From patchwork Tue Feb 16 02:13:07 2021 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Isaku Yamahata X-Patchwork-Id: 12089479 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-13.9 required=3.0 tests=BAYES_00, HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_CR_TRAILER,INCLUDES_PATCH, MAILING_LIST_MULTI,SPF_HELO_NONE,SPF_PASS,UNWANTED_LANGUAGE_BODY, URIBL_BLOCKED,USER_AGENT_GIT autolearn=unavailable autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 1413CC433E6 for ; Tue, 16 Feb 2021 02:25:09 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 9836564DE8 for ; Tue, 16 Feb 2021 02:25:08 +0000 (UTC) DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 9836564DE8 Authentication-Results: mail.kernel.org; dmarc=fail (p=none dis=none) header.from=intel.com Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Received: from localhost ([::1]:58526 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1lBq31-0001eq-MT for qemu-devel@archiver.kernel.org; Mon, 15 Feb 2021 21:25:07 -0500 Received: from eggs.gnu.org ([2001:470:142:3::10]:46118) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1lBpta-0006PF-47 for qemu-devel@nongnu.org; Mon, 15 Feb 2021 21:15:22 -0500 Received: from mga17.intel.com ([192.55.52.151]:25640) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1lBptQ-0001b1-Tl for qemu-devel@nongnu.org; Mon, 15 Feb 2021 21:15:17 -0500 IronPort-SDR: WkzmCC4TeRIsBhVVo2t07Dm91a4SlVqsnwss+WPHJe2KFvhe45qRE3T9NRm7WWBKti1+Ej39dY Ga9WW6Q3KANQ== X-IronPort-AV: E=McAfee;i="6000,8403,9896"; a="162558763" X-IronPort-AV: E=Sophos;i="5.81,182,1610438400"; d="scan'208";a="162558763" Received: from fmsmga005.fm.intel.com ([10.253.24.32]) by fmsmga107.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 15 Feb 2021 18:14:52 -0800 IronPort-SDR: UmonXQH++N8o4hbOsGrnbLSSo3CqOwTGYPfHNzkWv0Lvzw6w9homoIgVADRsuj3qfKEyBMnhNM AujP0HDQAVMA== X-IronPort-AV: E=Sophos;i="5.81,182,1610438400"; d="scan'208";a="591705415" Received: from ls.sc.intel.com (HELO localhost) ([143.183.96.54]) by fmsmga005-auth.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 15 Feb 2021 18:14:51 -0800 From: Isaku Yamahata To: qemu-devel@nongnu.org, pbonzini@redhat.com, alistair@alistair23.me, ehabkost@redhat.com, marcel.apfelbaum@gmail.com, mst@redhat.com, cohuck@redhat.com, mtosatti@redhat.com, xiaoyao.li@intel.com, seanjc@google.com Subject: [RFC PATCH 11/23] hw/i386: Initialize TDX via KVM ioctl() when kvm_type is TDX Date: Mon, 15 Feb 2021 18:13:07 -0800 Message-Id: X-Mailer: git-send-email 2.17.1 In-Reply-To: References: In-Reply-To: References: Received-SPF: pass client-ip=192.55.52.151; envelope-from=isaku.yamahata@intel.com; helo=mga17.intel.com X-Spam_score_int: -41 X-Spam_score: -4.2 X-Spam_bar: ---- X-Spam_report: (-4.2 / 5.0 requ) BAYES_00=-1.9, RCVD_IN_DNSWL_MED=-2.3, SPF_HELO_NONE=0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: isaku.yamahata@intel.com, Sean Christopherson , isaku.yamahata@gmail.com, kvm@vger.kernel.org Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" From: Xiaoyao Li Introduce tdx_ioctl() to invoke TDX specific sub-ioctls of KVM_MEMORY_ENCRYPT_OP. Use tdx_ioctl() to invoke KVM_TDX_INIT, by way of tdx_init(), during