From patchwork Wed Aug 30 12:42:40 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Maxim Levitsky X-Patchwork-Id: 13370389 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1A0B7C83F01 for ; Wed, 30 Aug 2023 18:32:27 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S231538AbjH3Sc1 (ORCPT ); Wed, 30 Aug 2023 14:32:27 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43622 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S244194AbjH3Mnp (ORCPT ); Wed, 30 Aug 2023 08:43:45 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 92424193 for ; Wed, 30 Aug 2023 05:42:56 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1693399375; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=JSiaWd3XTuuzrZMx0m5UZPazKAadIjf8HRjOqCLLZZ8=; b=aVed3lDRnj77wzN4y+cqVubJKgCePg5ANVumUeKzQ8wEbExErNvjfx7zanpdAIPE/xS8ng mJeKoZk4lP7sB4pQrJHETCser/bOPZmY+3vHsSyT7XSz0/dA1PlJSuBlMYMo9Wd8WZPrh1 2hD0wj77lrtKlymwk5982Cz3Hc5+mew= Received: from mimecast-mx02.redhat.com (66.187.233.73 [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-86-2BVjFtiyPIK734y9nBcB9w-1; Wed, 30 Aug 2023 08:42:51 -0400 X-MC-Unique: 2BVjFtiyPIK734y9nBcB9w-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id DEFAE1C0E4C1; Wed, 30 Aug 2023 12:42:50 +0000 (UTC) Received: from localhost.localdomain (unknown [10.45.224.27]) by smtp.corp.redhat.com (Postfix) with ESMTP id 4B1282026D35; Wed, 30 Aug 2023 12:42:48 +0000 (UTC) From: Maxim Levitsky To: kvm@vger.kernel.org Cc: "H. Peter Anvin" , linux-kernel@vger.kernel.org, Borislav Petkov , Thomas Gleixner , x86@kernel.org, Sean Christopherson , Ingo Molnar , Paolo Bonzini , Dave Hansen , Maxim Levitsky Subject: [PATCH 1/4] KVM: x86: refactor req_immediate_exit logic Date: Wed, 30 Aug 2023 15:42:40 +0300 Message-Id: <20230830124243.671152-2-mlevitsk@redhat.com> In-Reply-To: <20230830124243.671152-1-mlevitsk@redhat.com> References: <20230830124243.671152-1-mlevitsk@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org - move req_immediate_exit variable from arch specific to common code. - remove arch specific callback .request_immediate_exit and move the code down to the arch's vcpu_run's code. No functional change is intended. Signed-off-by: Maxim Levitsky --- arch/x86/include/asm/kvm-x86-ops.h | 1 - arch/x86/include/asm/kvm_host.h | 5 ++--- arch/x86/kvm/svm/svm.c | 5 +++-- arch/x86/kvm/vmx/vmx.c | 18 ++++++----------- arch/x86/kvm/vmx/vmx.h | 2 -- arch/x86/kvm/x86.c | 31 +++++++++++------------------- 6 files changed, 22 insertions(+), 40 deletions(-) diff --git a/arch/x86/include/asm/kvm-x86-ops.h b/arch/x86/include/asm/kvm-x86-ops.h index e3054e3e46d52d..f654a7f4cc8c0c 100644 --- a/arch/x86/include/asm/kvm-x86-ops.h +++ b/arch/x86/include/asm/kvm-x86-ops.h @@ -101,7 +101,6 @@ KVM_X86_OP(write_tsc_multiplier) KVM_X86_OP(get_exit_info) KVM_X86_OP(check_intercept) KVM_X86_OP(handle_exit_irqoff) -KVM_X86_OP(request_immediate_exit) KVM_X86_OP(sched_in) KVM_X86_OP_OPTIONAL(update_cpu_dirty_logging) KVM_X86_OP_OPTIONAL(vcpu_blocking) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 3bc146dfd38da5..5f2a93f92a3df7 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -993,6 +993,8 @@ struct kvm_vcpu_arch { */ bool pdptrs_from_userspace; + bool req_immediate_exit; + #if IS_ENABLED(CONFIG_HYPERV) hpa_t hv_root_tdp; #endif @@ -1672,8 +1674,6 @@ struct kvm_x86_ops { struct x86_exception *exception); void (*handle_exit_irqoff)(struct kvm_vcpu *vcpu); - void (*request_immediate_exit)(struct kvm_vcpu *vcpu); - void (*sched_in)(struct kvm_vcpu *kvm, int cpu); /* @@ -2157,7 +2157,6 @@ extern bool kvm_find_async_pf_gfn(struct kvm_vcpu *vcpu, gfn_t gfn); int kvm_skip_emulated_instruction(struct kvm_vcpu *vcpu); int kvm_complete_insn_gp(struct kvm_vcpu *vcpu, int err); -void __kvm_request_immediate_exit(struct kvm_vcpu *vcpu); void __user *__x86_set_memory_region(struct kvm *kvm, int id, gpa_t gpa, u32 size); diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index 956726d867aafc..18d0895975dfd8 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -4063,6 +4063,9 @@ static __no_kcsan fastpath_t svm_vcpu_run(struct kvm_vcpu *vcpu) clgi(); kvm_load_guest_xsave_state(vcpu); + if (vcpu->arch.req_immediate_exit) + smp_send_reschedule(vcpu->cpu); + kvm_wait_lapic_expire(vcpu); /* @@ -4878,8 +4881,6 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .check_intercept = svm_check_intercept, .handle_exit_irqoff = svm_handle_exit_irqoff, - .request_immediate_exit = __kvm_request_immediate_exit, - .sched_in = svm_sched_in, .nested_ops = &svm_nested_ops, diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index df461f387e20d4..77ef6f02b3f42f 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -68,6 +68,8 @@ #include "x86.h" #include "smm.h" +#include + MODULE_AUTHOR("Qumranet"); MODULE_LICENSE("GPL"); @@ -1274,8 +1276,6 @@ void vmx_prepare_switch_to_guest(struct kvm_vcpu *vcpu) u16 fs_sel, gs_sel; int i; - vmx->req_immediate_exit = false; - /* * Note that guest MSRs to be saved/restored can also be changed * when guest state is loaded. This happens when guest transitions @@ -5997,7 +5997,7 @@ static fastpath_t handle_fastpath_preemption_timer(struct kvm_vcpu *vcpu) { struct vcpu_vmx *vmx = to_vmx(vcpu); - if (!vmx->req_immediate_exit && + if (!vcpu->arch.req_immediate_exit && !unlikely(vmx->loaded_vmcs->hv_timer_soft_disabled)) { kvm_lapic_expired_hv_timer(vcpu); return EXIT_FASTPATH_REENTER_GUEST; @@ -7153,7 +7153,7 @@ static void vmx_update_hv_timer(struct kvm_vcpu *vcpu) u64 tscl; u32 delta_tsc; - if (vmx->req_immediate_exit) { + if (vcpu->arch.req_immediate_exit) { vmcs_write32(VMX_PREEMPTION_TIMER_VALUE, 0); vmx->loaded_vmcs->hv_timer_soft_disabled = false; } else if (vmx->hv_deadline_tsc != -1) { @@ -7348,6 +7348,8 @@ static fastpath_t vmx_vcpu_run(struct kvm_vcpu *vcpu) if (enable_preemption_timer) vmx_update_hv_timer(vcpu); + else if (vcpu->arch.req_immediate_exit) + smp_send_reschedule(vcpu->cpu); kvm_wait_lapic_expire(vcpu); @@ -7892,11 +7894,6 @@ static __init void vmx_set_cpu_caps(void) kvm_cpu_cap_check_and_set(X86_FEATURE_WAITPKG); } -static void vmx_request_immediate_exit(struct kvm_vcpu *vcpu) -{ - to_vmx(vcpu)->req_immediate_exit = true; -} - static int vmx_check_intercept_io(struct kvm_vcpu *vcpu, struct x86_instruction_info *info) { @@ -8305,8 +8302,6 @@ static struct kvm_x86_ops vmx_x86_ops __initdata = { .check_intercept = vmx_check_intercept, .handle_exit_irqoff = vmx_handle_exit_irqoff, - .request_immediate_exit = vmx_request_immediate_exit, - .sched_in = vmx_sched_in, .cpu_dirty_log_size = PML_ENTITY_NUM, @@ -8564,7 +8559,6 @@ static __init int hardware_setup(void) if (!enable_preemption_timer) { vmx_x86_ops.set_hv_timer = NULL; vmx_x86_ops.cancel_hv_timer = NULL; - vmx_x86_ops.request_immediate_exit = __kvm_request_immediate_exit; } kvm_caps.supported_mce_cap |= MCG_LMCE_P; diff --git a/arch/x86/kvm/vmx/vmx.h b/arch/x86/kvm/vmx/vmx.h index 32384ba3849949..96c94f69ef8311 100644 --- a/arch/x86/kvm/vmx/vmx.h +++ b/arch/x86/kvm/vmx/vmx.h @@ -330,8 +330,6 @@ struct vcpu_vmx { unsigned int ple_window; bool ple_window_dirty; - bool req_immediate_exit; - /* Support for PML */ #define PML_ENTITY_NUM 512 struct page *pml_pg; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 278dbd37dab244..53f0a5f493be37 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -10052,8 +10052,7 @@ static void kvm_inject_exception(struct kvm_vcpu *vcpu) * ordering between that side effect, the instruction completing, _and_ the * delivery of the asynchronous event. */ -static int kvm_check_and_inject_events(struct kvm_vcpu *vcpu, - bool *req_immediate_exit) +static int kvm_check_and_inject_events(struct