From nobody Sat Jun 20 00:55:26 2026 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 70D3EC433FE for ; Fri, 25 Mar 2022 13:59:41 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1358819AbiCYOBM (ORCPT ); Fri, 25 Mar 2022 10:01:12 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34170 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1352923AbiCYOBB (ORCPT ); Fri, 25 Mar 2022 10:01:01 -0400 Received: from mail-pj1-x102f.google.com (mail-pj1-x102f.google.com [IPv6:2607:f8b0:4864:20::102f]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id DBFEC5E75D; Fri, 25 Mar 2022 06:59:27 -0700 (PDT) Received: by mail-pj1-x102f.google.com with SMTP id n7-20020a17090aab8700b001c6aa871860so8413846pjq.2; Fri, 25 Mar 2022 06:59:27 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=gDqiN5pf4YDkXe9dP0rfveBOw3P+FW2+px95EQefoNY=; b=kkYNqJpY8u9gDoA9DnmCRX7/62AGWobuqKlzAlcYN+uWXouh/BeH2tZLvpP1Pe5x8N odUpBRTSCOad2wiq7N7r6BVHd7Tq3kfkf1E02MNx4v1o6iot5eEohU3h2MArIvqoLUNO WEYzgPqODwFOQ94RBABFjlSNGNeRGTu6VPYltarL+8TlZXyC74qo2nKyoUnVDwakPW1v bgK6qRtDpIXuMhwsEByfEpza4BqRiySjBynWM/jG0csg6eMtKEosH73rYuew+iXlXyw8 Dkt8JcgmXT3tshkJhLi3xirQ6/0pQ6HDHktRByei3kU53y6tGeytGASCqa3e86ICPyt5 nkCg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=gDqiN5pf4YDkXe9dP0rfveBOw3P+FW2+px95EQefoNY=; b=cgy9X8k9rGAq5k/iZsEGBSrlG+hsmnsGziCqJUfCX9jYhDdiL8pTynxpV+UWsFk4Cp jLfF0fFHu69qtJSuMeDtCgfvHJFZBwKPqv/Mj7eko2R7sbYS2XrD+jbUDRe4/qrhmmGP a4oIlASfLVYcbpot+rxWjGkpOaseavxmdBQWzfw3HGqzFxSycQ+b8yiikcSapBQcEZkm eFVXhDxJMpe2OY2QFePhwt3MRN17bclRENU/p3R2yfH7251fNs31Q4rIf5doeMIMcdbu L+uR+1Za1rGWstnIVwJ4Rm3XvrR6MDfo/EtTW9GQfKXvPifmTycexs1Yzwpe2Yv7Hk28 yYeA== X-Gm-Message-State: AOAM5324bCPH6u99iSnT8HmKOWienGgMvC/LFWcxQ3UqjU1cReOJComO eBr9aiEGFNwG4X+iHTPv9rMj3/EJEO0= X-Google-Smtp-Source: ABdhPJz7rYWP7B/KvxYgG8AMv5MkxbfY9Bt+GYvPawceO5AABPa4vilU0dOYqMC6kplGuZmI2PtUzg== X-Received: by 2002:a17:902:7888:b0:148:adf2:9725 with SMTP id q8-20020a170902788800b00148adf29725mr11724634pll.136.1648216767272; Fri, 25 Mar 2022 06:59:27 -0700 (PDT) Received: from localhost.localdomain ([203.205.141.113]) by smtp.googlemail.com with ESMTPSA id lw4-20020a17090b180400b001c7327d09c3sm14470875pjb.53.2022.03.25.06.59.24 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Fri, 25 Mar 2022 06:59:27 -0700 (PDT) From: Wanpeng Li X-Google-Original-From: Wanpeng Li To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org Cc: Paolo Bonzini , Sean Christopherson , Vitaly Kuznetsov , Wanpeng Li , Jim Mattson , Joerg Roedel Subject: [PATCH RESEND 1/5] KVM: X86: Add MSR_KVM_PREEMPT_COUNT support Date: Fri, 25 Mar 2022 06:58:25 -0700 Message-Id: <1648216709-44755-2-git-send-email-wanpengli@tencent.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> References: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" From: Wanpeng Li x86 preempt_count is per-cpu, any non-zero value for preempt_count indicates that either preemption has been disabled explicitly or the=20 CPU is currently servicing some sort of interrupt. The guest will=20 pass this value to the hypervisor, so the hypervisor knows whether=20 the guest is running in the critical section. Signed-off-by: Wanpeng Li --- arch/x86/include/asm/kvm_host.h | 6 +++++ arch/x86/include/uapi/asm/kvm_para.h | 2 ++ arch/x86/kvm/x86.c | 35 ++++++++++++++++++++++++++++ 3 files changed, 43 insertions(+) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_hos= t.h index f72e80178ffc..50f011a7445a 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -856,6 +856,12 @@ struct kvm_vcpu_arch { =20 u64 msr_kvm_poll_control; =20 + struct { + u64 msr_val; + bool preempt_count_enabled; + struct gfn_to_hva_cache preempt_count_cache; + } pv_pc; + /* * Indicates the guest is trying to write a gfn that contains one or * more of the PTEs used to translate the write itself, i.e. the access diff --git a/arch/x86/include/uapi/asm/kvm_para.h b/arch/x86/include/uapi/a= sm/kvm_para.h index 6e64b27b2c1e..f99fa4407604 100644 --- a/arch/x86/include/uapi/asm/kvm_para.h +++ b/arch/x86/include/uapi/asm/kvm_para.h @@ -36,6 +36,7 @@ #define KVM_FEATURE_MSI_EXT_DEST_ID 15 #define KVM_FEATURE_HC_MAP_GPA_RANGE 16 #define KVM_FEATURE_MIGRATION_CONTROL 17 +#define KVM_FEATURE_PREEMPT_COUNT 18 =20 #define KVM_HINTS_REALTIME 0 =20 @@ -58,6 +59,7 @@ #define MSR_KVM_ASYNC_PF_INT 0x4b564d06 #define MSR_KVM_ASYNC_PF_ACK 0x4b564d07 #define MSR_KVM_MIGRATION_CONTROL 0x4b564d08 +#define MSR_KVM_PREEMPT_COUNT 0x4b564d09 =20 struct kvm_steal_time { __u64 steal; diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 51106d32f04e..af75e273cb32 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -1456,6 +1456,7 @@ static const u32 emulated_msrs_all[] =3D { =20 MSR_KVM_ASYNC_PF_EN, MSR_KVM_STEAL_TIME, MSR_KVM_PV_EOI_EN, MSR_KVM_ASYNC_PF_INT, MSR_KVM_ASYNC_PF_ACK, + MSR_KVM_PREEMPT_COUNT, =20 MSR_IA32_TSC_ADJUST, MSR_IA32_TSC_DEADLINE, @@ -3433,6 +3434,25 @@ static void record_steal_time(struct kvm_vcpu *vcpu) mark_page_dirty_in_slot(vcpu->kvm, ghc->memslot, gpa_to_gfn(ghc->gpa)); } =20 +static int kvm_pv_enable_preempt_count(struct kvm_vcpu *vcpu, u64 data) +{ + u64 addr =3D data & ~KVM_MSR_ENABLED; + struct gfn_to_hva_cache *ghc =3D &vcpu->arch.pv_pc.preempt_count_cache; + + vcpu->arch.pv_pc.preempt_count_enabled =3D false; + vcpu->arch.pv_pc.msr_val =3D data; + + if (!(data & KVM_MSR_ENABLED)) + return 0; + + if (kvm_gfn_to_hva_cache_init(vcpu->kvm, ghc, addr, sizeof(int))) + return 1; + + vcpu->arch.pv_pc.preempt_count_enabled =3D true; + + return 0; +} + int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct msr_data *msr_info) { bool pr =3D false; @@ -3652,6 +3672,14 @@ int kvm_set_msr_common(struct kvm_vcpu *vcpu, struct= msr_data *msr_info) vcpu->arch.msr_kvm_poll_control =3D data; break; =20 + case MSR_KVM_PREEMPT_COUNT: + if (!guest_pv_has(vcpu, KVM_FEATURE_PREEMPT_COUNT)) + return 1; + + if (kvm_pv_enable_preempt_count(vcpu, data)) + return 1; + break; + case MSR_IA32_MCG_CTL: case MSR_IA32_MCG_STATUS: case MSR_IA32_MC0_CTL ... MSR_IA32_MCx_CTL(KVM_MAX_MCE_BANKS) - 1: @@ -3992,6 +4020,12 @@ int kvm_get_msr_common(struct kvm_vcpu *vcpu, struct= msr_data *msr_info) =20 msr_info->data =3D vcpu->arch.msr_kvm_poll_control; break; + case MSR_KVM_PREEMPT_COUNT: + if (!guest_pv_has(vcpu, KVM_FEATURE_PREEMPT_COUNT)) + return 1; + + msr_info->data =3D vcpu->arch.pv_pc.msr_val; + break; case MSR_IA32_P5_MC_ADDR: case MSR_IA32_P5_MC_TYPE: case MSR_IA32_MCG_CAP: @@ -11190,6 +11224,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu) =20 vcpu->arch.pending_external_vector =3D -1; vcpu->arch.preempted_in_kernel =3D false; + vcpu->arch.pv_pc.preempt_count_enabled =3D false; =20 #if IS_ENABLED(CONFIG_HYPERV) vcpu->arch.hv_root_tdp =3D INVALID_PAGE; --=20 2.25.1 From nobody Sat Jun 20 00:55:26 2026 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 AF250C433FE for ; Fri, 25 Mar 2022 13:59:47 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1359078AbiCYOBT (ORCPT ); Fri, 25 Mar 2022 10:01:19 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34382 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S243565AbiCYOBE (ORCPT ); Fri, 25 Mar 2022 10:01:04 -0400 Received: from mail-pl1-x632.google.com (mail-pl1-x632.google.com [IPv6:2607:f8b0:4864:20::632]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7AE195F274; Fri, 25 Mar 2022 06:59:30 -0700 (PDT) Received: by mail-pl1-x632.google.com with SMTP id w8so8093717pll.10; Fri, 25 Mar 2022 06:59:30 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=bg55DIKCupebm5f4NqZ1g4YC1+NWU4JZUPE3B9his68=; b=AaqdW4yCybWq9oeRTyuwcMjcgay7UvUWcDwjQS7vzblvc3BBYtlsBPKtpSP97wC0YG KEhu667dydDhzy1pNmdvCsCZDomlYCNwSA1DMezLRj7Pq/65P2q5b75eJkskFCgHupcC mE0mjOYtaU5jCpJakMmHqigh1PXXQIM02iVieoZr375Nqrel6RZyTZ0LyGIP91IH11pN LAmGzWNZp5S2yDgG/Pppsnv5lBqeFS/RGx3EPNoEysuNPNztKCdN4BnhmlkfXg/jQ1bx fJS8u/HEGtb60VDNx4Sjr7lKCxXsQZ7BnWAEOg6RCdqmwKygB6DqUSZDkZc/KsB5fL6h 1F9A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=bg55DIKCupebm5f4NqZ1g4YC1+NWU4JZUPE3B9his68=; b=Bz6TXbYMmYF3htiYKJSuztjQkdkVbQTnfLCstFUYuLRXQmFwM4kzwc5aBTFo/QEkos xkFC/Rr9JsSM0K+3qyquJh0oktXzAUwKa1Jvljox3tcSS0y472BUg/X8FFSBd/VS7UmZ 2uz8EWwaPIkNLPHKmW1hx97hST0otktyfiwm5Sd8+2bLJIdTr6xLDICNtOHRtbSmJvKR BB/MjkNy5syHOADEgjqRBBcCFMRrC8nagRKT547TIIS2z5Mw+yE5oXxwcYavKMw3phng gu4dOBPkU++A4NNhU6YXHoY3tR7IgABMP1MPMFqx0LS2jYLZw4ga4rfk24kJLOhrqqTZ tnLQ== X-Gm-Message-State: AOAM531zKRTFE0oAEFCraTQWXl4aUzlN28ooHDVwQ9ycabaVaP2+yzJT c8hNLdk8A0vcXK6p+i4NvXU+iYlqxms= X-Google-Smtp-Source: ABdhPJz5BQba3hY0HQZbweCxsAW/gWaHhHMfuPXeOppEO/6EwRv27Kh+wlMeGZnn9C+jU7MoD//2qg== X-Received: by 2002:a17:902:d486:b0:154:50c6:a0b5 with SMTP id c6-20020a170902d48600b0015450c6a0b5mr11622659plg.108.1648216769928; Fri, 25 Mar 2022 06:59:29 -0700 (PDT) Received: from localhost.localdomain ([203.205.141.113]) by smtp.googlemail.com with ESMTPSA id lw4-20020a17090b180400b001c7327d09c3sm14470875pjb.53.2022.03.25.06.59.27 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Fri, 25 Mar 2022 06:59:29 -0700 (PDT) From: Wanpeng Li X-Google-Original-From: Wanpeng Li To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org Cc: Paolo Bonzini , Sean Christopherson , Vitaly Kuznetsov , Wanpeng Li , Jim Mattson , Joerg Roedel Subject: [PATCH RESEND 2/5] KVM: X86: Add guest interrupt disable state support Date: Fri, 25 Mar 2022 06:58:26 -0700 Message-Id: <1648216709-44755-3-git-send-email-wanpengli@tencent.