From nobody Fri Dec 19 14:10:42 2025 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 36502C83F29 for ; Thu, 31 Aug 2023 08:30:53 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S245591AbjHaIax (ORCPT ); Thu, 31 Aug 2023 04:30:53 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:53880 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S245298AbjHaIai (ORCPT ); Thu, 31 Aug 2023 04:30:38 -0400 Received: from mail.loongson.cn (mail.loongson.cn [114.242.206.163]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id 1DACD19A; Thu, 31 Aug 2023 01:30:33 -0700 (PDT) Received: from loongson.cn (unknown [10.2.5.185]) by gateway (Coremail) with SMTP id _____8Cx5_GlT_BkIl4dAA--.60020S3; Thu, 31 Aug 2023 16:30:29 +0800 (CST) Received: from localhost.localdomain (unknown [10.2.5.185]) by localhost.localdomain (Coremail) with SMTP id AQAAf8Ax3c6gT_BkOPVnAA--.55892S9; Thu, 31 Aug 2023 16:30:28 +0800 (CST) From: Tianrui Zhao To: linux-kernel@vger.kernel.org, kvm@vger.kernel.org Cc: Paolo Bonzini , Huacai Chen , WANG Xuerui , Greg Kroah-Hartman , loongarch@lists.linux.dev, Jens Axboe , Mark Brown , Alex Deucher , Oliver Upton , maobibo@loongson.cn, Xi Ruoyao , zhaotianrui@loongson.cn Subject: [PATCH v20 07/30] LoongArch: KVM: Implement vcpu run interface Date: Thu, 31 Aug 2023 16:29:57 +0800 Message-Id: <20230831083020.2187109-8-zhaotianrui@loongson.cn> X-Mailer: git-send-email 2.39.1 In-Reply-To: <20230831083020.2187109-1-zhaotianrui@loongson.cn> References: <20230831083020.2187109-1-zhaotianrui@loongson.cn> MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable X-CM-TRANSID: AQAAf8Ax3c6gT_BkOPVnAA--.55892S9 X-CM-SenderInfo: p2kd03xldq233l6o00pqjv00gofq/ X-Coremail-Antispam: 1Uk129KBjDUn29KB7ZKAUJUUUUU529EdanIXcx71UUUUU7KY7 ZEXasCq-sGcSsGvfJ3UbIjqfuFe4nvWSU5nxnvy29KBjDU0xBIdaVrnUUvcSsGvfC2Kfnx nUUI43ZEXa7xR_UUUUUUUUU== Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" Implement vcpu run interface, handling mmio, iocsr reading fault and deliver interrupt, lose fpu before vcpu enter guest. Reviewed-by: Bibo Mao Signed-off-by: Tianrui Zhao --- arch/loongarch/kvm/vcpu.c | 130 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 130 insertions(+) diff --git a/arch/loongarch/kvm/vcpu.c b/arch/loongarch/kvm/vcpu.c index 545b18cd1c..83f2988ea6 100644 --- a/arch/loongarch/kvm/vcpu.c +++ b/arch/loongarch/kvm/vcpu.c @@ -18,6 +18,91 @@ int kvm_arch_vcpu_precreate(struct kvm *kvm, unsigned in= t id) return 0; } =20 +/* + * _kvm_check_requests - check and handle pending vCPU requests + * + * Return: RESUME_GUEST if we should enter the guest + * RESUME_HOST if we should exit to userspace + */ +static int _kvm_check_requests(struct kvm_vcpu *vcpu) +{ + if (!kvm_request_pending(vcpu)) + return RESUME_GUEST; + + if (kvm_check_request(KVM_REQ_TLB_FLUSH, vcpu)) + /* Drop vpid for this vCPU */ + vcpu->arch.vpid =3D 0; + + if (kvm_dirty_ring_check_request(vcpu)) + return RESUME_HOST; + + return RESUME_GUEST; +} + +/* + * Check and handle pending signal and vCPU requests etc + * Run with irq enabled and preempt enabled + * + * Return: RESUME_GUEST if we should enter the guest + * RESUME_HOST if we should exit to userspace + * < 0 if we should exit to userspace, where the return value + * indicates an error + */ +static int kvm_enter_guest_check(struct kvm_vcpu *vcpu) +{ + int ret; + + /* + * Check conditions before entering the guest + */ + ret =3D xfer_to_guest_mode_handle_work(vcpu); + if (ret < 0) + return ret; + + ret =3D _kvm_check_requests(vcpu); + return ret; +} + +/* + * called with irq enabled + * + * Return: RESUME_GUEST if we should enter the guest, and irq disabled + * Others if we should exit to userspace + */ +static int kvm_pre_enter_guest(struct kvm_vcpu *vcpu) +{ + int ret; + + do { + ret =3D kvm_enter_guest_check(vcpu); + if (ret !=3D RESUME_GUEST) + break; + + /* + * handle vcpu timer, interrupts, check requests and + * check vmid before vcpu enter guest + */ + local_irq_disable(); + kvm_acquire_timer(vcpu); + _kvm_deliver_intr(vcpu); + /* make sure the vcpu mode has been written */ + smp_store_mb(vcpu->mode, IN_GUEST_MODE); + _kvm_check_vmid(vcpu); + vcpu->arch.host_eentry =3D csr_read64(LOONGARCH_CSR_EENTRY); + /* clear KVM_LARCH_CSR as csr will change when enter guest */ + vcpu->arch.aux_inuse &=3D ~KVM_LARCH_CSR; + + if (kvm_request_pending(vcpu) || xfer_to_guest_mode_work_pending()) { + /* make sure the vcpu mode has been written */ + smp_store_mb(vcpu->mode, OUTSIDE_GUEST_MODE); + local_irq_enable(); + ret =3D -EAGAIN; + } + } while (ret !=3D RESUME_GUEST); + + return ret; +} + int kvm_arch_vcpu_create(struct kvm_vcpu *vcpu) { unsigned long timer_hz; @@ -85,3 +170,48 @@ void kvm_arch_vcpu_destroy(struct kvm_vcpu *vcpu) context->last_vcpu =3D NULL; } } + +int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu) +{ + int r =3D -EINTR; + struct kvm_run *run =3D vcpu->run; + + if (vcpu->mmio_needed) { + if (!vcpu->mmio_is_write) + _kvm_complete_mmio_read(vcpu, run); + vcpu->mmio_needed =3D 0; + } + + if (run->exit_reason =3D=3D KVM_EXIT_LOONGARCH_IOCSR) { + if (!run->iocsr_io.is_write) + _kvm_complete_iocsr_read(vcpu, run); + } + + /* clear exit_reason */ + run->exit_reason =3D KVM_EXIT_UNKNOWN; + if (run->immediate_exit) + return r; + + lose_fpu(1); + vcpu_load(vcpu); + kvm_sigset_activate(vcpu); + r =3D kvm_pre_enter_guest(vcpu); + if (r !=3D RESUME_GUEST) + goto out; + + guest_timing_enter_irqoff(); + guest_state_enter_irqoff(); + trace_kvm_enter(vcpu); + r =3D kvm_loongarch_ops->enter_guest(run, vcpu); + + trace_kvm_out(vcpu); + /* + * guest exit is already recorded at _kvm_handle_exit + * return val must not be RESUME_GUEST + */ + local_irq_enable(); +out: + kvm_sigset_deactivate(vcpu); + vcpu_put(vcpu); + return r; +} --=20 2.27.0