Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1760227AbYFZMo3 (ORCPT ); Thu, 26 Jun 2008 08:44:29 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754923AbYFZM3x (ORCPT ); Thu, 26 Jun 2008 08:29:53 -0400 Received: from il.qumranet.com ([212.179.150.194]:32345 "EHLO il.qumranet.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752140AbYFZM2j (ORCPT ); Thu, 26 Jun 2008 08:28:39 -0400 From: Avi Kivity To: kvm@vger.kernel.org Cc: linux-kernel@vger.kernel.org, Sheng Yang Subject: [PATCH 19/50] KVM: VMX: Enable NMI with in-kernel irqchip Date: Thu, 26 Jun 2008 15:28:01 +0300 Message-Id: <1214483312-9265-20-git-send-email-avi@qumranet.com> X-Mailer: git-send-email 1.5.6 In-Reply-To: <1214483312-9265-1-git-send-email-avi@qumranet.com> References: <1214483312-9265-1-git-send-email-avi@qumranet.com> Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 10983 Lines: 311 From: Sheng Yang Signed-off-by: Sheng Yang Signed-off-by: Avi Kivity --- arch/x86/kvm/vmx.c | 124 +++++++++++++++++++++++++++++++++++++------ arch/x86/kvm/vmx.h | 12 ++++- arch/x86/kvm/x86.c | 1 + include/asm-x86/kvm_host.h | 1 + 4 files changed, 119 insertions(+), 19 deletions(-) diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c index 96377bf..332ed24 100644 --- a/arch/x86/kvm/vmx.c +++ b/arch/x86/kvm/vmx.c @@ -264,6 +264,11 @@ static inline int cpu_has_vmx_vpid(void) SECONDARY_EXEC_ENABLE_VPID); } +static inline int cpu_has_virtual_nmis(void) +{ + return vmcs_config.pin_based_exec_ctrl & PIN_BASED_VIRTUAL_NMIS; +} + static int __find_msr_index(struct vcpu_vmx *vmx, u32 msr) { int i; @@ -1088,7 +1093,7 @@ static __init int setup_vmcs_config(struct vmcs_config *vmcs_conf) u32 _vmentry_control = 0; min = PIN_BASED_EXT_INTR_MASK | PIN_BASED_NMI_EXITING; - opt = 0; + opt = PIN_BASED_VIRTUAL_NMIS; if (adjust_vmx_controls(min, opt, MSR_IA32_VMX_PINBASED_CTLS, &_pin_based_exec_control) < 0) return -EIO; @@ -2130,6 +2135,13 @@ static void vmx_inject_irq(struct kvm_vcpu *vcpu, int irq) irq | INTR_TYPE_EXT_INTR | INTR_INFO_VALID_MASK); } +static void vmx_inject_nmi(struct kvm_vcpu *vcpu) +{ + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, + INTR_TYPE_NMI_INTR | INTR_INFO_VALID_MASK | NMI_VECTOR); + vcpu->arch.nmi_pending = 0; +} + static void kvm_do_inject_irq(struct kvm_vcpu *vcpu) { int word_index = __ffs(vcpu->arch.irq_summary); @@ -2653,6 +2665,19 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) return 1; } +static int handle_nmi_window(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) +{ + u32 cpu_based_vm_exec_control; + + /* clear pending NMI */ + cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); + cpu_based_vm_exec_control &= ~CPU_BASED_VIRTUAL_NMI_PENDING; + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); + ++vcpu->stat.nmi_window_exits; + + return 1; +} + /* * The exit handlers return 1 if the exit was handled fully and guest execution * may resume. Otherwise they set the kvm_run parameter to indicate what needs @@ -2663,6 +2688,7 @@ static int (*kvm_vmx_exit_handlers[])(struct kvm_vcpu *vcpu, [EXIT_REASON_EXCEPTION_NMI] = handle_exception, [EXIT_REASON_EXTERNAL_INTERRUPT] = handle_external_interrupt, [EXIT_REASON_TRIPLE_FAULT] = handle_triple_fault, + [EXIT_REASON_NMI_WINDOW] = handle_nmi_window, [EXIT_REASON_IO_INSTRUCTION] = handle_io, [EXIT_REASON_CR_ACCESS] = handle_cr, [EXIT_REASON_DR_ACCESS] = handle_dr, @@ -2750,17 +2776,52 @@ static void enable_irq_window(struct kvm_vcpu *vcpu) vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); } +static void enable_nmi_window(struct kvm_vcpu *vcpu) +{ + u32 cpu_based_vm_exec_control; + + if (!cpu_has_virtual_nmis()) + return; + + cpu_based_vm_exec_control = vmcs_read32(CPU_BASED_VM_EXEC_CONTROL); + cpu_based_vm_exec_control |= CPU_BASED_VIRTUAL_NMI_PENDING; + vmcs_write32(CPU_BASED_VM_EXEC_CONTROL, cpu_based_vm_exec_control); +} + +static int vmx_nmi_enabled(struct kvm_vcpu *vcpu) +{ + u32 guest_intr = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO); + return !