Received: by 2002:a6b:500f:0:0:0:0:0 with SMTP id e15csp128324iob; Tue, 17 May 2022 21:10:40 -0700 (PDT) X-Google-Smtp-Source: ABdhPJzyNfN8//WN4IjDlF2UvyBBC8/njuIudNk/1zeKlzUk2GW2BbGhunwqqIULqwr/PR9IqTIb X-Received: by 2002:a17:90a:c002:b0:1d9:250a:73c8 with SMTP id p2-20020a17090ac00200b001d9250a73c8mr28214280pjt.133.1652847039800; Tue, 17 May 2022 21:10:39 -0700 (PDT) ARC-Seal: i=1; a=rsa-sha256; t=1652847039; cv=none; d=google.com; s=arc-20160816; b=KcyfYyFgIyPBqz8r+6ObD4sxEvqWYPwFHvQ1CWZl7nb1INT5lrAmwbO6uDjFNO5+6j RCX2Q3NdyiDQKhJq5lMhr4cEFQUZx/GtMkr3mhoZGqcuCMOsufc1bbtMIf2G8bRhGCQ8 6/gVydBYXB4vdzT94016kkSO/86/ibQiMgATrqr0dkgctrNgf0kFGKAnC6ft44vA/m/x Go1IGW6oK5/tC05jnYRnsHmq7Rfn1LpNJ+vexvd/1gxNyE8C1iwXWrEurJpUI+ygwSZW kpqRtrpt8D3Bd/rHhMVU1mr53SgDwKtrvHTMMlc6qR5eI2ngtgoN0RgbfWHrLxTcniDe OtmQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d=google.com; s=arc-20160816; h=list-id:precedence:content-transfer-encoding:user-agent:date :mime-version:references:in-reply-to:cc:to:from:subject:message-id :dkim-signature; bh=oaG4XsDjsKIUJ+dwnlpwAuMILYCC53DkZgFKCWJNmZg=; b=TDX7Gl87Bk91pI7qzyVyJVNmL1IY5XkXdw0bc8zUIn86p4h0ZA6hJJIvcZ/UI59h10 Ze3Wp8/HZFok7woqMJHPFVg2aqI2C4N2EgSh2lulAGFPnPKyKgH4PiwYIz26qxvjg8Gt hnzzQLgIf7QJvinA8lJWp+FvFFTLgmwTxq8fgPFKnECgkpBW6DF5uGOLwlj9R2DzZUkf Wz1V1PMYezbXb6PH/nZQADhCk7dKg+Gs30VVZGbMdgG+/0v2lD/n7QfW0s8/sHg2L80t 4NsWjAWufzYgjSkKABMYeEejSBv3aKib6QIIKmo69nxw/O14zh6ssqWWLe7mCqDoJhQW 1rrw== ARC-Authentication-Results: i=1; mx.google.com; dkim=pass header.i=@intel.com header.s=Intel header.b=nu7R6n0L; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=intel.com Return-Path: Received: from lindbergh.monkeyblade.net (lindbergh.monkeyblade.net. [2620:137:e000::1:18]) by mx.google.com with ESMTPS id c23-20020a63da17000000b003c1bb53b4b1si1177034pgh.248.2022.05.17.21.10.39 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Tue, 17 May 2022 21:10:39 -0700 (PDT) Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) client-ip=2620:137:e000::1:18; Authentication-Results: mx.google.com; dkim=pass header.i=@intel.com header.s=Intel header.b=nu7R6n0L; spf=pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:18 as permitted sender) smtp.mailfrom=linux-kernel-owner@vger.kernel.org; dmarc=pass (p=NONE sp=NONE dis=NONE) header.from=intel.com Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by lindbergh.monkeyblade.net (Postfix) with ESMTP id C4AB012D1D; Tue, 17 May 2022 20:41:24 -0700 (PDT) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S238500AbiEQCz3 (ORCPT + 99 others); Mon, 16 May 2022 22:55:29 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:46754 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S234471AbiEQCzZ (ORCPT ); Mon, 16 May 2022 22:55:25 -0400 Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 2657C457AC for ; Mon, 16 May 2022 19:55:23 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1652756123; x=1684292123; h=message-id:subject:from:to:cc:in-reply-to:references: mime-version:date:content-transfer-encoding; bh=5RyhjO8+v5haZ+WOtijCUg3ji41pHOSJp3sI5CcUB+E=; b=nu7R6n0LdgC7lxvO8u8dVZCuKHb8kvBDka8MjNqJkbjJE8kreJDRWa+9 JAgrFXIlUU2MD1DhXNZ+MtcGuc5sJmSCsZipkwxCfvEPf/3LquBSbJfLm 9xQ6uER1dT3gn34vHLd1aP+0rso0Mve1fgDjECnRyObRPjw2TC4Gx7KRM Dj3ffI6zkbIQ1WkgVCH9jVB7lnW9nBVoGd9EZZAwiMXDSOeeGyCz+N+tL NHkK6mnHAXUigquCV+nL43w8O38Ap0umz3aUVOS0gxjAe2CWdBGYic5zV UHQBYYp6tCV9JjC03Xm7Uc1gzRAkodqHxEKciOPc4yDZhMmF9q93LJeaL w==; X-IronPort-AV: E=McAfee;i="6400,9594,10349"; a="270728835" X-IronPort-AV: E=Sophos;i="5.91,231,1647327600"; d="scan'208";a="270728835" Received: from orsmga007.jf.intel.com ([10.7.209.58]) by orsmga102.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 16 May 2022 19:55:22 -0700 X-IronPort-AV: E=Sophos;i="5.91,231,1647327600"; d="scan'208";a="568646591" Received: from sandrocx-mobl1.amr.corp.intel.com (HELO khuang2-desk.gar.corp.intel.com) ([10.254.33.66]) by orsmga007-auth.jf.