } /** - * kvm_mmu_clear_dirty_pt_masked - clear MMU D-bit for PT level pages + * kvm_mmu_clear_dirty_pt_masked - clear MMU D-bit for PT level pages, or write + * protect the page if the D-bit isn't supported. * @kvm: kvm instance * @slot: slot to clear D-bit * @gfn_offset: start of the BITS_PER_LONG pages we care about @@ -1766,18 +1821,9 @@ static int kvm_test_age_rmapp(struct kvm *kvm, struct kvm_rmap_head *rmap_head, u64 *sptep; struct rmap_iterator iter; - /* - * If there's no access bit in the secondary pte set by the hardware and - * fast access tracking is also not enabled, it's up to gup-fast/gup to - * set the access bit in the primary pte or in the page structure. - */ - if (!shadow_accessed_mask && !shadow_acc_track_mask) - goto out; - for_each_rmap_spte(rmap_head, &iter, sptep) if (is_accessed_spte(*sptep)) return 1; -out: return 0; } @@ -1798,18 +1844,6 @@ static void rmap_recycle(struct kvm_vcpu *vcpu, u64 *spte, gfn_t gfn) int kvm_age_hva(struct kvm *kvm, unsigned long start, unsigned long end) { - /* - * In case of absence of EPT Access and Dirty Bits supports, - * emulate the accessed bit for EPT, by checking if this page has - * an EPT mapping, and clearing it if it does. On the next access, - * a new EPT mapping will be established. - * This has some overhead, but not as much as the cost of swapping - * out actively used pages or breaking up actively used hugepages. - */ - if (!shadow_accessed_mask && !shadow_acc_track_mask) - return kvm_handle_hva_range(kvm, start, end, 0, - kvm_unmap_rmapp); - return kvm_handle_hva_range(kvm, start, end, 0, kvm_age_rmapp); } @@ -2398,7 +2432,12 @@ static void link_shadow_page(struct kvm_vcpu *vcpu, u64 *sptep, BUILD_BUG_ON(VMX_EPT_WRITABLE_MASK != PT_WRITABLE_MASK); spte = __pa(sp->spt) | shadow_present_mask | PT_WRITABLE_MASK | - shadow_user_mask | shadow_x_mask | shadow_accessed_mask; + shadow_user_mask | shadow_x_mask; + + if (sp_ad_disabled(sp)) + spte |= shadow_acc_track_value; + else + spte |= shadow_accessed_mask; mmu_spte_set(sptep, spte); @@ -2666,10 +2705,15 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, { u64 spte = 0; int ret = 0; + struct kvm_mmu_page *sp; if (set_mmio_spte(vcpu, sptep, gfn, pfn, pte_access)) return 0; + sp = page_header(__pa(sptep)); + if (sp_ad_disabled(sp)) + spte |= shadow_acc_track_value; + /* * For the EPT case, shadow_present_mask is 0 if hardware * supports exec-only page table entries. In that case, @@ -2678,7 +2722,7 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, */ spte |= shadow_present_mask; if (!speculative) - spte |= shadow_accessed_mask; + spte |= spte_shadow_accessed_mask(spte); if (pte_access & ACC_EXEC_MASK) spte |= shadow_x_mask; @@ -2735,7 +2779,7 @@ static int set_spte(struct kvm_vcpu *vcpu, u64 *sptep, if (pte_access & ACC_WRITE_MASK) { kvm_vcpu_mark_page_dirty(vcpu, gfn); - spte |= shadow_dirty_mask; + spte |= spte_shadow_dirty_mask(spte); } if (speculative) @@ -2877,16 +2921,16 @@ static void direct_pte_prefetch(struct kvm_vcpu *vcpu, u64 *sptep) { struct kvm_mmu_page *sp; + sp = page_header(__pa(sptep)); + /* - * Since it's no accessed bit on EPT, it's no way to - * distinguish between actually accessed translations - * and prefetched, so disable pte prefetch if EPT is - * enabled. + * Without accessed bits, there's no way to distinguish between + * actually accessed translations and prefetched, so disable pte + * prefetch if accessed bits aren't available. */ - if (!shadow_accessed_mask) + if (sp_ad_disabled(sp)) return; - sp = page_header(__pa(sptep)); if (sp->role.level > PT_PAGE_TABLE_LEVEL) return; @@ -4290,6 +4334,7 @@ static void init_kvm_tdp_mmu(struct kvm_vcpu *vcpu) context->base_role.word = 0; context->base_role.smm = is_smm(vcpu); + context->base_role.ad_disabled = (shadow_accessed_mask == 0); context->page_fault = tdp_page_fault; context->sync_page = nonpaging_sync_page; context->invlpg = nonpaging_invlpg; @@ -4377,6 +4422,7 @@ void kvm_init_shadow_ept_mmu(struct kvm_vcpu *vcpu, bool execonly, context->root_level = context->shadow_root_level; context->root_hpa = INVALID_PAGE; context->direct_map = false; + context->base_role.ad_disabled = !accessed_dirty; update_permission_bitmask(vcpu, context, true); update_pkru_bitmask(vcpu, context, true); @@ -4636,6 +4682,7 @@ static void kvm_mmu_pte_write(struct kvm_vcpu *vcpu, gpa_t gpa, mask.smep_andnot_wp = 1; mask.smap_andnot_wp = 1; mask.smm = 1; + mask.ad_disabled = 1; /* * If we don't have indirect shadow pages, it means no page is diff --git a/arch/x86/kvm/mmu.h b/arch/x86/kvm/mmu.h index 330bf3a811fb..a276834950c1 100644 --- a/arch/x86/kvm/mmu.h +++ b/arch/x86/kvm/mmu.h @@ -51,7 +51,7 @@ static inline u64 rsvd_bits(int s, int e) return ((1ULL << (e - s + 1)) - 1) << s; } -void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask); +void kvm_mmu_set_mmio_spte_mask(u64 mmio_mask, u64 mmio_value); void reset_shadow_zero_bits_mask(struct kvm_vcpu *vcpu, struct kvm_mmu *context); diff --git a/arch/x86/kvm/mmutrace.h b/arch/x86/kvm/mmutrace.h index 5a24b846a1cb..8b97a6cba8d1 100644 --- a/arch/x86/kvm/mmutrace.h +++ b/arch/x86/kvm/mmutrace.h @@ -30,8 +30,9 @@ \ role.word = __entry->role; \ \ - trace_seq_printf(p, "sp gen %lx gfn %llx %u%s q%u%s %s%s" \ - " %snxe root %u %s%c", __entry->mmu_valid_gen, \ + trace_seq_printf(p, "sp gen %lx gfn %llx l%u%s q%u%s %s%s" \ + " %snxe %sad root %u %s%c", \ + __entry->mmu_valid_gen, \ __entry->gfn, role.level, \ role.cr4_pae ? " pae" : "", \ role.quadrant, \ @@ -39,6 +40,7 @@ access_str[role.access], \ role.invalid ? " invalid" : "", \ role.nxe ? "" : "!", \ + role.ad_disabled ? "!" : "", \ __entry->root_count, \ __entry->unsync ? "unsync" : "sync", 0); \ saved_ptr; \ diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c index 33460fcdeef9..905ea6052517 100644 --- a/arch/x86/kvm/svm.c +++ b/arch/x86/kvm/svm.c @@ -190,6 +190,7 @@ struct vcpu_svm { struct nested_state nested; bool nmi_singlestep; + u64 nmi_singlestep_guest_rflags; unsigned int3_injected; unsigned long int3_rip; @@ -964,6 +965,18 @@ static void svm_disable_lbrv(struct vcpu_svm *svm) set_msr_interception(msrpm, MSR_IA32_LASTINTTOIP, 0, 0); } +static void disable_nmi_singlestep(struct vcpu_svm *svm) +{ + svm->nmi_singlestep = false; + if (!(svm->vcpu.guest_debug & KVM_GUESTDBG_SINGLESTEP)) { + /* Clear our flags if they were not set by the guest */ + if (!(svm->nmi_singlestep_guest_rflags & X86_EFLAGS_TF)) + svm->vmcb->save.rflags &= ~X86_EFLAGS_TF; + if (!(svm->nmi_singlestep_guest_rflags & X86_EFLAGS_RF)) + svm->vmcb->save.rflags &= ~X86_EFLAGS_RF; + } +} + /* Note: * This hash table is used to map VM_ID to a struct kvm_arch, * when handling AMD IOMMU GALOG notification to schedule in @@ -1713,11 +1726,24 @@ static void svm_vcpu_unblocking(struct kvm_vcpu *vcpu) static unsigned long svm_get_rflags(struct kvm_vcpu *vcpu) { - return to_svm(vcpu)->vmcb->save.rflags; + struct vcpu_svm *svm = to_svm(vcpu); + unsigned long rflags = svm->vmcb->save.rflags; + + if (svm->nmi_singlestep) { + /* Hide our flags if they were not set by the guest */ + if (!(svm->nmi_singlestep_guest_rflags & X86_EFLAGS_TF)) + rflags &= ~X86_EFLAGS_TF; + if (!(svm->nmi_singlestep_guest_rflags & X86_EFLAGS_RF)) + rflags &= ~X86_EFLAGS_RF; + } + return rflags; } static void svm_set_rflags(struct kvm_vcpu *vcpu, unsigned long rflags) { + if (to_svm(vcpu)->nmi_singlestep) + rflags |= (X86_EFLAGS_TF | X86_EFLAGS_RF); + /* * Any change of EFLAGS.VM is accompanied by a reload of SS * (caused by either a task switch or an inter-privilege IRET), @@ -2112,10 +2138,7 @@ static int db_interception(struct vcpu_svm *svm) } if (svm->nmi_singlestep) { - svm->nmi_singlestep = false; - if (!(svm->vcpu.guest_debug & KVM_GUESTDBG_SINGLESTEP)) - svm->vmcb->save.rflags &= - ~(X86_EFLAGS_TF | X86_EFLAGS_RF); + disable_nmi_singlestep(svm); } if (svm->vcpu.guest_debug & @@ -2370,8 +2393,8 @@ static void nested_svm_uninit_mmu_context(struct kvm_vcpu *vcpu) static int nested_svm_check_permissions(struct vcpu_svm *svm) { - if (!(svm->vcpu.arch.efer & EFER_SVME) - || !is_paging(&svm->vcpu)) { + if (!(svm->vcpu.arch.efer & EFER_SVME) || + !is_paging(&svm->vcpu)) { kvm_queue_exception(&svm->vcpu, UD_VECTOR); return 1; } @@ -2381,7 +2404,7 @@ static int nested_svm_check_permissions(struct vcpu_svm *svm) return 1; } - return 0; + return 0; } static int nested_svm_check_exception(struct vcpu_svm *svm, unsigned nr, @@ -2534,6 +2557,31 @@ static int nested_svm_exit_handled_msr(struct vcpu_svm *svm) return (value & mask) ? NESTED_EXIT_DONE : NESTED_EXIT_HOST; } +/* DB exceptions for our internal use must not cause vmexit */ +static int nested_svm_intercept_db(struct vcpu_svm *svm) +{ + unsigned long dr6; + + /* if we're not singlestepping, it's not ours */ + if (!svm->nmi_singlestep) + return NESTED_EXIT_DONE; + + /* if it's not a singlestep exception, it's not ours */ + if (kvm_get_dr(&svm->vcpu, 6, &dr6)) + return NESTED_EXIT_DONE; + if (!(dr6 & DR6_BS)) + return NESTED_EXIT_DONE; + + /* if the guest is singlestepping, it should get the vmexit */ + if (svm->nmi_singlestep_guest_rflags & X86_EFLAGS_TF) { + disable_nmi_singlestep(svm); + return NESTED_EXIT_DONE; + } + + /* it's ours, the nested hypervisor must not see this one */ + return NESTED_EXIT_HOST; +} + static int nested_svm_exit_special(struct vcpu_svm *svm) { u32 exit_code = svm->vmcb->control.exit_code; @@ -2589,8 +2637,12 @@ static int nested_svm_intercept(struct vcpu_svm *svm) } case SVM_EXIT_EXCP_BASE ... SVM_EXIT_EXCP_BASE + 0x1f: { u32 excp_bits = 1 << (exit_code - SVM_EXIT_EXCP_BASE); - if (svm->nested.intercept_exceptions & excp_bits) - vmexit = NESTED_EXIT_DONE; + if (svm->nested.intercept_exceptions & excp_bits) { + if (exit_code == SVM_EXIT_EXCP_BASE + DB_VECTOR) + vmexit = nested_svm_intercept_db(svm); + else + vmexit = NESTED_EXIT_DONE; + } /* async page fault always cause vmexit */ else if ((exit_code == SVM_EXIT_EXCP_BASE + PF_VECTOR) && svm->apf_reason != 0) @@ -4627,10 +4679,17 @@ static void enable_nmi_window(struct kvm_vcpu *vcpu) == HF_NMI_MASK) return; /* IRET will cause a vm exit */ + if ((svm->vcpu.arch.hflags & HF_GIF_MASK) == 0) + return; /* STGI will cause a vm exit */ + + if (svm->nested.exit_required) + return; /* we're