(guest_intr & (GUEST_INTR_STATE_NMI | + GUEST_INTR_STATE_MOV_SS | + GUEST_INTR_STATE_STI)); +} + +static int vmx_irq_enabled(struct kvm_vcpu *vcpu) +{ + u32 guest_intr = vmcs_read32(GUEST_INTERRUPTIBILITY_INFO); + return (!(guest_intr & (GUEST_INTR_STATE_MOV_SS | + GUEST_INTR_STATE_STI)) && + (vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF)); +} + +static void enable_intr_window(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.nmi_pending) + enable_nmi_window(vcpu); + else if (kvm_cpu_has_interrupt(vcpu)) + enable_irq_window(vcpu); +} + static void vmx_intr_assist(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); - u32 idtv_info_field, intr_info_field; - int has_ext_irq, interrupt_window_open; + u32 idtv_info_field, intr_info_field, exit_intr_info_field; int vector; update_tpr_threshold(vcpu); - has_ext_irq = kvm_cpu_has_interrupt(vcpu); intr_info_field = vmcs_read32(VM_ENTRY_INTR_INFO_FIELD); + exit_intr_info_field = vmcs_read32(VM_EXIT_INTR_INFO); idtv_info_field = vmx->idt_vectoring_info; if (intr_info_field & INTR_INFO_VALID_MASK) { if (idtv_info_field & INTR_INFO_VALID_MASK) { @@ -2768,8 +2829,7 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) if (printk_ratelimit()) printk(KERN_ERR "Fault when IDT_Vectoring\n"); } - if (has_ext_irq) - enable_irq_window(vcpu); + enable_intr_window(vcpu); return; } if (unlikely(idtv_info_field & INTR_INFO_VALID_MASK)) { @@ -2779,30 +2839,56 @@ static void vmx_intr_assist(struct kvm_vcpu *vcpu) u8 vect = idtv_info_field & VECTORING_INFO_VECTOR_MASK; vmx_inject_irq(vcpu, vect); - if (unlikely(has_ext_irq)) - enable_irq_window(vcpu); + enable_intr_window(vcpu); return; } KVMTRACE_1D(REDELIVER_EVT, vcpu, idtv_info_field, handler); - vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field); + /* + * SDM 3: 25.7.1.2 + * Clear bit "block by NMI" before VM entry if a NMI delivery + * faulted. + */ + if ((idtv_info_field & VECTORING_INFO_TYPE_MASK) + == INTR_TYPE_NMI_INTR && cpu_has_virtual_nmis()) + vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, + vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & + ~GUEST_INTR_STATE_NMI); + + vmcs_write32(VM_ENTRY_INTR_INFO_FIELD, idtv_info_field + & ~INTR_INFO_RESVD_BITS_MASK); vmcs_write32(VM_ENTRY_INSTRUCTION_LEN, vmcs_read32(VM_EXIT_INSTRUCTION_LEN)); if (unlikely(idtv_info_field & INTR_INFO_DELIVER_CODE_MASK)) vmcs_write32(VM_ENTRY_EXCEPTION_ERROR_CODE, vmcs_read32(IDT_VECTORING_ERROR_CODE)); - if (unlikely(has_ext_irq)) - enable_irq_window(vcpu); + enable_intr_window(vcpu); return; } - if (!has_ext_irq) + if (cpu_has_virtual_nmis()) { + /* + * SDM 3: 25.7.1.2 + * Re-set bit "block by NMI" before VM entry if vmexit caused by + * a guest IRET fault. + */ + if ((exit_intr_info_field & INTR_INFO_UNBLOCK_NMI) && + (exit_intr_info_field & INTR_INFO_VECTOR_MASK) != 8) + vmcs_write32(GUEST_INTERRUPTIBILITY_INFO, + vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) | + GUEST_INTR_STATE_NMI); + else if (vcpu->arch.nmi_pending) { + if (vmx_nmi_enabled(vcpu)) + vmx_inject_nmi(vcpu); + enable_intr_window(vcpu); + return; + } + + } + if (!kvm_cpu_has_interrupt(vcpu)) return; - interrupt_window_open = - ((vmcs_readl(GUEST_RFLAGS) & X86_EFLAGS_IF) && - (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0); - if (interrupt_window_open) { + if (vmx_irq_enabled(vcpu)) { vector = kvm_cpu_get_interrupt(vcpu); vmx_inject_irq(vcpu, vector); kvm_timer_intr_post(vcpu, vector); @@ -2963,7 +3049,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) fixup_rmode_irq(vmx); vcpu->arch.interrupt_window_open = - (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & 3) == 0; + (vmcs_read32(GUEST_INTERRUPTIBILITY_INFO) & + (GUEST_INTR_STATE_STI | GUEST_INTR_STATE_MOV_SS)) == 0; asm("mov %0, %%ds; mov %0, %%es" : : "r"(__USER_DS)); vmx->launched = 1; @@ -2971,7 +3058,8 @@ static void vmx_vcpu_run(struct kvm_vcpu *vcpu, struct kvm_run *kvm_run) intr_info = vmcs_read32(VM_EXIT_INTR_INFO); /* We need to handle NMIs before interrupts are enabled */ - if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200) { /* nmi */ + if ((intr_info & INTR_INFO_INTR_TYPE_MASK) == 0x200 && + (intr_info & INTR_INFO_VALID_MASK)) { KVMTRACE_0D(NMI, vcpu, handler); asm("int $2"); } diff --git a/arch/x86/kvm/vmx.h b/arch/x86/kvm/vmx.h index 79d94c6..425a134 100644 --- a/arch/x86/kvm/vmx.h +++ b/arch/x86/kvm/vmx.h @@ -40,6 +40,7 @@ #define CPU_BASED_CR8_LOAD_EXITING 0x00080000 #define CPU_BASED_CR8_STORE_EXITING 0x00100000 #define CPU_BASED_TPR_SHADOW 0x00200000 +#define CPU_BASED_VIRTUAL_NMI_PENDING 0x00400000 #define CPU_BASED_MOV_DR_EXITING 0x00800000 #define CPU_BASED_UNCOND_IO_EXITING 0x01000000 #define CPU_BASED_USE_IO_BITMAPS 0x02000000 @@ -216,7 +217,7 @@ enum vmcs_field { #define EXIT_REASON_TRIPLE_FAULT 2 #define EXIT_REASON_PENDING_INTERRUPT 7 - +#define EXIT_REASON_NMI_WINDOW 8 #define EXIT_REASON_TASK_SWITCH 9 #define EXIT_REASON_CPUID 10 #define EXIT_REASON_HLT 12 @@ -251,7 +252,9 @@ enum vmcs_field { #define INTR_INFO_VECTOR_MASK 0xff /* 7:0 */ #define INTR_INFO_INTR_TYPE_MASK 0x700 /* 10:8 */ #define INTR_INFO_DELIVER_CODE_MASK 0x800 /* 11 */ +#define INTR_INFO_UNBLOCK_NMI 0x1000 /* 12 */ #define INTR_INFO_VALID_MASK 0x80000000 /* 31 */ +#define INTR_INFO_RESVD_BITS_MASK 0x7ffff000 #define VECTORING_INFO_VECTOR_MASK INTR_INFO_VECTOR_MASK #define VECTORING_INFO_TYPE_MASK INTR_INFO_INTR_TYPE_MASK @@ -259,9 +262,16 @@ enum vmcs_field { #define VECTORING_INFO_VALID_MASK INTR_INFO_VALID_MASK #define INTR_TYPE_EXT_INTR (0 << 8) /* external interrupt */ +#define INTR_TYPE_NMI_INTR (2 << 8) /* NMI */ #define INTR_TYPE_EXCEPTION (3 << 8) /* processor exception */ #define INTR_TYPE_SOFT_INTR (4 << 8) /* software interrupt */ +/* GUEST_INTERRUPTIBILITY_INFO flags. */ +#define GUEST_INTR_STATE_STI 0x00000001 +#define GUEST_INTR_STATE_MOV_SS 0x00000002 +#define GUEST_INTR_STATE_SMI 0x00000004 +#define GUEST_INTR_STATE_NMI 0x00000008 + /* * Exit Qualifications for MOV for Control Register Access */ diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 2c308f0..cab5f92 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -72,6 +72,7 @@ struct kvm_stats_debugfs_item debugfs_entries[] = { { "mmio_exits", VCPU_STAT(mmio_exits) }, { "signal_exits", VCPU_STAT(signal_exits) }, { "irq_window", VCPU_STAT(irq_window_exits) }, + { "nmi_window", VCPU_STAT(nmi_window_exits) }, { "halt_exits", VCPU_STAT(halt_exits) }, { "halt_wakeup", VCPU_STAT(halt_wakeup) }, { "hypercalls", VCPU_STAT(hypercalls) }, diff --git a/include/asm-x86/kvm_host.h b/include/asm-x86/kvm_host.h index b666219..bacb1e2 100644 --- a/include/asm-x86/kvm_host.h +++ b/include/asm-x86/kvm_host.h @@ -347,6 +347,7 @@ struct kvm_vcpu_stat { u32 mmio_exits; u32 signal_exits; u32 irq_window_exits; + u32 nmi_window_exits; u32 halt_exits; u32 halt_wakeup; u32 request_irq_exits; -- 1.5.6 -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/