kvm_arch_init(). KVM_TDX_INIT configures global TD state, e.g. the canonical CPUID config, and must be executed prior to creating vCPUs. Note, this doesn't address the fact that Qemu may change the CPUID configuration when creating vCPUs, i.e. punts on refactoring Qemu to provide a stable CPUID config prior to kvm_arch_init(). Explicitly set subleaf index and flags when adding CPUID Set the index and flags when adding a CPUID entry to avoid propagating stale state from a removed entry, e.g. when the CPUID 0x4 loop bails, it can leave non-zero index and flags in the array. Signed-off-by: Xiaoyao Li Co-developed-by: Sean Christopherson Signed-off-by: Sean Christopherson --- accel/kvm/kvm-all.c | 2 + include/sysemu/tdx.h | 2 + target/i386/kvm/tdx-stub.c | 4 ++ target/i386/kvm/tdx.c | 128 +++++++++++++++++++++++++++++++++++++ target/i386/kvm/tdx.h | 24 +++++++ 5 files changed, 160 insertions(+) create mode 100644 target/i386/kvm/tdx.h diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c index 351c25a5cb..cd13b8c94d 100644 --- a/accel/kvm/kvm-all.c +++ b/accel/kvm/kvm-all.c @@ -40,6 +40,7 @@ #include "trace.h" #include "hw/irq.h" #include "sysemu/sev.h" +#include "sysemu/tdx.h" #include "qapi/visitor.h" #include "qapi/qapi-types-common.h" #include "qapi/qapi-visit-common.h" @@ -428,6 +429,7 @@ int kvm_init_vcpu(CPUState *cpu, Error **errp) trace_kvm_init_vcpu(cpu->cpu_index, kvm_arch_vcpu_id(cpu)); + tdx_pre_create_vcpu(cpu); ret = kvm_get_vcpu(s, kvm_arch_vcpu_id(cpu)); if (ret < 0) { error_setg_errno(errp, -ret, "kvm_init_vcpu: kvm_get_vcpu failed (%lu)", diff --git a/include/sysemu/tdx.h b/include/sysemu/tdx.h index 60ebded851..36a901e723 100644 --- a/include/sysemu/tdx.h +++ b/include/sysemu/tdx.h @@ -7,4 +7,6 @@ bool kvm_has_tdx(KVMState *s); #endif +void tdx_pre_create_vcpu(CPUState *cpu); + #endif diff --git a/target/i386/kvm/tdx-stub.c b/target/i386/kvm/tdx-stub.c index e1eb09cae1..93d5913c89 100644 --- a/target/i386/kvm/tdx-stub.c +++ b/target/i386/kvm/tdx-stub.c @@ -8,3 +8,7 @@ bool kvm_has_tdx(KVMState *s) return false; } #endif + +void tdx_pre_create_vcpu(CPUState *cpu) +{ +} diff --git a/target/i386/kvm/tdx.c b/target/i386/kvm/tdx.c index e62a570f75..00eda80725 100644 --- a/target/i386/kvm/tdx.c +++ b/target/i386/kvm/tdx.c @@ -14,8 +14,10 @@ #include "qemu/osdep.h" #include +#include #include "cpu.h" +#include "kvm_i386.h" #include "hw/boards.h" #include "qapi/error.h" #include "qom/object_interfaces.h" @@ -23,8 +25,134 @@ #include "sysemu/kvm.h" #include "sysemu/kvm_int.h" #include "sysemu/tdx.h" +#include "tdx.h" + +#define TDX1_TD_ATTRIBUTE_DEBUG BIT_ULL(0) +#define TDX1_TD_ATTRIBUTE_PERFMON BIT_ULL(63) bool kvm_has_tdx(KVMState *s) { return !!(kvm_check_extension(s, KVM_CAP_VM_TYPES) & BIT(KVM_X86_TDX_VM)); } + +static void __tdx_ioctl(int ioctl_no, const char *ioctl_name, + __u32 metadata, void *data) +{ + struct kvm_tdx_cmd tdx_cmd; + int r; + + memset(&tdx_cmd, 0x0, sizeof(tdx_cmd)); + + tdx_cmd.id = ioctl_no; + tdx_cmd.metadata = metadata; + tdx_cmd.data = (__u64)(unsigned long)data; + + r = kvm_vm_ioctl(kvm_state, KVM_MEMORY_ENCRYPT_OP, &tdx_cmd); + if (r) { + error_report("%s failed: %s", ioctl_name, strerror(-r)); + exit(1); + } +} +#define