not going to run the guest yet */ + /* * Something prevents NMI from been injected. Single step over possible * problem (IRET or exception injection or interrupt shadow) */ + svm->nmi_singlestep_guest_rflags = svm_get_rflags(vcpu); svm->nmi_singlestep = true; svm->vmcb->save.rflags |= (X86_EFLAGS_TF | X86_EFLAGS_RF); } @@ -4771,6 +4830,22 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu) if (unlikely(svm->nested.exit_required)) return; + /* + * Disable singlestep if we're injecting an interrupt/exception. + * We don't want our modified rflags to be pushed on the stack where + * we might not be able to easily reset them if we disabled NMI + * singlestep later. + */ + if (svm->nmi_singlestep && svm->vmcb->control.event_inj) { + /* + * Event injection happens before external interrupts cause a + * vmexit and interrupts are disabled here, so smp_send_reschedule + * is enough to force an immediate vmexit. + */ + disable_nmi_singlestep(svm); + smp_send_reschedule(vcpu->cpu); + } + pre_svm_run(svm); sync_lapic_to_cr8(vcpu); diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 6dcc4873e435..f76efad248ab 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -913,8 +913,9 @@ static void nested_release_page_clean(struct page *page) kvm_release_page_clean(page); } +static bool nested_ept_ad_enabled(struct kvm_vcpu *vcpu); static unsigned long nested_ept_get_cr3(struct kvm_vcpu *vcpu); -static u64 construct_eptp(unsigned long root_hpa); +static u64 construct_eptp(struct kvm_vcpu *vcpu, unsigned long root_hpa); static bool vmx_xsaves_supported(void); static int vmx_set_tss_addr(struct kvm *kvm, unsigned int addr); static void vmx_set_segment(struct kvm_vcpu *vcpu, @@ -2772,7 +2773,7 @@ static void nested_vmx_setup_ctls_msrs(struct vcpu_vmx *vmx) if (enable_ept_ad_bits) { vmx->nested.nested_vmx_secondary_ctls_high |= SECONDARY_EXEC_ENABLE_PML; - vmx->nested.nested_vmx_ept_caps |= VMX_EPT_AD_BIT; + vmx->nested.nested_vmx_ept_caps |= VMX_EPT_AD_BIT; } } else vmx->nested.nested_vmx_ept_caps = 0; @@ -3198,7 +3199,8 @@ static int vmx_get_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) msr_info->data = vmcs_readl(GUEST_SYSENTER_ESP); break; case MSR_IA32_BNDCFGS: - if (!kvm_mpx_supported()) + if (!kvm_mpx_supported() || + (!msr_info->host_initiated && !guest_cpuid_has_mpx(vcpu))) return 1; msr_info->data = vmcs_read64(GUEST_BNDCFGS); break; @@ -3280,7 +3282,11 @@ static int vmx_set_msr(struct kvm_vcpu *vcpu, struct msr_data *msr_info) vmcs_writel(GUEST_SYSENTER_ESP, data); break; case MSR_IA32_BNDCFGS: - if (!kvm_mpx_supported()) + if (!kvm_mpx_supported() || + (!msr_info->host_initiated && !guest_cpuid_has_mpx(vcpu))) + return 1; + if (is_noncanonical_address(data & PAGE_MASK) || + (data & MSR_IA32_BNDCFGS_RSVD)) return 1; vmcs_write64(GUEST_BNDCFGS, data); break; @@ -4013,7 +4019,7 @@ static inline void __vmx_flush_tlb(struct kvm_vcpu *vcpu, int vpid) if (enable_ept) { if (!VALID_PAGE(vcpu->arch.mmu.root_hpa)) return; - ept_sync_context(construct_eptp(vcpu->arch.mmu.root_hpa)); + ept_sync_context(construct_eptp(vcpu, vcpu->arch.mmu.root_hpa)); } else { vpid_sync_context(vpid); } @@ -4188,14 +4194,15 @@ static void vmx_set_cr0(struct kvm_vcpu *vcpu, unsigned long cr0) vmx->emulation_required = emulation_required(vcpu); } -static u64 construct_eptp(unsigned long root_hpa) +static u64 construct_eptp(struct kvm_vcpu *vcpu, unsigned long root_hpa) { u64 eptp; /* TODO write the value reading from MSR */ eptp = VMX_EPT_DEFAULT_MT | VMX_EPT_DEFAULT_GAW << VMX_EPT_GAW_EPTP_SHIFT; - if (enable_ept_ad_bits) + if (enable_ept_ad_bits && + (!is_guest_mode(vcpu) || nested_ept_ad_enabled(vcpu))) eptp |= VMX_EPT_AD_ENABLE_BIT; eptp |= (root_hpa & PAGE_MASK); @@ -4209,7 +4216,7 @@ static void vmx_set_cr3(struct kvm_vcpu *vcpu, unsigned long cr3) guest_cr3 = cr3; if (enable_ept) { - eptp = construct_eptp(cr3); + eptp = construct_eptp(vcpu, cr3); vmcs_write64(EPT_POINTER, eptp); if (is_paging(vcpu) || is_guest_mode(vcpu)) guest_cr3 = kvm_read_cr3(vcpu); @@ -5170,7 +5177,8 @@ static void ept_set_mmio_spte_mask(void) * EPT Misconfigurations can be generated if the value of bits 2:0 * of an EPT paging-structure entry is 110b (write/execute). */ - kvm_mmu_set_mmio_spte_mask(VMX_EPT_MISCONFIG_WX_VALUE); + kvm_mmu_set_mmio_spte_mask(VMX_EPT_RWX_MASK, + VMX_EPT_MISCONFIG_WX_VALUE); } #define VMX_XSS_EXIT_BITMAP 0 @@ -6220,17 +6228,6 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu) exit_qualification = vmcs_readl(EXIT_QUALIFICATION); - if (is_guest_mode(vcpu) - && !(exit_qualification & EPT_VIOLATION_GVA_TRANSLATED)) { - /* - * Fix up exit_qualification according to whether guest - * page table accesses are reads or writes. - */ - u64 eptp = nested_ept_get_cr3(vcpu); - if (!(eptp & VMX_EPT_AD_ENABLE_BIT)) - exit_qualification &= ~EPT_VIOLATION_ACC_WRITE; - } - /* * EPT violation happened while executing iret from NMI, * "blocked by NMI" bit has to be set before next VM entry. @@ -6453,7 +6450,7 @@ void vmx_enable_tdp(void) enable_ept_ad_bits ? VMX_EPT_DIRTY_BIT : 0ull, 0ull, VMX_EPT_EXECUTABLE_MASK, cpu_has_vmx_ept_execute_only() ? 