kvm_vcpu *vcpu) { bool can_inject; int r; @@ -10230,8 +10229,9 @@ static int kvm_check_and_inject_events(struct kvm_vcpu *vcpu, if (is_guest_mode(vcpu) && kvm_x86_ops.nested_ops->has_events && - kvm_x86_ops.nested_ops->has_events(vcpu)) - *req_immediate_exit = true; + kvm_x86_ops.nested_ops->has_events(vcpu)) { + vcpu->arch.req_immediate_exit = true; + } /* * KVM must never queue a new exception while injecting an event; KVM @@ -10248,10 +10248,9 @@ static int kvm_check_and_inject_events(struct kvm_vcpu *vcpu, WARN_ON_ONCE(vcpu->arch.exception.pending || vcpu->arch.exception_vmexit.pending); return 0; - out: if (r == -EBUSY) { - *req_immediate_exit = true; + vcpu->arch.req_immediate_exit = true; r = 0; } return r; @@ -10478,12 +10477,6 @@ static void kvm_vcpu_reload_apic_access_page(struct kvm_vcpu *vcpu) static_call_cond(kvm_x86_set_apic_access_page_addr)(vcpu); } -void __kvm_request_immediate_exit(struct kvm_vcpu *vcpu) -{ - smp_send_reschedule(vcpu->cpu); -} -EXPORT_SYMBOL_GPL(__kvm_request_immediate_exit); - /* * Called within kvm->srcu read side. * Returns 1 to let vcpu_run() continue the guest execution loop without @@ -10498,8 +10491,6 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) kvm_cpu_accept_dm_intr(vcpu); fastpath_t exit_fastpath; - bool req_immediate_exit = false; - if (kvm_request_pending(vcpu)) { if (kvm_check_request(KVM_REQ_VM_DEAD, vcpu)) { r = -EIO; @@ -10660,7 +10651,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) goto out; } - r = kvm_check_and_inject_events(vcpu, &req_immediate_exit); + r = kvm_check_and_inject_events(vcpu); if (r < 0) { r = 0; goto out; @@ -10729,10 +10720,9 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) goto cancel_injection; } - if (req_immediate_exit) { + + if (vcpu->arch.req_immediate_exit) kvm_make_request(KVM_REQ_EVENT, vcpu); - static_call(kvm_x86_request_immediate_exit)(vcpu); - } fpregs_assert_state_consistent(); if (test_thread_flag(TIF_NEED_FPU_LOAD)) @@ -10764,6 +10754,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) (kvm_get_apic_mode(vcpu) != LAPIC_MODE_DISABLED)); exit_fastpath = static_call(kvm_x86_vcpu_run)(vcpu); + if (likely(exit_fastpath != EXIT_FASTPATH_REENTER_GUEST)) break; @@ -10779,6 +10770,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) ++vcpu->stat.exits; } + vcpu->arch.req_immediate_exit = false; /* * Do this here before restoring debug registers on the host. And * since we do this before handling the vmexit, a DR access vmexit @@ -10866,8 +10858,7 @@ static int vcpu_enter_guest(struct kvm_vcpu *vcpu) return r; cancel_injection: - if (req_immediate_exit) - kvm_make_request(KVM_REQ_EVENT, vcpu); + vcpu->arch.req_immediate_exit = false; static_call(kvm_x86_cancel_injection)(vcpu); if (unlikely(vcpu->arch.apic_attention)) kvm_lapic_sync_from_vapic(vcpu); From patchwork Wed Aug 30 12:42:41 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Maxim Levitsky X-Patchwork-Id: 13370392 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id DF1A5C83F21 for ; Wed, 30 Aug 2023 18:32:59 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S235788AbjH3SdA (ORCPT ); Wed, 30 Aug 2023 14:33:00 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:43634 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S244193AbjH3Mnp (ORCPT ); Wed, 30 Aug 2023 08:43:45 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 38569CC5 for ; Wed, 30 Aug 2023 05:42:57 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1693399376; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=cXmGRleFgxAJpob9+MyMwqaQ6p2zsPjd+uLwoaCoCBo=; b=SbMfEnnKvyvoe2wqtKK8nrYXFlIAJMh/6NdYAaMGcGrDVtd0eds77uWU86IlU17tp47JPl SGH7ZwicX1t8tY+WGJNUVgTFvSLEYIuTE9voV9/04ICOqECQyoHQY9pfAtO8YeYjx+VCyj n2Tmhts/Zh57/iBDPnXJ7UGXxz+lGTA= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-63-e4BA0LOoNSePE1uuIFVhQQ-1; Wed, 