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> References: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" From: Wanpeng Li Let's get the information whether or not guests disable interruptions. Signed-off-by: Wanpeng Li --- arch/x86/include/asm/kvm_host.h | 1 + arch/x86/kvm/x86.c | 3 +++ 2 files changed, 4 insertions(+) diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_hos= t.h index 50f011a7445a..8e05cbfa9827 100644 --- a/arch/x86/include/asm/kvm_host.h +++ b/arch/x86/include/asm/kvm_host.h @@ -861,6 +861,7 @@ struct kvm_vcpu_arch { bool preempt_count_enabled; struct gfn_to_hva_cache preempt_count_cache; } pv_pc; + bool irq_disabled; =20 /* * Indicates the guest is trying to write a gfn that contains one or diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index af75e273cb32..425fd7f38fa9 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -4576,6 +4576,7 @@ void kvm_arch_vcpu_load(struct kvm_vcpu *vcpu, int cp= u) =20 static_call(kvm_x86_vcpu_load)(vcpu, cpu); =20 + vcpu->arch.irq_disabled =3D false; /* Save host pkru register if supported */ vcpu->arch.host_pkru =3D read_pkru(); =20 @@ -4668,6 +4669,7 @@ void kvm_arch_vcpu_put(struct kvm_vcpu *vcpu) =20 static_call(kvm_x86_vcpu_put)(vcpu); vcpu->arch.last_host_tsc =3D rdtsc(); + vcpu->arch.irq_disabled =3D !static_call(kvm_x86_get_if_flag)(vcpu); } =20 static int kvm_vcpu_ioctl_get_lapic(struct kvm_vcpu *vcpu, @@ -11225,6 +11227,7 @@ int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu) vcpu->arch.pending_external_vector =3D -1; vcpu->arch.preempted_in_kernel =3D false; vcpu->arch.pv_pc.preempt_count_enabled =3D false; + vcpu->arch.irq_disabled =3D false; =20 #if IS_ENABLED(CONFIG_HYPERV) vcpu->arch.hv_root_tdp =3D INVALID_PAGE; --=20 2.25.1 From nobody Sat Jun 20 00:55:26 2026 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 945EBC433F5 for ; Fri, 25 Mar 2022 13:59:51 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1359091AbiCYOBX (ORCPT ); Fri, 25 Mar 2022 10:01:23 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34462 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1358899AbiCYOBI (ORCPT ); Fri, 25 Mar 2022 10:01:08 -0400 Received: from mail-pj1-x1033.google.com (mail-pj1-x1033.google.com [IPv6:2607:f8b0:4864:20::1033]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 34B88606CA; Fri, 25 Mar 2022 06:59:33 -0700 (PDT) Received: by mail-pj1-x1033.google.com with SMTP id g9-20020a17090ace8900b001c7cce3c0aeso3597732pju.2; Fri, 25 Mar 2022 06:59:33 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=zFHRAV31Ch1tp+2iNpA7ALNISwTJjRRUfUkhPuq13Zo=; b=Cu4xcWiyozPIH5KA4agN/rdwpv5i8VxFgBjItP6j8wz9TTGY2XuJfrGQw+GTMtCW9T /ORB7tR1sdl/Q2caWDHAQLD00WcfLj/Q606h4CIpEdO2KXSfZ9epi8QKsbG4RrC27HCW 3bcl81VVQeIQU2umHw3Ml2DfmM9cksdbbAXFKXmAVCYkOjZUbl8WR/SoKc3xHU0i/42Z qNfFvvi8PJNXfrH9OZYWGhsUPsdWgKVzeXvo+qiDoC1NHkGRcUiMWqov+A8N+zmYPMvN pReCRc0a1pUtGV3hu6WFntKrJ/dt5aL5paVw+ZjZUcCKD5WEcyr7E4+H3R6phBGZCNmb NX4A== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=zFHRAV31Ch1tp+2iNpA7ALNISwTJjRRUfUkhPuq13Zo=; b=YWRaIob8XCMvb8TyPiyEwR/IQPB5SnxSoM/QJ9EgglA5mclhAL20D/hD3ab5iV4Mkt dM13p00ITYXkLmgoTM38h4wsfcwO4x+PzXLL1powV6P7mA+RjIQ5i+frQ/O5li1v9A+S XmA44ZLpILVgUhYGAmTzH7R3Zjkfpw6hWVdbaDNqss1L8JzgaGXXgTpmST2XKnG8BHuD bIYQepicZHtSq8o5uMkZC7K6/KDUPab1Xy4cg2uFR6GmTFjJII2S7Os/IC6O32OGG79q gugVlmciZnk2LHNvFXlAjYiWMCHSx0+O2H9Dx+8+4fYg+njkMuVltBe4VKElur3YlPjM wm+g== X-Gm-Message-State: AOAM5305/nHeDuhksIINso8nrjFJjjEf2KPxHPu1H3IBnNKp98wjwfei CFE86V35fsucKpkrHwaWKYlWWFsVusM= X-Google-Smtp-Source: ABdhPJwLEuM4gsxnbwwEPABdOHYRI14l6NbAdRqXSGJjqYX+3gfuV43mgriZl0bNGYiGc+e7x+iErg== X-Received: by 2002:a17:90b:1809:b0:1c7:2032:4b34 with SMTP id lw9-20020a17090b180900b001c720324b34mr25034960pjb.4.1648216772580; Fri, 25 Mar 2022 06:59:32 -0700 (PDT) Received: from localhost.localdomain ([203.205.141.113]) by smtp.googlemail.com with ESMTPSA id lw4-20020a17090b180400b001c7327d09c3sm14470875pjb.53.2022.03.25.06.59.30 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Fri, 25 Mar 2022 06:59:32 -0700 (PDT) From: Wanpeng Li X-Google-Original-From: Wanpeng Li To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org Cc: Paolo Bonzini , Sean Christopherson , Vitaly Kuznetsov , Wanpeng Li , Jim Mattson , Joerg Roedel Subject: [PATCH RESEND 3/5] KVM: X86: Boost vCPU which is in critical section Date: Fri, 25 Mar 2022 06:58:27 -0700 Message-Id: <1648216709-44755-4-git-send-email-wanpengli@tencent.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> References: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" From: Wanpeng Li The missing semantic gap that occurs when a guest OS is preempted=20 when executing its own critical section, this leads to degradation=20 of application scalability. We try to bridge this semantic gap in=20 some ways, by passing guest preempt_count to the host and checking=20 guest irq disable state, the hypervisor now knows whether guest=20 OSes are running in the critical section, the hypervisor yield-on-spin=20 heuristics can be more smart this time to boost the vCPU candidate=20 who is in the critical section to mitigate this preemption problem,=20 in addition, it is more likely to be a potential lock holder. Testing on 96 HT 2 socket Xeon CLX server, with 96 vCPUs VM 100GB RAM, one VM running benchmark, the other(none-2) VMs running cpu-bound=20 workloads, There is no performance regression for other benchmarks=20 like Unixbench etc. 1VM vanilla optimized improved hackbench -l 50000 28 21.45 30.5% ebizzy -M 12189 12354 1.4% dbench 712 MB/sec 722 MB/sec 1.4% 2VM: vanilla optimized improved hackbench -l 10000 29.4 26 13% ebizzy -M 3834 4033 5% dbench 42.3 MB/sec 44.1 MB/sec 4.3% 3VM: vanilla optimized improved hackbench -l 10000 47 35.46 33% ebizzy -M 3828 4031 5% dbench=20 30.5 MB/sec 31.16 MB/sec 2.3% Signed-off-by: Wanpeng Li --- arch/x86/kvm/x86.c | 22 ++++++++++++++++++++++ include/linux/kvm_host.h | 1 + virt/kvm/kvm_main.c | 7 +++++++ 3 files changed, 30 insertions(+) diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c index 425fd7f38fa9..6b300496bbd0 100644 --- a/arch/x86/kvm/x86.c +++ b/arch/x86/kvm/x86.c @@ -10375,6 +10375,28 @@ static int vcpu_run(struct kvm_vcpu *vcpu) return r; } =20 +static int kvm_vcpu_non_preemptable(struct kvm_vcpu *vcpu) +{ + int count; + + if (!vcpu->arch.pv_pc.preempt_count_enabled) + return 0; + + if (!kvm_read_guest_cached(vcpu->kvm, &vcpu->arch.pv_pc.preempt_count_cac= he, + &count, sizeof(int))) + return (count & ~PREEMPT_NEED_RESCHED); + + return 0; +} + +bool kvm_arch_boost_candidate(struct kvm_vcpu *vcpu) +{ + if (vcpu->arch.irq_disabled || kvm_vcpu_non_preemptable(vcpu)) + return true; + + return false; +} + static inline int complete_emulated_io(struct kvm_vcpu *vcpu) { int r; diff --git a/include/linux/kvm_host.h b/include/linux/kvm_host.h index 252ee4a61b58..9f1a7d9540de 100644 --- a/include/linux/kvm_host.h +++ b/include/linux/kvm_host.h @@ -1428,6 +1428,7 @@ bool kvm_arch_dy_has_pending_interrupt(struct kvm_vcp= u *vcpu); int kvm_arch_post_init_vm(struct kvm *kvm); void kvm_arch_pre_destroy_vm(struct kvm *kvm); int kvm_arch_create_vm_debugfs(struct kvm *kvm); +bool kvm_arch_boost_candidate(struct kvm_vcpu *vcpu); =20 #ifndef __KVM_HAVE_ARCH_VM_ALLOC /* diff --git a/virt/kvm/kvm_main.c b/virt/kvm/kvm_main.c index 9581a24c3d17..ee5a788892e0 100644 --- a/virt/kvm/kvm_main.c +++ b/virt/kvm/kvm_main.c @@ -3545,6 +3545,11 @@ bool __weak kvm_arch_dy_has_pending_interrupt(struct= kvm_vcpu *vcpu) return false; } =20 +bool __weak kvm_arch_boost_candidate(struct kvm_vcpu *vcpu) +{ + return true; +} + void kvm_vcpu_on_spin(struct kvm_vcpu *me, bool yield_to_kernel_mode) { struct kvm *kvm =3D me->kvm; @@ -3580,6 +3585,8 @@ void kvm_vcpu_on_spin(struct kvm_vcpu *me, bool yield= _to_kernel_mode) !kvm_arch_dy_has_pending_interrupt(vcpu) && !kvm_arch_vcpu_in_kernel(vcpu)) continue; + if (!kvm_arch_boost_candidate(vcpu)) + continue; if (!kvm_vcpu_eligible_for_directed_yield(vcpu)) continue; =20 --=20 2.25.1 From nobody Sat Jun 20 00:55:26 2026 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 63BD8C433EF for ; Fri, 25 Mar 2022 13:59:55 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1359252AbiCYOB1 (ORCPT ); Fri, 25 Mar 2022 10:01:27 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34938 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1358938AbiCYOBN (ORCPT ); Fri, 25 Mar 2022 10:01:13 -0400 Received: from mail-pl1-x62c.google.com (mail-pl1-x62c.google.com [IPv6:2607:f8b0:4864:20::62c]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 06A9360A92; Fri, 25 Mar 2022 06:59:35 -0700 (PDT) Received: by mail-pl1-x62c.google.com with SMTP id m18so3214506plx.3; Fri, 25 Mar 2022 06:59:35 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=4jbCwzr0mct8OKs/Lgqse4ZkWggPrV/4VXWHupxqHk0=; b=nT728+6kR20X1SoIl0V5jnB6v7lr0+IpMj+2Vafc9rc9beExEQmjLTm2p2TOZLz2rN VfynJ8uduC92IgQRDDXKz/VEJvV4l2XdlT1yKY7wOAjqJQcsXG7iuyruqR/wOpQGAsWo pjPvTowRroLFlAwJab3PdPh3/3TplhmVsRinWI3HXC2hBRNC0SVDAc47xfj17McKbNPW bXB3QzVuCLiyXA92OtOv2/lh7VxfeEDuVGshLX8LhatAr+lEyXIByScvhSnMFYduKAr1 LvsMoIO5xvdO8X0BTnDrDXMrmaS76+bW51+kwbvFDCCQkvEDbKng68ovq4G5ueo8VW4c wZLw== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=4jbCwzr0mct8OKs/Lgqse4ZkWggPrV/4VXWHupxqHk0=; b=jj1j6yKEs+dk5k8FXd9ddVXbz9DfaWWqGV+5ggJOXig+2KIRnohCxRHgxn5Y4pdqc6 dJZYjKo28dqJZkTpImpyGl0m8u0mSAy/S14CNSocvFj2O+iOPdVbth/MUKJFQke2G4IT WpogTS0HeJbwIpNXRPFUjKY0DIQDPtr5xnFmQSLiv1OSWeiRsMWjlgeKSYzx84Qq/gbQ +s0FxDBa/4cpqZQqll0TOx9XM7iLq27U0zIcM9a3aaCg76+PjF9fr1YbbuR3HnKib/nZ k3jOEkaaBa10ZvZF5sEkiXQ+EZAp+0z+nJy/oyKnsy65H8GQsIte74/fpUyZqBZvNEC7 R9/w== X-Gm-Message-State: AOAM533hMVg9f024QL3ppAJWcydy4pleHivKxZvMFxs6z1HLXxxM6jqK P4N3wJrqpMgdKW8eLS0IpR2rwA7Uml8= X-Google-Smtp-Source: ABdhPJxjmkgO4hGCqv5qcydxjfqbeQere/uio1LGQxcMv22iZukZrTDZQvjrF16csQcLHjItIV2f8w== X-Received: by 2002:a17:90b:1d08:b0:1c7:5523:6a1f with SMTP id on8-20020a17090b1d0800b001c755236a1fmr25263357pjb.225.1648216775134; Fri, 25 