0ull : VMX_EPT_READABLE_MASK, - enable_ept_ad_bits ? 0ull : VMX_EPT_RWX_MASK); + VMX_EPT_RWX_MASK); ept_set_mmio_spte_mask(); kvm_enable_tdp(); @@ -6557,7 +6554,6 @@ static __init int hardware_setup(void) vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_CS, false); vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_ESP, false); vmx_disable_intercept_for_msr(MSR_IA32_SYSENTER_EIP, false); - vmx_disable_intercept_for_msr(MSR_IA32_BNDCFGS, true); memcpy(vmx_msr_bitmap_legacy_x2apic_apicv, vmx_msr_bitmap_legacy, PAGE_SIZE); @@ -7661,7 +7657,10 @@ static int handle_invvpid(struct kvm_vcpu *vcpu) unsigned long type, types; gva_t gva; struct x86_exception e; - int vpid; + struct { + u64 vpid; + u64 gla; + } operand; if (!(vmx->nested.nested_vmx_secondary_ctls_high & SECONDARY_EXEC_ENABLE_VPID) || @@ -7691,17 +7690,28 @@ static int handle_invvpid(struct kvm_vcpu *vcpu) if (get_vmx_mem_address(vcpu, vmcs_readl(EXIT_QUALIFICATION), vmx_instruction_info, false, &gva)) return 1; - if (kvm_read_guest_virt(&vcpu->arch.emulate_ctxt, gva, &vpid, - sizeof(u32), &e)) { + if (kvm_read_guest_virt(&vcpu->arch.emulate_ctxt, gva, &operand, + sizeof(operand), &e)) { kvm_inject_page_fault(vcpu, &e); return 1; } + if (operand.vpid >> 16) { + nested_vmx_failValid(vcpu, + VMXERR_INVALID_OPERAND_TO_INVEPT_INVVPID); + return kvm_skip_emulated_instruction(vcpu); + } switch (type) { case VMX_VPID_EXTENT_INDIVIDUAL_ADDR: + if (is_noncanonical_address(operand.gla)) { + nested_vmx_failValid(vcpu, + VMXERR_INVALID_OPERAND_TO_INVEPT_INVVPID); + return kvm_skip_emulated_instruction(vcpu); + } + /* fall through */ case VMX_VPID_EXTENT_SINGLE_CONTEXT: case VMX_VPID_EXTENT_SINGLE_NON_GLOBAL: - if (!vpid) { + if (!operand.vpid) { nested_vmx_failValid(vcpu, VMXERR_INVALID_OPERAND_TO_INVEPT_INVVPID); return kvm_skip_emulated_instruction(vcpu); @@ -9394,6 +9404,11 @@ static void nested_ept_inject_page_fault(struct kvm_vcpu *vcpu, vmcs12->guest_physical_address = fault->address; } +static bool nested_ept_ad_enabled(struct kvm_vcpu *vcpu) +{ + return nested_ept_get_cr3(vcpu) & VMX_EPT_AD_ENABLE_BIT; +} + /* Callbacks for nested_ept_init_mmu_context: */ static unsigned long nested_ept_get_cr3(struct kvm_vcpu *vcpu) @@ -9404,18 +9419,18 @@ static unsigned long nested_ept_get_cr3(struct kvm_vcpu *vcpu) static int nested_ept_init_mmu_context(struct kvm_vcpu *vcpu) { - u64 eptp; + bool wants_ad; WARN_ON(mmu_is_nested(vcpu)); - eptp = nested_ept_get_cr3(vcpu); - if ((eptp & VMX_EPT_AD_ENABLE_BIT) && !enable_ept_ad_bits) + wants_ad = nested_ept_ad_enabled(vcpu); + if (wants_ad && !enable_ept_ad_bits) return 1; kvm_mmu_unload(vcpu); kvm_init_shadow_ept_mmu(vcpu, to_vmx(vcpu)->nested.nested_vmx_ept_caps & VMX_EPT_EXECUTE_ONLY_BIT, - eptp & VMX_EPT_AD_ENABLE_BIT); + wants_ad); vcpu->arch.mmu.set_cr3 = vmx_set_cr3; vcpu->arch.mmu.get_cr3 = nested_ept_get_cr3; vcpu->arch.mmu.inject_page_fault = nested_ept_inject_page_fault; @@ -10728,8 +10743,7 @@ static void sync_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) vmcs12->guest_pdptr3 = vmcs_read64(GUEST_PDPTR3); } - if (nested_cpu_has_ept(vmcs12)) - vmcs12->guest_linear_address = vmcs_readl(GUEST_LINEAR_ADDRESS); + vmcs12->guest_linear_address = vmcs_readl(GUEST_LINEAR_ADDRESS); if (nested_cpu_has_vid(vmcs12)) vmcs12->guest_intr_status = vmcs_read16(GUEST_INTR_STATUS); @@ -10754,8 +10768,6 @@ static void sync_vmcs12(struct kvm_vcpu *vcpu, struct vmcs12 *vmcs12) vmcs12->guest_sysenter_eip = vmcs_readl(GUEST_SYSENTER_EIP); if (kvm_mpx_supported()) vmcs12->guest_bndcfgs = vmcs_read64(GUEST_BNDCFGS); - if (nested_cpu_has_xsaves(vmcs12)) - vmcs12->xss_exit_bitmap = vmcs_read64(XSS_EXIT_BITMAP); } /* @@ -11152,7 +11164,8 @@ static int vmx_set_hv_timer(struct kvm_vcpu *vcpu, u64 guest_deadline_tsc) vmx->hv_deadline_tsc = tscl + delta_tsc; vmcs_set_bits(PIN_BASED_VM_EXEC_CONTROL, PIN_BASED_VMX_PREEMPTION_TIMER); - return 0; + + return delta_tsc == 0; } static void vmx_cancel_hv_timer(struct kvm_vcpu *vcpu) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 0e846f0cb83b..6c7266f7766d 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -2841,10 +2841,10 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cpu) kvm_vcpu_write_tsc_offset(vcpu, offset); vcpu->arch.tsc_catchup = 1; } - if (kvm_lapic_hv_timer_in_use(vcpu) && - kvm_x86_ops->set_hv_timer(vcpu, - kvm_get_lapic_target_expiration_tsc(vcpu))) - kvm_lapic_switch_to_sw_timer(vcpu); + + if (kvm_lapic_hv_timer_in_use(vcpu)) + kvm_lapic_restart_hv_timer(vcpu); + /* * On a host with synchronized TSC, there is no need to update * kvmclock on vcpu->cpu migration @@ -6011,7 +6011,7 @@ static void kvm_set_mmio_spte_mask(void) mask &= ~1ull; #endif - kvm_mmu_set_mmio_spte_mask(mask); + kvm_mmu_set_mmio_spte_mask(mask, mask); } #ifdef CONFIG_X86_64 @@ -6733,7 +6733,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) bool req_immediate_exit = false; - if (vcpu->requests) { + if (kvm_request_pending(vcpu)) { if (kvm_check_request(KVM_REQ_MMU_RELOAD, vcpu)) kvm_mmu_unload(vcpu); if (kvm_check_request(KVM_REQ_MIGRATE_TIMER, vcpu)) @@ -6897,7 +6897,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) kvm_x86_ops->sync_pir_to_irr(vcpu); } - if (vcpu->mode == EXITING_GUEST_MODE || vcpu->requests + if (vcpu->mode == EXITING_GUEST_MODE || kvm_request_pending(vcpu) || need_resched() || signal_pending(current)) { vcpu->mode = OUTSIDE_GUEST_MODE; smp_wmb(); diff --git a/arch/x86/pci/sta2x11-fixup.c b/arch/x86/pci/sta2x11-fixup.c index ec008e800b45..53d600217973 100644 --- a/arch/x86/pci/sta2x11-fixup.c +++ b/arch/x86/pci/sta2x11-fixup.c @@ -26,6 +26,7 @@ #include <linux/pci_ids.h> #include <linux/export.h> #include <linux/list.h> +#include <asm/iommu.h> #define STA2X11_SWIOTLB_SIZE (4*1024*1024) extern int swiotlb_late_init_with_default_size(size_t default_size); @@ -191,7 +192,7 @@ static const struct dma_map_ops sta2x11_dma_ops = { .sync_sg_for_cpu = swiotlb_sync_sg_for_cpu, .sync_sg_for_device = swiotlb_sync_sg_for_device, .mapping_error = swiotlb_dma_mapping_error, - .dma_supported = NULL, /* FIXME: we should use this instead! */ + .dma_supported = x86_dma_supported, }; /* At setup time, we use our own ops if the device is a ConneXt one */ diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c index a5ffcbb20cc0..0e7ef69e8531 100644 --- a/arch/x86/xen/enlighten.c +++ b/arch/x86/xen/enlighten.c @@ -106,15 +106,83 @@ int xen_cpuhp_setup(int (*cpu_up_prepare_cb)(unsigned int), return rc >= 0 ? 0 : rc; } -static void clamp_max_cpus(void) +static int xen_vcpu_setup_restore(int cpu) { -#ifdef CONFIG_SMP - if (setup_max_cpus > MAX_VIRT_CPUS) - setup_max_cpus = MAX_VIRT_CPUS; -#endif + int rc = 0; + + /* Any per_cpu(xen_vcpu) is stale, so reset it */ + xen_vcpu_info_reset(cpu); + + /* + * For PVH and PVHVM, setup online VCPUs only. The rest will + * be handled by hotplug. + */ + if (xen_pv_domain() || + (xen_hvm_domain() && cpu_online(cpu))) { + rc = xen_vcpu_setup(cpu); + } + + return rc; +} + +/* + * On restore, set the vcpu placement up again. + * If it fails, then we're in a bad state, since + * we can't back out from using it... + */ +void xen_vcpu_restore(void) +{ + int cpu, rc; + + for_each_possible_cpu(cpu) { + bool other_cpu = (cpu != smp_processor_id()); + bool is_up; + + if (xen_vcpu_nr(cpu) == XEN_VCPU_ID_INVALID) + continue; + + /* Only Xen 4.5 and higher support this. */ + is_up = HYPERVISOR_vcpu_op(VCPUOP_is_up, + xen_vcpu_nr(cpu), NULL) > 0; + + if (other_cpu && is_up && + HYPERVISOR_vcpu_op(VCPUOP_down, xen_vcpu_nr(cpu), NULL)) + BUG(); + + if (xen_pv_domain() || xen_feature(XENFEAT_hvm_safe_pvclock)) + xen_setup_runstate_info(cpu); + + rc = xen_vcpu_setup_restore(cpu); + if (rc) + pr_emerg_once("vcpu restore failed for cpu=%d err=%d. " + "System will hang.\n", cpu, rc); + /* + * In case xen_vcpu_setup_restore() fails, do not bring up the + * VCPU. This helps us avoid the resulting OOPS when the VCPU + * accesses pvclock_vcpu_time via xen_vcpu (which is NULL.) + * Note that this does not improve the situation much -- now the + * VM hangs instead of OOPSing -- with the VCPUs that did not + * fail, spinning in stop_machine(), waiting for the failed + * VCPUs to come up. + */ + if (other_cpu && is_up && (rc == 0) && + HYPERVISOR_vcpu_op(VCPUOP_up, xen_vcpu_nr(cpu), NULL)) + BUG(); + } } -void xen_vcpu_setup(int cpu) +void xen_vcpu_info_reset(int cpu) +{ + if (xen_vcpu_nr(cpu) < MAX_VIRT_CPUS) { + per_cpu(xen_vcpu, cpu) = + &HYPERVISOR_shared_info->vcpu_info[xen_vcpu_nr(cpu)]; + } else { + /* Set to NULL so that if somebody accesses it we get an OOPS */ + per_cpu(xen_vcpu, cpu) = NULL; + } +} + +int xen_vcpu_setup(int cpu) { struct vcpu_register_vcpu_info info; int err; @@ -123,11 +191,11 @@ void xen_vcpu_setup(int cpu) BUG_ON(HYPERVISOR_shared_info == &xen_dummy_shared_info); /* - * This path is called twice on PVHVM - first during bootup via - * smp_init -> xen_hvm_cpu_notify, and then if the VCPU is being - * hotplugged: cpu_up -> xen_hvm_cpu_notify. - * As we can only do the VCPUOP_register_vcpu_info once lets - * not over-write its result. + * This path is called on PVHVM at bootup (xen_hvm_smp_prepare_boot_cpu) + * and at restore (xen_vcpu_restore). Also called for hotplugged + * VCPUs (cpu_init -> xen_hvm_cpu_prepare_hvm). + * However, the hypercall can only be done once (see below) so if a VCPU + * is offlined and comes back online then let's not redo the hypercall. * * For PV it is called during restore (xen_vcpu_restore) and bootup * (xen_setup_vcpu_info_placement). The hotplug mechanism does not @@ -135,42 +203,44 @@ void xen_vcpu_setup(int cpu) */ if (xen_hvm_domain()) { if (per_cpu(xen_vcpu, cpu) == &per_cpu(xen_vcpu_info, cpu)) - return; + return 0; } - if (xen_vcpu_nr(cpu) < MAX_VIRT_CPUS) - per_cpu(xen_vcpu, cpu) = - &HYPERVISOR_shared_info->vcpu_info[xen_vcpu_nr(cpu)]; - if (!xen_have_vcpu_info_placement) { - if (cpu >= MAX_VIRT_CPUS) - clamp_max_cpus(); - return; + if (xen_have_vcpu_info_placement) { + vcpup = &per_cpu(xen_vcpu_info, cpu); + info.mfn = arbitrary_virt_to_mfn(vcpup); + info.offset = offset_in_page(vcpup); + + /* + * Check to see if the hypervisor will put the vcpu_info + * structure where we want it, which allows direct access