30 Aug 2023 08:42:54 -0400 X-MC-Unique: e4BA0LOoNSePE1uuIFVhQQ-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id A694E800CAB; Wed, 30 Aug 2023 12:42:53 +0000 (UTC) Received: from localhost.localdomain (unknown [10.45.224.27]) by smtp.corp.redhat.com (Postfix) with ESMTP id 429ED2026D35; Wed, 30 Aug 2023 12:42:51 +0000 (UTC) From: Maxim Levitsky To: kvm@vger.kernel.org Cc: "H. Peter Anvin" , linux-kernel@vger.kernel.org, Borislav Petkov , Thomas Gleixner , x86@kernel.org, Sean Christopherson , Ingo Molnar , Paolo Bonzini , Dave Hansen , Maxim Levitsky Subject: [PATCH 2/4] KVM: x86: add more information to the kvm_entry tracepoint Date: Wed, 30 Aug 2023 15:42:41 +0300 Message-Id: <20230830124243.671152-3-mlevitsk@redhat.com> In-Reply-To: <20230830124243.671152-1-mlevitsk@redhat.com> References: <20230830124243.671152-1-mlevitsk@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Add: - Flag showing that VM is in a guest mode on entry. - Flag showing that immediate vm exit is set to happen after the entry. - VMX/SVM specific interrupt injection info (like in a vm exit). Signed-off-by: Maxim Levitsky --- arch/x86/include/asm/kvm-x86-ops.h | 1 + arch/x86/include/asm/kvm_host.h | 5 ++++- arch/x86/kvm/svm/svm.c | 17 +++++++++++++++++ arch/x86/kvm/trace.h | 19 +++++++++++++++++-- arch/x86/kvm/vmx/vmx.c | 12 ++++++++++++ 5 files changed, 51 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/kvm-x86-ops.h b/arch/x86/include/asm/kvm-x86-ops.h index f654a7f4cc8c0c..346fed6e3c33aa 100644 --- a/arch/x86/include/asm/kvm-x86-ops.h +++ b/arch/x86/include/asm/kvm-x86-ops.h @@ -99,6 +99,7 @@ KVM_X86_OP(get_l2_tsc_multiplier) KVM_X86_OP(write_tsc_offset) KVM_X86_OP(write_tsc_multiplier) KVM_X86_OP(get_exit_info) +KVM_X86_OP(get_entry_info) KVM_X86_OP(check_intercept) KVM_X86_OP(handle_exit_irqoff) KVM_X86_OP(sched_in) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 5f2a93f92a3df7..9cc5ac420f4052 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -1661,13 +1661,16 @@ struct kvm_x86_ops { void (*write_tsc_multiplier)(struct kvm_vcpu *vcpu, u64 multiplier); /* - * Retrieve somewhat arbitrary exit information. Intended to + * Retrieve somewhat arbitrary exit/entry information. Intended to * be used only from within tracepoints or error paths. */ void (*get_exit_info)(struct kvm_vcpu *vcpu, u32 *reason, u64 *info1, u64 *info2, u32 *exit_int_info, u32 *exit_int_info_err_code); + void (*get_entry_info)(struct kvm_vcpu *vcpu, + u32 *inj_info, u32 *inj_info_error_code); + int (*check_intercept)(struct kvm_vcpu *vcpu, struct x86_instruction_info *info, enum x86_intercept_stage stage, diff --git a/arch/x86/kvm/svm/svm.c b/arch/x86/kvm/svm/svm.c index 18d0895975dfd8..dd09cb8d7eaebf 100644 --- a/arch/x86/kvm/svm/svm.c +++ b/arch/x86/kvm/svm/svm.c @@ -3405,6 +3405,22 @@ static void svm_get_exit_info(struct kvm_vcpu *vcpu, u32 *reason, *error_code = 0; } +static void svm_get_entry_info(struct kvm_vcpu *vcpu, + u32 *inj_info, + u32 *inj_info_error_code) +{ + struct vmcb_control_area *control = &to_svm(vcpu)->vmcb->control; + + *inj_info = control->event_inj; + + if ((*inj_info & SVM_EXITINTINFO_VALID) && + (*inj_info & SVM_EXITINTINFO_VALID_ERR)) + *inj_info_error_code = control->event_inj_err; + else + *inj_info_error_code = 0; + +} + static int svm_handle_exit(struct kvm_vcpu *vcpu, fastpath_t exit_fastpath) { struct vcpu_svm *svm = to_svm(vcpu); @@ -4866,6 +4882,7 @@ static struct kvm_x86_ops svm_x86_ops __initdata = { .required_apicv_inhibits = AVIC_REQUIRED_APICV_INHIBITS, .get_exit_info = svm_get_exit_info, + .get_entry_info = svm_get_entry_info, .vcpu_after_set_cpuid = svm_vcpu_after_set_cpuid, diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h index 83843379813ee3..f4c56f59f5c11b 100644 --- a/arch/x86/kvm/trace.h +++ b/arch/x86/kvm/trace.h @@ -21,14 +21,29 @@ TRACE_EVENT(kvm_entry, TP_STRUCT__entry( __field( unsigned int, vcpu_id ) __field( unsigned long, rip ) - ), + __field( u32, inj_info ) + __field( u32, inj_info_err ) + __field( bool, guest_mode ) + __field( bool, req_imm_exit ) + ), TP_fast_assign( __entry->vcpu_id = vcpu->vcpu_id; __entry->rip = kvm_rip_read(vcpu); + + static_call(kvm_x86_get_entry_info)(vcpu, + &__entry->inj_info, + &__entry->inj_info_err); + + __entry->req_imm_exit = vcpu->arch.req_immediate_exit; + __entry->guest_mode = is_guest_mode(vcpu); ), - TP_printk("vcpu %u, rip 0x%lx", __entry->vcpu_id, __entry->rip) + TP_printk("vcpu %u, rip 0x%lx inj 0x%08x inj_error_code 0x%08x%s%s", + __entry->vcpu_id, __entry->rip, + __entry->inj_info, __entry->inj_info_err, + __entry->req_imm_exit ? " [imm exit]" : "", + __entry->guest_mode ? " [guest]" : "") ); /* diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index 77ef6f02b3f42f..d662d546232171 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -6157,6 +6157,17 @@ static void vmx_get_exit_info(struct kvm_vcpu *vcpu, u32 *reason, } } +static void vmx_get_entry_info(struct kvm_vcpu *vcpu, + u32 *inj_info, + u32 *inj_info_error_code) +{ + *inj_info = vmcs_read32(VM_ENTRY_INTR_INFO_FIELD); + if (is_exception_with_error_code(*inj_info)) + *inj_info_error_code = vmcs_read32(VM_ENTRY_EXCEPTION_ERROR_CODE); + else + *inj_info_error_code = 0; +} + static void vmx_destroy_pml_buffer(struct vcpu_vmx *vmx) { if (vmx->pml_pg) { @@ -8287,6 +8298,7 @@ static struct kvm_x86_ops vmx_x86_ops __initdata = { .get_mt_mask = vmx_get_mt_mask, .get_exit_info = vmx_get_exit_info, + .get_entry_info = vmx_get_entry_info, .vcpu_after_set_cpuid = vmx_vcpu_after_set_cpuid, From patchwork Wed Aug 30 12:42:42 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Maxim Levitsky X-Patchwork-Id: 13370393 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 5497FC6FA8F for ; Wed, 30 Aug 2023 18:33:05 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S237059AbjH3SdD (ORCPT ); Wed, 30 Aug 2023 14:33:03 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52342 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S244201AbjH3Mo2 (ORCPT ); Wed, 30 Aug 2023 08:44:28 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.133.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 181E7CDA for ; Wed, 30 Aug 2023 05:43:03 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1693399382; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=woWU04N+glUzD420ymk+4+G9vxtbf/gGrQWI0ejKWmM=; b=Qd9KWMU74M4LEi1UbERFh3n1U/aBoTZUNwZH1iyElXHxlzH4VllVVcsD9DquCslNsq8L9K bWLQ/Ume0xSWrR17uMLpvYSg/2rimbRq8a3IGUyshz0K1Fpg5iKgpiVAdcjVc7pmICFlX1 lIcnsBZUjZE/j+fWRtQDxnPgRhsI+nQ= Received: from mimecast-mx02.redhat.com (66.187.233.73 [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-442-tEX4fSVcNq-3tP4VwYPVwQ-1; Wed, 30 Aug 2023 08:42:57 -0400 X-MC-Unique: tEX4fSVcNq-3tP4VwYPVwQ-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 793751C0E4C4; Wed, 30 Aug 2023 12:42:56 +0000 (UTC) Received: from localhost.localdomain (unknown [10.45.224.27]) by smtp.corp.redhat.com (Postfix) with ESMTP id 0F98E2026D35; Wed, 30 Aug 2023 12:42:53 +0000 (UTC) From: Maxim Levitsky To: kvm@vger.kernel.org Cc: "H. Peter Anvin" , linux-kernel@vger.kernel.org, Borislav Petkov , Thomas Gleixner , x86@kernel.org, Sean Christopherson , Ingo Molnar , Paolo Bonzini , Dave Hansen , Maxim Levitsky Subject: [PATCH 3/4] KVM: x86: add more information to kvm_exit tracepoint Date: Wed, 30 Aug 2023 15:42:42 +0300 Message-Id: <20230830124243.671152-4-mlevitsk@redhat.com> In-Reply-To: <20230830124243.671152-1-mlevitsk@redhat.com> References: <20230830124243.671152-1-mlevitsk@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Add: - Flag that shows that the VM is in guest mode - Bitmap of