tdx_ioctl(ioctl_no, metadata, data) \ + __tdx_ioctl(ioctl_no, stringify(ioctl_no), metadata, data) + +void tdx_pre_create_vcpu(CPUState *cpu) +{ + struct { + struct kvm_cpuid2 cpuid; + struct kvm_cpuid_entry2 entries[KVM_MAX_CPUID_ENTRIES]; + } cpuid_data; + + /* + * The kernel defines these structs with padding fields so there + * should be no extra padding in our cpuid_data struct. + */ + QEMU_BUILD_BUG_ON(sizeof(cpuid_data) != + sizeof(struct kvm_cpuid2) + + sizeof(struct kvm_cpuid_entry2) * + KVM_MAX_CPUID_ENTRIES); + + MachineState *ms = MACHINE(qdev_get_machine()); + X86CPU *x86cpu = X86_CPU(cpu); + CPUX86State *env = &x86cpu->env; + TdxGuest *tdx = (TdxGuest *)object_dynamic_cast(OBJECT(ms->cgs), + TYPE_TDX_GUEST); + struct kvm_tdx_init_vm init_vm; + + if (!tdx) { + return; + } + + /* HACK: Remove MPX support, which is not allowed by TDX. */ + env->features[FEAT_XSAVE_COMP_LO] &= ~(XSTATE_BNDREGS_MASK | + XSTATE_BNDCSR_MASK); + + if (!(env->features[FEAT_1_ECX] & CPUID_EXT_XSAVE)) { + error_report("TDX VM must support XSAVE features"); + exit(1); + } + + qemu_mutex_lock(&tdx->lock); + if (tdx->initialized) { + goto out; + } + tdx->initialized = true; + + memset(&cpuid_data, 0, sizeof(cpuid_data)); + + cpuid_data.cpuid.nent = kvm_x86_arch_cpuid(env, cpuid_data.entries, 0); + cpuid_data.cpuid.padding = 0; + + init_vm.max_vcpus = ms->smp.cpus; + init_vm.attributes = 0; + init_vm.attributes |= tdx->debug ? TDX1_TD_ATTRIBUTE_DEBUG : 0; + init_vm.attributes |= x86cpu->enable_pmu ? TDX1_TD_ATTRIBUTE_PERFMON : 0; + + init_vm.cpuid = (__u64)(&cpuid_data); + tdx_ioctl(KVM_TDX_INIT_VM, 0, &init_vm); +out: + qemu_mutex_unlock(&tdx->lock); +} + +static bool tdx_guest_get_debug(Object *obj, Error **errp) +{ + TdxGuest *tdx = TDX_GUEST(obj); + + return tdx->debug; +} + +static void tdx_guest_set_debug(Object *obj, bool value, Error **errp) +{ + TdxGuest *tdx = TDX_GUEST(obj); + + tdx->debug = value; +} + +/* tdx guest */ +OBJECT_DEFINE_TYPE_WITH_INTERFACES(TdxGuest, + tdx_guest, + TDX_GUEST, + CONFIDENTIAL_GUEST_SUPPORT, + { TYPE_USER_CREATABLE }, + { NULL }) + +static void tdx_guest_init(Object *obj) +{ + TdxGuest *tdx = TDX_GUEST(obj); + + qemu_mutex_init(&tdx->lock); + tdx->debug = false; + object_property_add_bool(obj, "debug", tdx_guest_get_debug, + tdx_guest_set_debug); + + /* TODO: move this after fully TD initialized */ + tdx->parent_obj.ready = true; +} + +static void tdx_guest_finalize(Object *obj) +{ +} + +static void tdx_guest_class_init(ObjectClass *oc, void *data) +{ +} diff --git a/target/i386/kvm/tdx.h b/target/i386/kvm/tdx.h new file mode 100644 index 0000000000..6ad6c9a313 --- /dev/null +++ b/target/i386/kvm/tdx.h @@ -0,0 +1,24 @@ +#ifndef QEMU_I386_TDX_H +#define QEMU_I386_TDX_H + +#include "qom/object.h" +#include "exec/confidential-guest-support.h" + +#define TYPE_TDX_GUEST "tdx-guest" +#define TDX_GUEST(obj) \ + OBJECT_CHECK(TdxGuest, (obj), TYPE_TDX_GUEST) + +typedef struct TdxGuestClass { + ConfidentialGuestSupportClass parent_class; +} TdxGuestClass; + +typedef struct TdxGuest { + ConfidentialGuestSupport parent_obj; + + QemuMutex lock; + + bool initialized; + bool debug; +} TdxGuest; + +#endif