via + * a percpu-variable. + * N.B. This hypercall can _only_ be called once per CPU. + * Subsequent calls will error out with -EINVAL. This is due to + * the fact that hypervisor has no unregister variant and this + * hypercall does not allow to over-write info.mfn and + * info.offset. + */ + err = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_info, + xen_vcpu_nr(cpu), &info); + + if (err) { + pr_warn_once("register_vcpu_info failed: cpu=%d err=%d\n", + cpu, err); + xen_have_vcpu_info_placement = 0; + } else { + /* + * This cpu is using the registered vcpu info, even if + * later ones fail to. + */ + per_cpu(xen_vcpu, cpu) = vcpup; + } } - vcpup = &per_cpu(xen_vcpu_info, cpu); - info.mfn = arbitrary_virt_to_mfn(vcpup); - info.offset = offset_in_page(vcpup); - - /* Check to see if the hypervisor will put the vcpu_info - structure where we want it, which allows direct access via - a percpu-variable. - N.B. This hypercall can _only_ be called once per CPU. Subsequent - calls will error out with -EINVAL. This is due to the fact that - hypervisor has no unregister variant and this hypercall does not - allow to over-write info.mfn and info.offset. - */ - err = HYPERVISOR_vcpu_op(VCPUOP_register_vcpu_info, xen_vcpu_nr(cpu), - &info); + if (!xen_have_vcpu_info_placement) + xen_vcpu_info_reset(cpu); - if (err) { - printk(KERN_DEBUG "register_vcpu_info failed: err=%d\n", err); - xen_have_vcpu_info_placement = 0; - clamp_max_cpus(); - } else { - /* This cpu is using the registered vcpu info, even if - later ones fail to. */ - per_cpu(xen_vcpu, cpu) = vcpup; - } + return ((per_cpu(xen_vcpu, cpu) == NULL) ? -ENODEV : 0); } void xen_reboot(int reason) diff --git a/arch/x86/xen/enlighten_hvm.c b/arch/x86/xen/enlighten_hvm.c index a6d014f47e52..87d791356ea9 100644 --- a/arch/x86/xen/enlighten_hvm.c +++ b/arch/x86/xen/enlighten_hvm.c @@ -1,5 +1,6 @@ #include <linux/cpu.h> #include <linux/kexec.h> +#include <linux/memblock.h> #include <xen/features.h> #include <xen/events.h> @@ -10,9 +11,11 @@ #include <asm/reboot.h> #include <asm/setup.h> #include <asm/hypervisor.h> +#include <asm/e820/api.h> #include <asm/xen/cpuid.h> #include <asm/xen/hypervisor.h> +#include <asm/xen/page.h> #include "xen-ops.h" #include "mmu.h" @@ -20,37 +23,34 @@ void __ref xen_hvm_init_shared_info(void) { - int cpu; struct xen_add_to_physmap xatp; - static struct shared_info *shared_info_page; + u64 pa; + + if (HYPERVISOR_shared_info == &xen_dummy_shared_info) { + /* + * Search for a free page starting at 4kB physical address. + * Low memory is preferred to avoid an EPT large page split up + * by the mapping. + * Starting below X86_RESERVE_LOW (usually 64kB) is fine as + * the BIOS used for HVM guests is well behaved and won't + * clobber memory other than the first 4kB. + */ + for (pa = PAGE_SIZE; + !e820__mapped_all(pa, pa + PAGE_SIZE, E820_TYPE_RAM) || + memblock_is_reserved(pa); + pa += PAGE_SIZE) + ; + + memblock_reserve(pa, PAGE_SIZE); + HYPERVISOR_shared_info = __va(pa); + } - if (!shared_info_page) - shared_info_page = (struct shared_info *) - extend_brk(PAGE_SIZE, PAGE_SIZE); xatp.domid = DOMID_SELF; xatp.idx = 0; xatp.space = XENMAPSPACE_shared_info; - xatp.gpfn = __pa(shared_info_page) >> PAGE_SHIFT; + xatp.gpfn = virt_to_pfn(HYPERVISOR_shared_info); if (HYPERVISOR_memory_op(XENMEM_add_to_physmap, &xatp)) BUG(); - - HYPERVISOR_shared_info = (struct shared_info *)shared_info_page; - - /* xen_vcpu is a pointer to the vcpu_info struct in the shared_info - * page, we use it in the event channel upcall and in some pvclock - * related functions. We don't need the vcpu_info placement - * optimizations because we don't use any pv_mmu or pv_irq op on - * HVM. - * When xen_hvm_init_shared_info is run at boot time only vcpu 0 is - * online but xen_hvm_init_shared_info is run at resume time too and - * in that case multiple vcpus might be online. */ - for_each_online_cpu(cpu) { - /* Leave it to be NULL. */ - if (xen_vcpu_nr(cpu) >= MAX_VIRT_CPUS) - continue; - per_cpu(xen_vcpu, cpu) = - &HYPERVISOR_shared_info->vcpu_info[xen_vcpu_nr(cpu)]; - } } static void __init init_hvm_pv_info(void) @@ -106,7 +106,7 @@ static void xen_hvm_crash_shutdown(struct pt_regs *regs) static int xen_cpu_up_prepare_hvm(unsigned int cpu) { - int rc; + int rc = 0; /* * This can happen if CPU was offlined earlier and @@ -121,7 +121,9 @@ static int xen_cpu_up_prepare_hvm(unsigned int cpu) per_cpu(xen_vcpu_id, cpu) = cpu_acpi_id(cpu); else per_cpu(xen_vcpu_id, cpu) = cpu; - xen_vcpu_setup(cpu); + rc = xen_vcpu_setup(cpu); + if (rc) + return rc; if (xen_have_vector_callback && xen_feature(XENFEAT_hvm_safe_pvclock)) xen_setup_timer(cpu); @@ -130,9 +132,8 @@ static int xen_cpu_up_prepare_hvm(unsigned int cpu) if (rc) { WARN(1, "xen_smp_intr_init() for CPU %d failed: %d\n", cpu, rc); - return rc; } - return 0; + return rc; } static int xen_cpu_dead_hvm(unsigned int cpu) @@ -154,6 +155,13 @@ static void __init xen_hvm_guest_init(void) xen_hvm_init_shared_info(); + /* + * xen_vcpu is a pointer to the vcpu_info struct in the shared_info + * page, we use it in the event channel