pending kvm requests. - Flag showing that this VM exit is due to request to have an immediate VM exit. Signed-off-by: Maxim Levitsky --- arch/x86/kvm/trace.h | 14 ++++++++++++-- 1 file changed, 12 insertions(+), 2 deletions(-) diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h index f4c56f59f5c11b..0657a3a348b4ae 100644 --- a/arch/x86/kvm/trace.h +++ b/arch/x86/kvm/trace.h @@ -320,12 +320,18 @@ TRACE_EVENT(name, \ __field( u32, intr_info ) \ __field( u32, error_code ) \ __field( unsigned int, vcpu_id ) \ + __field( bool, guest_mode ) \ + __field( u64, requests ) \ + __field( bool, req_imm_exit ) \ ), \ \ TP_fast_assign( \ __entry->guest_rip = kvm_rip_read(vcpu); \ __entry->isa = isa; \ __entry->vcpu_id = vcpu->vcpu_id; \ + __entry->guest_mode = is_guest_mode(vcpu); \ + __entry->requests = READ_ONCE(vcpu->requests); \ + __entry->req_imm_exit = vcpu->arch.req_immediate_exit; \ static_call(kvm_x86_get_exit_info)(vcpu, \ &__entry->exit_reason, \ &__entry->info1, \ @@ -335,11 +341,15 @@ TRACE_EVENT(name, \ ), \ \ TP_printk("vcpu %u reason %s%s%s rip 0x%lx info1 0x%016llx " \ - "info2 0x%016llx intr_info 0x%08x error_code 0x%08x", \ + "info2 0x%016llx intr_info 0x%08x error_code 0x%08x " \ + "requests 0x%016llx%s%s", \ __entry->vcpu_id, \ kvm_print_exit_reason(__entry->exit_reason, __entry->isa), \ __entry->guest_rip, __entry->info1, __entry->info2, \ - __entry->intr_info, __entry->error_code) \ + __entry->intr_info, __entry->error_code, \ + __entry->requests, \ + __entry->guest_mode ? " [guest]" : "", \ + __entry->req_imm_exit ? " [imm exit]" : "") \ ) /* From patchwork Wed Aug 30 12:42:43 2023 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Maxim Levitsky X-Patchwork-Id: 13370390 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id 88F36C83F01 for ; Wed, 30 Aug 2023 18:32:43 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S232160AbjH3Scg (ORCPT ); Wed, 30 Aug 2023 14:32:36 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52354 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S244199AbjH3Mo2 (ORCPT ); Wed, 30 Aug 2023 08:44:28 -0400 Received: from us-smtp-delivery-124.mimecast.com (us-smtp-delivery-124.mimecast.com [170.10.129.124]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id EB18ACD8 for ; Wed, 30 Aug 2023 05:43:05 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=redhat.com; s=mimecast20190719; t=1693399385; h=from:from:reply-to:subject:subject:date:date:message-id:message-id: to:to:cc:cc:mime-version:mime-version: content-transfer-encoding:content-transfer-encoding: in-reply-to:in-reply-to:references:references; bh=krCDjpp4Y6dP2ieUItA5y6v8pExPMlLnXu/D3aGCLS0=; b=YhAC1g3HsTU1iuBjwd/zcM/8X5d/J6jgbERNOTMF4eYER9/PfIdAz/DJusipcNb+yEkkrI 9BMR/ZZTgLBS6LkjJx4pABcaVej+VOpv8cmtVEH+W3sbB0tkdsNweijvLuW1+tgSp1Hba0 JoDH7J8Udkik6pQmVdyV2E8mRSoig5Y= Received: from mimecast-mx02.redhat.com (mimecast-mx02.redhat.com [66.187.233.88]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.2, cipher=TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384) id us-mta-175-SOhcxdbeMd-iHx-79YsjWg-1; Wed, 30 Aug 2023 08:42:59 -0400 X-MC-Unique: SOhcxdbeMd-iHx-79YsjWg-1 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 40F0880B1EA; Wed, 30 Aug 2023 12:42:59 +0000 (UTC) Received: from localhost.localdomain (unknown [10.45.224.27]) by smtp.corp.redhat.com (Postfix) with ESMTP id D19782026D35; Wed, 30 Aug 2023 12:42:56 +0000 (UTC) From: Maxim Levitsky To: kvm@vger.kernel.org Cc: "H. Peter Anvin" , linux-kernel@vger.kernel.org, Borislav Petkov , Thomas Gleixner , x86@kernel.org, Sean Christopherson , Ingo Molnar , Paolo Bonzini , Dave Hansen , Maxim Levitsky Subject: [PATCH 4/4] KVM: x86: add new nested vmexit tracepoints Date: Wed, 30 Aug 2023 15:42:43 +0300 Message-Id: <20230830124243.671152-5-mlevitsk@redhat.com> In-Reply-To: <20230830124243.671152-1-mlevitsk@redhat.com> References: <20230830124243.671152-1-mlevitsk@redhat.