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 16 May 2022 19:55:18 -0700 Message-ID: <1292fe3206bef08304dc1bfe3185a9e6cec690fd.camel@intel.com> Subject: Re: [PATCH v6 1/5] x86/tdx: Add TDX Guest attestation interface driver From: Kai Huang To: Kuppuswamy Sathyanarayanan , Thomas Gleixner , Ingo Molnar , Borislav Petkov , Dave Hansen , x86@kernel.org Cc: "H . Peter Anvin" , "Kirill A . Shutemov" , Tony Luck , Andi Kleen , Wander Lairson Costa , Isaku Yamahata , marcelo.cerri@canonical.com, tim.gardner@canonical.com, khalid.elmously@canonical.com, philip.cox@canonical.com, linux-kernel@vger.kernel.org In-Reply-To: <20220512221952.3647598-2-sathyanarayanan.kuppuswamy@linux.intel.com> References: <20220512221952.3647598-1-sathyanarayanan.kuppuswamy@linux.intel.com> <20220512221952.3647598-2-sathyanarayanan.kuppuswamy@linux.intel.com> Content-Type: text/plain; charset="UTF-8" MIME-Version: 1.0 Date: Tue, 17 May 2022 14:54:45 +1200 User-Agent: Evolution 3.42.4 (3.42.4-2.fc35) Content-Transfer-Encoding: 8bit X-Spam-Status: No, score=-2.1 required=5.0 tests=BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,DKIM_VALID_AU,HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,RDNS_NONE,SPF_HELO_NONE,T_SCC_BODY_TEXT_LINE autolearn=no autolearn_force=no version=3.4.6 X-Spam-Checker-Version: SpamAssassin 3.4.6 (2021-04-09) on lindbergh.monkeyblade.net Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Thu, 2022-05-12 at 15:19 -0700, Kuppuswamy Sathyanarayanan wrote: > In TDX guest, attestation is used to verify the trustworthiness of a TD > to other entities before provisioning secrets to the TD. > > One usage example is, when a TD guest uses encrypted drive and if the > decryption keys required to access the drive are stored in a secure 3rd > party keyserver, the key server can use attestation to verify TD's > trustworthiness and release the decryption keys to the TD. > > The attestation process consists of two steps: TDREPORT generation and > Quote generation. > > TDREPORT (TDREPORT_STRUCT) is a fixed-size data structure generated by > the TDX module which contains TD-specific information (such as TD > measurements), platform security version, and the MAC to protect the > integrity of the TDREPORT. The TD kernel uses TDCALL[TDG.MR.REPORT] to > get the TDREPORT from the TDX module. A user-provided 64-Byte > REPORTDATA is used as input and included in the TDREPORT. Typically it > can be some nonce provided by attestation service so the TDREPORT can > be verified uniquely. More details about TDREPORT can be found in > Intel TDX Module specification, section titled "TDG.MR.REPORT Leaf". > > Also note that the MAC added to the TDREPORT is bound to the platform. > So TDREPORT can only be verified locally.  > "bound to the platform, so ...". > Intel SGX Quote Enclave (QE) > is leveraged to verify the TDREPORT locally and convert it to a remote > verifiable Quote to support remote attestation of the TDREPORT. First of all, sorry for having to modify this back and forth for multi-times. However this sounds like besides the SGX QE, there are other ways can be leveraged to generate the Quote. This isn't true based on current TDX architecture. So I still think below is slightly better: TDREPORT can only be verified on local platform as the MAC key is bound to the platform. To support remote verification of the TDREPORT, TDX leverages Intel SGX Quote Enclave (QE) to verify the TDREPORT locally and convert it to a remote verifiable Quote. > > After getting the TDREPORT, the second step of the attestation process > is to send it to QE or Quote Generation Service (QGS) to generate a TD > Quote. The QE sends the Quote back after it is generated. How is the > data sent and received is QE implementation and deployment specific.TD > userspace attestation software can use whatever communication channel > available (i.e. tcp/ip, vsock) to communicate with the QE using whatever > data format. TDX also defines TDVMCALLs to allow TD to ask VMM to > facilitate sending/receiving data between TD attestation software and > the QE. This support is documented in GHCI 1.0 spec "5.4 TD attestation". This paragraph is used to provide more information to help to justify the decision to provide a way to let userspace to get the TDREPORT. Now looks this paragraph has details