Mar 2022 06:59:35 -0700 (PDT) Received: from localhost.localdomain ([203.205.141.113]) by smtp.googlemail.com with ESMTPSA id lw4-20020a17090b180400b001c7327d09c3sm14470875pjb.53.2022.03.25.06.59.32 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Fri, 25 Mar 2022 06:59:34 -0700 (PDT) From: Wanpeng Li X-Google-Original-From: Wanpeng Li To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org Cc: Paolo Bonzini , Sean Christopherson , Vitaly Kuznetsov , Wanpeng Li , Jim Mattson , Joerg Roedel Subject: [PATCH RESEND 4/5] x86/kvm: Add MSR_KVM_PREEMPT_COUNT guest support Date: Fri, 25 Mar 2022 06:58:28 -0700 Message-Id: <1648216709-44755-5-git-send-email-wanpengli@tencent.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> References: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" From: Wanpeng Li The x86 guest passes the per-cpu preempt_count value to the hypervisor, so the hypervisor knows whether the guest is running in the critical=20 section. Signed-off-by: Wanpeng Li --- arch/x86/kernel/kvm.c | 10 ++++++++++ 1 file changed, 10 insertions(+) diff --git a/arch/x86/kernel/kvm.c b/arch/x86/kernel/kvm.c index 21933095a10e..e389fa4393ae 100644 --- a/arch/x86/kernel/kvm.c +++ b/arch/x86/kernel/kvm.c @@ -366,6 +366,14 @@ static void kvm_guest_cpu_init(void) =20 if (has_steal_clock) kvm_register_steal_time(); + + if (kvm_para_has_feature(KVM_FEATURE_PREEMPT_COUNT)) { + u64 pa =3D slow_virt_to_phys(this_cpu_ptr(&__preempt_count)) + | KVM_MSR_ENABLED; + wrmsrl(MSR_KVM_PREEMPT_COUNT, pa); + + pr_debug("setup pv preempt_count: cpu %d\n", smp_processor_id()); + } } =20 static void kvm_pv_disable_apf(void) @@ -442,6 +450,8 @@ static void kvm_guest_cpu_offline(bool shutdown) if (!shutdown) apf_task_wake_all(); kvmclock_disable(); + if (kvm_para_has_feature(KVM_FEATURE_PREEMPT_COUNT)) + wrmsrl(MSR_KVM_PREEMPT_COUNT, 0); } =20 static int kvm_cpu_online(unsigned int cpu) --=20 2.25.1 From nobody Sat Jun 20 00:55:26 2026 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 E8FB4C433F5 for ; Fri, 25 Mar 2022 14:00:02 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1358886AbiCYOBf (ORCPT ); Fri, 25 Mar 2022 10:01:35 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:34948 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1358940AbiCYOBN (ORCPT ); Fri, 25 Mar 2022 10:01:13 -0400 Received: from mail-pl1-x636.google.com (mail-pl1-x636.google.com [IPv6:2607:f8b0:4864:20::636]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 7D3ED5F253; Fri, 25 Mar 2022 06:59:38 -0700 (PDT) Received: by mail-pl1-x636.google.com with SMTP id j13so8095732plj.8; Fri, 25 Mar 2022 06:59:38 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; h=from:to:cc:subject:date:message-id:in-reply-to:references; bh=EV0wmNwRVxX8NQcoop+BEUS5js89FYD4nmxPHK49pWQ=; b=MaTdW3To90ijPhW/yZNfIQWH8ozmuI4LzhKZZvBDJlsmUM8Q8KOcBH9ehtDzmtC3YW eBh8n4j26HCJ6UIWcx/uCWY44xhL2PQ0sfAA1+jRChFcFazSPdkTIVADj9Dv0lsOLPG8 UI6O41pQBpGMsUUKkhL0jM5O/hLzOmYIEJMDkkVr9krBTgPBzhVDEh6+yZt7JQooPEVz /pAkUQoqYT62KUT1s8stfPcwAp1cuKUBU4np52nZnrjZQZwV4kFEl9zVyssIJ+qwZFc+ sgY7F2ftMg5zXJ7ohNd1pPwnjmwomWxeC8VnAE0powpbiCvktOMAr80vds0yyEliKNyN 3zww== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; h=x-gm-message-state:from:to:cc:subject:date:message-id:in-reply-to :references; bh=EV0wmNwRVxX8NQcoop+BEUS5js89FYD4nmxPHK49pWQ=; b=YELN5Lrk0E0oFZCpvqtW5YYzQesUqFNMWJwSBo8q2O+JfBgWeLOmPVJWTVU34EQK7A 6AfDYd1nZo6wOIZHNyrO+CkycXq4xM/vep5R99obENEN9fIZE9AyvKN+mkcmV5sq+mYL DasVZAjawP5iWN8kNlB0xXW/Bm5fl0fUya2vB7Gdh0P1ekBt8aKPmoF4++rG31bcr7z6 5gRAbPbpU/GxWxkYh2CDYuEkEJrrHIk/hnZr8PTDlYovaD6xIKxUcDXU20e1mCBkPby1 2Ci/GY4+57P/Ze4zkGg5jcCyed+fVICsQIbtmiV9kvDHPPkD9OocWpFzi4Jwug/QQiXE VHmA== X-Gm-Message-State: AOAM530lb9ID88TBVF4W42yiIU1Xk27QaYX6O0wH2xqqqQBDV8wUnf0x Xh8u5K6UAlnFOBCY2On2e8Ga7rqAQn0= X-Google-Smtp-Source: ABdhPJzz7K+/5uo4/3em0DsU11ggYGjYid3uyziE4s7FG3CHnGx59o2D7meiWG5Ps8LDY3KfTEYf5g== X-Received: by 2002:a17:90b:1e43:b0:1c7:46bf:ba29 with SMTP id pi3-20020a17090b1e4300b001c746bfba29mr12861850pjb.100.1648216777881; Fri, 25 Mar 2022 06:59:37 -0700 (PDT) Received: from localhost.localdomain ([203.205.141.113]) by smtp.googlemail.com with ESMTPSA id lw4-20020a17090b180400b001c7327d09c3sm14470875pjb.53.2022.03.25.06.59.35 (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Fri, 25 Mar 2022 06:59:37 -0700 (PDT) From: Wanpeng Li X-Google-Original-From: Wanpeng Li To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org Cc: Paolo Bonzini , Sean Christopherson , Vitaly Kuznetsov , Wanpeng Li , Jim Mattson , Joerg Roedel Subject: [PATCH RESEND 5/5] KVM: X86: Expose PREEMT_COUNT CPUID feature bit to guest Date: Fri, 25 Mar 2022 06:58:29 -0700 Message-Id: <1648216709-44755-6-git-send-email-wanpengli@tencent.com> X-Mailer: git-send-email 2.7.4 In-Reply-To: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> References: <1648216709-44755-1-git-send-email-wanpengli@tencent.com> Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Content-Type: text/plain; charset="utf-8" From: Wanpeng Li Expose the PREEMPT_COUNT feature bit to the guest, the guest can check this feature bit before using MSR_KVM_PREEMPT_COUNT. Signed-off-by: Wanpeng Li --- Documentation/virt/kvm/cpuid.rst | 3 +++ arch/x86/kvm/cpuid.c | 3 ++- 2 files changed, 5 insertions(+), 1 deletion(-) diff --git a/Documentation/virt/kvm/cpuid.rst b/Documentation/virt/kvm/cpui= d.rst index bda3e3e737d7..c45158af98a7 100644 --- a/Documentation/virt/kvm/cpuid.rst +++ b/Documentation/virt/kvm/cpuid.rst @@ -103,6 +103,9 @@ KVM_FEATURE_HC_MAP_GPA_RANGE 16 guest ch= ecks this feature bit bef KVM_FEATURE_MIGRATION_CONTROL 17 guest checks this feature b= it before using MSR_KVM_MIGRATION_CON= TROL =20 +KVM_FEATURE_PREEMPT_COUNT 18 guest checks this feature b= it before + using MSR_KVM_PREEMPT_COUNT + KVM_FEATURE_CLOCKSOURCE_STABLE_BIT 24 host will warn if no guest-= side per-cpu warps are expected = in kvmclock diff --git a/arch/x86/kvm/cpuid.c b/arch/x86/kvm/cpuid.c index 58b0b4e0263c..4785f5a63d8d 100644 --- a/arch/x86/kvm/cpuid.c +++ b/arch/x86/kvm/cpuid.c @@ -1071,7 +1071,8 @@ static inline int __do_cpuid_func(struct kvm_cpuid_ar= ray *array, u32 function) (1 << KVM_FEATURE_PV_SEND_IPI) | (1 << KVM_FEATURE_POLL_CONTROL) | (1 << KVM_FEATURE_PV_SCHED_YIELD) | - (1 << KVM_FEATURE_ASYNC_PF_INT); + (1 << KVM_FEATURE_ASYNC_PF_INT) | + (1 << KVM_FEATURE_PREEMPT_COUNT); =20 if (sched_info_on()) entry->eax |=3D (1 << KVM_FEATURE_STEAL_TIME); --=20 2.25.1