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.1 on 10.11.54.4 Precedence: bulk List-ID: X-Mailing-List: kvm@vger.kernel.org Add 3 new tracepoints for nested VM exits which are intended to capture extra information to gain insights about the nested guest behavior. The new tracepoints are: - kvm_nested_msr - kvm_nested_hypercall These tracepoints capture extra register state to be able to know which MSR or which hypercall was done. - kvm_nested_page_fault This tracepoint allows to capture extra info about which host pagefault error code caused the nested page fault. Signed-off-by: Maxim Levitsky --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/svm/nested.c | 22 +++++++++ arch/x86/kvm/trace.h | 82 +++++++++++++++++++++++++++++++-- arch/x86/kvm/vmx/nested.c | 21 +++++++++ arch/x86/kvm/vmx/vmx.c | 1 + arch/x86/kvm/x86.c | 3 ++ 6 files changed, 127 insertions(+), 3 deletions(-) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h index 9cc5ac420f4052..9a72a078d126c9 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -948,6 +948,7 @@ struct kvm_vcpu_arch { /* set at EPT violation at this point */ unsigned long exit_qualification; + u32 ept_fault_error_code; /* pv related host specific info */ struct { diff --git a/arch/x86/kvm/svm/nested.c b/arch/x86/kvm/svm/nested.c index 96936ddf1b3c56..d24bd21fbdc728 100644 --- a/arch/x86/kvm/svm/nested.c +++ b/arch/x86/kvm/svm/nested.c @@ -38,6 +38,8 @@ static void nested_svm_inject_npf_exit(struct kvm_vcpu *vcpu, { struct vcpu_svm *svm = to_svm(vcpu); struct vmcb *vmcb = svm->vmcb; + u64 host_error_code = vmcb->control.exit_info_1; + if (vmcb->control.exit_code != SVM_EXIT_NPF) { /* @@ -48,11 +50,15 @@ static void nested_svm_inject_npf_exit(struct kvm_vcpu *vcpu, vmcb->control.exit_code_hi = 0; vmcb->control.exit_info_1 = (1ULL << 32); vmcb->control.exit_info_2 = fault->address; + host_error_code = 0; } vmcb->control.exit_info_1 &= ~0xffffffffULL; vmcb->control.exit_info_1 |= fault->error_code; + trace_kvm_nested_page_fault(fault->address, host_error_code, + fault->error_code); + nested_svm_vmexit(svm); } @@ -1130,6 +1136,22 @@ int nested_svm_vmexit(struct vcpu_svm *svm) vmcb12->control.exit_int_info_err, KVM_ISA_SVM); + /* Collect some info about nested VM exits */ + switch (vmcb12->control.exit_code) { + case SVM_EXIT_MSR: + trace_kvm_nested_msr(vmcb12->control.exit_info_1 == 1, + kvm_rcx_read(vcpu), + (vmcb12->save.rax & -1u) | + (((u64)(kvm_rdx_read(vcpu) & -1u) << 32))); + break; + case SVM_EXIT_VMMCALL: + trace_kvm_nested_hypercall(vmcb12->save.rax, + kvm_rbx_read(vcpu), + kvm_rcx_read(vcpu), + kvm_rdx_read(vcpu)); + break; + } + kvm_vcpu_unmap(vcpu, &map, true); nested_svm_transition_tlb_flush(vcpu); diff --git a/arch/x86/kvm/trace.h b/arch/x86/kvm/trace.h index 0657a3a348b4ae..d08ae87c536324 100644 --- a/arch/x86/kvm/trace.h +++ b/arch/x86/kvm/trace.h @@ -620,7 +620,7 @@ TRACE_EVENT(kvm_pv_eoi, ); /* - * Tracepoint for nested VMRUN + * Tracepoint for nested VMRUN/VMENTER */ TRACE_EVENT(kvm_nested_vmenter, TP_PROTO(__u64 rip, __u64 vmcb, __u64 nested_rip, __u32 int_ctl, @@ -753,8 +753,84 @@ TRACE_EVENT(kvm_nested_intr_vmexit, TP_printk("rip: 0x%016llx", __entry->rip) ); + /* - * Tracepoint for nested #vmexit because of interrupt pending + * Tracepoint for nested guest MSR access. + */ +TRACE_EVENT(kvm_nested_msr, + TP_PROTO(bool write, u32 ecx, u64 data), + TP_ARGS(write, ecx, data), + + TP_STRUCT__entry( + __field( bool, write ) + __field( u32, ecx ) + __field( u64, data ) + ), + + TP_fast_assign( + __entry->write = write; + __entry->ecx = ecx; + __entry->data = data; + ), + + TP_printk("msr_%s %x = 0x%llx", + __entry->write ? "write" : "read", + __entry->ecx, __entry->data) +); + +/* + * Tracepoint