upcall and in some pvclock + * related functions. + */ + xen_vcpu_info_reset(0); + xen_panic_handler_init(); if (xen_feature(XENFEAT_hvm_callback_vector)) diff --git a/arch/x86/xen/enlighten_pv.c b/arch/x86/xen/enlighten_pv.c index f33eef4ebd12..811e4ddb3f37 100644 --- a/arch/x86/xen/enlighten_pv.c +++ b/arch/x86/xen/enlighten_pv.c @@ -89,8 +89,6 @@ void *xen_initial_gdt; -RESERVE_BRK(shared_info_page_brk, PAGE_SIZE); - static int xen_cpu_up_prepare_pv(unsigned int cpu); static int xen_cpu_dead_pv(unsigned int cpu); @@ -107,35 +105,6 @@ struct tls_descs { */ static DEFINE_PER_CPU(struct tls_descs, shadow_tls_desc); -/* - * On restore, set the vcpu placement up again. - * If it fails, then we're in a bad state, since - * we can't back out from using it... - */ -void xen_vcpu_restore(void) -{ - int cpu; - - for_each_possible_cpu(cpu) { - bool other_cpu = (cpu != smp_processor_id()); - bool is_up = HYPERVISOR_vcpu_op(VCPUOP_is_up, xen_vcpu_nr(cpu), - NULL); - - if (other_cpu && is_up && - HYPERVISOR_vcpu_op(VCPUOP_down, xen_vcpu_nr(cpu), NULL)) - BUG(); - - xen_setup_runstate_info(cpu); - - if (xen_have_vcpu_info_placement) - xen_vcpu_setup(cpu); - - if (other_cpu && is_up && - HYPERVISOR_vcpu_op(VCPUOP_up, xen_vcpu_nr(cpu), NULL)) - BUG(); - } -} - static void __init xen_banner(void) { unsigned version = HYPERVISOR_xen_version(XENVER_version, NULL); @@ -960,30 +929,43 @@ void xen_setup_shared_info(void) HYPERVISOR_shared_info = (struct shared_info *)fix_to_virt(FIX_PARAVIRT_BOOTMAP); -#ifndef CONFIG_SMP - /* In UP this is as good a place as any to set up shared info */ - xen_setup_vcpu_info_placement(); -#endif - xen_setup_mfn_list_list(); - /* - * Now that shared info is set up we can start using routines that - * point to pvclock area. - */ - if (system_state == SYSTEM_BOOTING) + if (system_state == SYSTEM_BOOTING) { +#ifndef CONFIG_SMP + /* + * In UP this is as good a place as any to set up shared info. + * Limit this to boot only, at restore vcpu setup is done via + * xen_vcpu_restore(). + */ + xen_setup_vcpu_info_placement(); +#endif + /* + * Now that shared info is set up we can start using routines + * that point to pvclock area. + */ xen_init_time_ops(); + } } /* This is called once we have the cpu_possible_mask */ -void xen_setup_vcpu_info_placement(void) +void __ref xen_setup_vcpu_info_placement(void) { int cpu; for_each_possible_cpu(cpu) { /* Set up direct vCPU id mapping for PV guests. */ per_cpu(xen_vcpu_id, cpu) = cpu; - xen_vcpu_setup(cpu); + + /* + * xen_vcpu_setup(cpu) can fail -- in which case it + * falls back to the shared_info version for cpus + * where xen_vcpu_nr(cpu) < MAX_VIRT_CPUS. + * + * xen_cpu_up_prepare_pv() handles the rest by failing + * them in hotplug. + */ + (void) xen_vcpu_setup(cpu); } /* @@ -1332,9 +1314,17 @@ asmlinkage __visible void __init xen_start_kernel(void) */ acpi_numa = -1; #endif - /* Don't do the full vcpu_info placement stuff until we have a - possible map and a non-dummy shared_info. */ - per_cpu(xen_vcpu, 0) = &HYPERVISOR_shared_info->vcpu_info[0]; + /* Let's presume PV guests always boot on vCPU with id 0. */ + per_cpu(xen_vcpu_id, 0) = 0; + + /* + * Setup xen_vcpu early because start_kernel needs it for + * local_irq_disable(), irqs_disabled(). + * + * Don't do the full vcpu_info placement stuff until we have + * the cpu_possible_mask and a non-dummy shared_info. + */ + xen_vcpu_info_reset(0); WARN_ON(xen_cpuhp_setup(xen_cpu_up_prepare_pv, xen_cpu_dead_pv)); @@ -1431,9 +1421,7 @@ asmlinkage __visible void __init xen_start_kernel(void) #endif xen_raw_console_write("about to get started...\n"); - /* Let's presume PV guests always boot on vCPU with id 0. */ - per_cpu(xen_vcpu_id, 0) = 0; - + /* We need this for printk timestamps */ xen_setup_runstate_info(0); xen_efi_init(); @@ -1451,6 +1439,9 @@ static int xen_cpu_up_prepare_pv(unsigned int cpu) { int rc; + if (per_cpu(xen_vcpu, cpu) == NULL) + return -ENODEV; + xen_setup_timer(cpu); rc = xen_smp_intr_init(cpu); diff --git a/arch/x86/xen/pci-swiotlb-xen.c b/arch/x86/xen/pci-swiotlb-xen.c index 42b08f8fc2ca..37c6056a7bba 100644 --- a/arch/x86/xen/pci-swiotlb-xen.c +++ b/arch/x86/xen/pci-swiotlb-xen.c @@ -18,20 +18,6 @@ int xen_swiotlb __read_mostly; -static const struct dma_map_ops xen_swiotlb_dma_ops = { - .alloc = xen_swiotlb_alloc_coherent, - .free = xen_swiotlb_free_coherent, - .sync_single_for_cpu = xen_swiotlb_sync_single_for_cpu, - .sync_single_for_device = xen_swiotlb_sync_single_for_device, - .sync_sg_for_cpu = xen_swiotlb_sync_sg_for_cpu, - .sync_sg_for_device = xen_swiotlb_sync_sg_for_device, - .map_sg = xen_swiotlb_map_sg_attrs, - .unmap_sg = xen_swiotlb_unmap_sg_attrs, - .map_page = xen_swiotlb_map_page, - .unmap_page = xen_swiotlb_unmap_page, - .dma_supported = xen_swiotlb_dma_supported, -}; - /* * pci_xen_swiotlb_detect - set xen_swiotlb to 1 if necessary * diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c index a5bf7c451435..c81046323ebc 100644 --- a/arch/x86/xen/setup.c +++ b/arch/x86/xen/setup.c @@ -499,7 +499,7 @@ static unsigned long __init xen_foreach_remap_area(unsigned long nr_pages, void __init xen_remap_memory(void) { unsigned long buf = (unsigned long)&xen_remap_buf; - unsigned long mfn_save, mfn, pfn; + unsigned long mfn_save, pfn; unsigned long remapped = 0; unsigned int i; unsigned long pfn_s = ~0UL; @@ -515,8 +515,7 @@ void __init xen_remap_memory(void) pfn = xen_remap_buf.target_pfn; for (i = 0; i < xen_remap_buf.size; i++) { - mfn = xen_remap_buf.mfns[i]; - xen_update_mem_tables(pfn, mfn); + xen_update_mem_tables(pfn, xen_remap_buf.mfns[i]); remapped++; pfn++; } @@ -530,8 +529,6 @@ void __init xen_remap_memory(void) pfn_s = xen_remap_buf.target_pfn; len = xen_remap_buf.size; } - - mfn = xen_remap_mfn; xen_remap_mfn = xen_remap_buf.next_area_mfn; } diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c index 82ac611f2fc1..e7f02eb73727 100644 --- a/arch/x86/xen/smp.c +++ b/arch/x86/xen/smp.c @@ -1,4 +1,5 @@ #include <linux/smp.h> +#include <linux/cpu.h> #include <linux/slab.h> #include <linux/cpumask.h> #include <linux/percpu.h> @@ -114,6 +115,36 @@ int xen_smp_intr_init(unsigned int cpu) return rc; } +void __init xen_smp_cpus_done(unsigned int max_cpus) +{ + int cpu, rc, count = 0; + + if (xen_hvm_domain()) + native_smp_cpus_done(max_cpus); + + if (xen_have_vcpu_info_placement) + return; + + for_each_online_cpu(cpu) { + if (xen_vcpu_nr(cpu) < MAX_VIRT_CPUS) + continue; + + rc = cpu_down(cpu); + + if (rc == 0) { + /* + * Reset vcpu_info so this cpu cannot be onlined again. + */ + xen_vcpu_info_reset(cpu); + count++; + } else { + pr_warn("%s: failed to bring CPU %d down, error %d\n", + __func__, cpu, rc); + } + } + WARN(count, "%s: brought %d CPUs offline\n", __func__, count); +} + void xen_smp_send_reschedule(int cpu) { xen_send_IPI_one(cpu, XEN_RESCHEDULE_VECTOR); diff --git a/arch/x86/xen/smp.h b/arch/x86/xen/smp.h index 8ebb6acca64a..87d3c76cba37 100644 --- a/arch/x86/xen/smp.h +++ b/arch/x86/xen/smp.h @@ -14,6 +14,8 @@ extern void xen_smp_intr_free(unsigned int cpu); int xen_smp_intr_init_pv(unsigned int cpu); void xen_smp_intr_free_pv(unsigned int cpu); +void xen_smp_cpus_done(unsigned int max_cpus); + void xen_smp_send_reschedule(int cpu); void xen_smp_send_call_function_ipi(const struct cpumask *mask); void xen_smp_send_call_function_single_ipi(int cpu); diff --git a/arch/x86/xen/smp_hvm.c b/arch/x86/xen/smp_hvm.c index f18561bbf5c9..fd60abedf658 100644 --- a/arch/x86/xen/smp_hvm.c +++ b/arch/x86/xen/smp_hvm.c @@ -12,7 +12,8 @@ static void __init xen_hvm_smp_prepare_boot_cpu(void) native_smp_prepare_boot_cpu(); /* - * Setup vcpu_info for boot CPU. + * Setup vcpu_info for boot CPU. Secondary CPUs get their vcpu_info + * in xen_cpu_up_prepare_hvm(). */ xen_vcpu_setup(0); @@ -27,10 +28,20 @@ static void __init xen_hvm_smp_prepare_boot_cpu(void) static void __init xen_hvm_smp_prepare_cpus(unsigned int max_cpus) { + int cpu; + native_smp_prepare_cpus(max_cpus); WARN_ON(xen_smp_intr_init(0)); xen_init_lock_cpu(0); + + for_each_possible_cpu(cpu) { + if (cpu == 0) + continue; + + /* Set default vcpu_id to make sure that we don't use cpu-0's */ + per_cpu(xen_vcpu_id, cpu) = XEN_VCPU_ID_INVALID; + } } #ifdef CONFIG_HOTPLUG_CPU @@ -60,4 +71,5 @@ void __init xen_hvm_smp_init(void) smp_ops.send_call_func_ipi = xen_smp_send_call_function_ipi; smp_ops.send_call_func_single_ipi = xen_smp_send_call_function_single_ipi; smp_ops.smp_prepare_boot_cpu = xen_hvm_smp_prepare_boot_cpu; + smp_ops.smp_cpus_done = xen_smp_cpus_done; } diff --git a/arch/x86/xen/smp_pv.c b/arch/x86/xen/smp_pv.c index aae32535f4ec..1ea598e5f030 100644 --- a/arch/x86/xen/smp_pv.c +++ b/arch/x86/xen/smp_pv.c @@ -371,10 +371,6 @@ static int xen_pv_cpu_up(unsigned int cpu, struct task_struct *idle) return 0; } -static void xen_pv_smp_cpus_done(unsigned int max_cpus) -{ -} - #ifdef CONFIG_HOTPLUG_CPU static int xen_pv_cpu_disable(void) { @@ -469,7 +465,7 @@ static irqreturn_t xen_irq_work_interrupt(int irq, void *dev_id) static const struct smp_ops xen_smp_ops __initconst = { .smp_prepare_boot_cpu = xen_pv_smp_prepare_boot_cpu, .smp_prepare_cpus = xen_pv_smp_prepare_cpus, - .smp_cpus_done = xen_pv_smp_cpus_done, + .smp_cpus_done = xen_smp_cpus_done, .cpu_up = xen_pv_cpu_up, .cpu_die = xen_pv_cpu_die, diff --git a/arch/x86/xen/suspend_hvm.c b/arch/x86/xen/suspend_hvm.c index 01afcadde50a..484999416d8b 100644 --- a/arch/x86/xen/suspend_hvm.c +++ b/arch/x86/xen/suspend_hvm.c @@ -8,15 +8,10 @@ void xen_hvm_post_suspend(int suspend_cancelled) { - int cpu; - - if (!suspend_cancelled) + if (!suspend_cancelled) { xen_hvm_init_shared_info(); + xen_vcpu_restore(); + } xen_callback_vector(); xen_unplug_emulated_devices(); - if (xen_feature(XENFEAT_hvm_safe_pvclock)) { - for_each_online_cpu(cpu) { - xen_setup_runstate_info(cpu); - } - } } diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h index 9a440a42c618..0d5004477db6 100644 --- a/arch/x86/xen/xen-ops.h +++ b/arch/x86/xen/xen-ops.h @@ -78,7 +78,8 @@ bool xen_vcpu_stolen(int vcpu); extern int xen_have_vcpu_info_placement; -void xen_vcpu_setup(int cpu); +int xen_vcpu_setup(int cpu); +void xen_vcpu_info_reset(int cpu); void xen_setup_vcpu_info_placement(void); #ifdef CONFIG_SMP |