for nested hypercalls, capturing generic info about the + * hypercall + */ + +TRACE_EVENT(kvm_nested_hypercall, + TP_PROTO(u64 rax, u64 rbx, u64 rcx, u64 rdx), + TP_ARGS(rax, rbx, rcx, rdx), + + TP_STRUCT__entry( + __field( u64, rax ) + __field( u64, rbx ) + __field( u64, rcx ) + __field( u64, rdx ) + ), + + TP_fast_assign( + __entry->rax = rax; + __entry->rbx = rbx; + __entry->rcx = rcx; + __entry->rdx = rdx; + ), + + TP_printk("rax 0x%llx rbx 0x%llx rcx 0x%llx rdx 0x%llx", + __entry->rax, __entry->rbx, __entry->rcx, __entry->rdx) +); + + +TRACE_EVENT(kvm_nested_page_fault, + TP_PROTO(u64 gpa, u64 host_error_code, u64 guest_error_code), + TP_ARGS(gpa, host_error_code, guest_error_code), + + TP_STRUCT__entry( + __field( u64, gpa ) + __field( u64, host_error_code ) + __field( u64, guest_errror_code ) + ), + + TP_fast_assign( + __entry->gpa = gpa; + __entry->host_error_code = host_error_code; + __entry->guest_errror_code = guest_error_code; + ), + + TP_printk("gpa 0x%llx host err 0x%llx guest err 0x%llx", + __entry->gpa, + __entry->host_error_code, + __entry->guest_errror_code) +); + + +/* + * Tracepoint for invlpga */ TRACE_EVENT(kvm_invlpga, TP_PROTO(__u64 rip, int asid, u64 address), @@ -777,7 +853,7 @@ TRACE_EVENT(kvm_invlpga, ); /* - * Tracepoint for nested #vmexit because of interrupt pending + * Tracepoint for skinit */ TRACE_EVENT(kvm_skinit, TP_PROTO(__u64 rip, __u32 slb), diff --git a/arch/x86/kvm/vmx/nested.c b/arch/x86/kvm/vmx/nested.c index 516391cc0d64fb..303472aa4d7e85 100644 --- a/arch/x86/kvm/vmx/nested.c +++ b/arch/x86/kvm/vmx/nested.c @@ -402,6 +402,10 @@ static void nested_ept_inject_page_fault(struct kvm_vcpu *vcpu, */ nested_ept_invalidate_addr(vcpu, vmcs12->ept_pointer, fault->address); + + trace_kvm_nested_page_fault(fault->address, + vcpu->arch.ept_fault_error_code, + fault->error_code); } nested_vmx_vmexit(vcpu, vm_exit_reason, 0, exit_qualification); @@ -4877,6 +4881,23 @@ void nested_vmx_vmexit(struct kvm_vcpu *vcpu, u32 vm_exit_reason, vmcs12->vm_exit_intr_error_code, KVM_ISA_VMX); + switch ((u16)vmcs12->vm_exit_reason) { + case EXIT_REASON_MSR_READ: + case EXIT_REASON_MSR_WRITE: + trace_kvm_nested_msr(vmcs12->vm_exit_reason == EXIT_REASON_MSR_WRITE, + kvm_rcx_read(vcpu), + (kvm_rax_read(vcpu) & -1u) | + (((u64)(kvm_rdx_read(vcpu) & -1u) << 32))); + break; + case EXIT_REASON_VMCALL: + trace_kvm_nested_hypercall(kvm_rax_read(vcpu), + kvm_rbx_read(vcpu), + kvm_rcx_read(vcpu), + kvm_rdx_read(vcpu)); + break; + + } + load_vmcs12_host_state(vcpu, vmcs12); return; diff --git a/arch/x86/kvm/vmx/vmx.c b/arch/x86/kvm/vmx/vmx.c index d662d546232171..649f09a350e9a1 100644 --- a/arch/x86/kvm/vmx/vmx.c +++ b/arch/x86/kvm/vmx/vmx.c @@ -5774,6 +5774,7 @@ static int handle_ept_violation(struct kvm_vcpu *vcpu) PFERR_GUEST_FINAL_MASK : PFERR_GUEST_PAGE_MASK; vcpu->arch.exit_qualification = exit_qualification; + vcpu->arch.ept_fault_error_code = error_code; /* * Check that the GPA doesn't exceed physical memory limits, as that is diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 53f0a5f493be37..d46b9f3f44428b 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -13628,6 +13628,9 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_vmenter); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_vmexit); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_vmexit_inject); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_intr_vmexit); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_hypercall); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_page_fault); +EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_msr); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_nested_vmenter_failed); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_invlpga); EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_skinit);