not quite related to this patch. For instance, I am not sure whether we need to mention TDVMCALL at all. Also, it _may_ be helpful to point out we cannot have QE inside the TD since TDX doesn't support SGX within the TD, otherwise it's completely possible that the TD attestation agent can just implement the QE by itself therefore doesn't need any communication channel at all. Anyway, perhaps just: " After getting the TDREPORT, the second step of the attestation process is to send it to the QE to generate the Quote. TDX doesn't support SGX inside the TD, so the QE can be deployed in the host, or in another legacy VM with SGX support. How to send the TDREPORT to QE and receive the Quote is implementation and deployment specific.   Implement a basic attestation driver to allow TD userspace to get the TDREPORT. The TD userspace attestation software can get the TDREPORT and then choose whatever communication channel available (i.e. vsock) to send the TDREPORT to QE and receive the Quote. " Anyway I am not good at writing changelog so will leave to others in the future. > > Implement a basic attestation driver to allow TD userspace to get the > TDREPORT, which is sent to QE by the attestation software to generate > a Quote for remote verification. > > Also note that explicit access permissions are not enforced in this > driver because the quote and measurements are not a secret. However > the access permissions of the device node can be used to set any > desired access policy. The udev default is usually root access > only. > > Operations like getting TDREPORT or Quote generation involves sending > a blob of data as input and getting another blob of data as output. It > was considered to use a sysfs interface for this, but it doesn't fit > well into the standard sysfs model for configuring values. It would be > possible to do read/write on files, but it would need multiple file > descriptors, which would be somewhat messy. IOCTLs seems to be the best > fitting and simplest model for this use case. Also, the REPORTDATA used > in TDREPORT generation can possibly come from attestation service to > uniquely verify the Quote (like per instance verification). In such > case, since REPORTDATA is a secret, using sysfs to share it is insecure > compared to sending it via IOCTL. > > Reviewed-by: Tony Luck > Reviewed-by: Andi Kleen > Acked-by: Kirill A. Shutemov > Signed-off-by: Kuppuswamy Sathyanarayanan > --- > arch/x86/coco/tdx/Makefile | 2 +- > arch/x86/coco/tdx/attest.c | 118 ++++++++++++++++++++++++++++++++ > arch/x86/include/uapi/asm/tdx.h | 42 ++++++++++++ > 3 files changed, 161 insertions(+), 1 deletion(-) > create mode 100644 arch/x86/coco/tdx/attest.c > create mode 100644 arch/x86/include/uapi/asm/tdx.h > > diff --git a/arch/x86/coco/tdx/Makefile b/arch/x86/coco/tdx/Makefile > index 46c55998557d..d2db3e6770e5 100644 > --- a/arch/x86/coco/tdx/Makefile > +++ b/arch/x86/coco/tdx/Makefile > @@ -1,3 +1,3 @@ > # SPDX-License-Identifier: GPL-2.0 > > -obj-y += tdx.o tdcall.o > +obj-y += tdx.o tdcall.o attest.o > diff --git a/arch/x86/coco/tdx/attest.c b/arch/x86/coco/tdx/attest.c > new file mode 100644 > index 000000000000..a5f4111f9b18 > --- /dev/null > +++ b/arch/x86/coco/tdx/attest.c > @@ -0,0 +1,118 @@ > +// SPDX-License-Identifier: GPL-2.0 > +/* > + * attest.c - TDX guest attestation interface driver. > + * > + * Implements user interface to trigger attestation process. > + * > + * Copyright (C) 2022 Intel Corporation > + * > + */ > + > +#define pr_fmt(fmt) "x86/tdx: attest: " fmt > + > +#include > +#include > +#include > +#include > +#include > + > +#define DRIVER_NAME "tdx-attest" > + > +/* TDREPORT module call leaf ID */ > +#define TDX_GET_REPORT 4 > + Looks more white spaces than needed. > +static struct miscdevice miscdev; > + > +static long tdx_get_report(void __user *argp) > +{ > + void *reportdata = NULL, *tdreport = NULL; > + long ret = 0; > + > + /* Allocate buffer space for REPORTDATA */ > + reportdata = kmalloc(TDX_REPORTDATA_LEN, GFP_KERNEL); > + if (!reportdata) > + return -ENOMEM; > + > + /* Allocate buffer space for TDREPORT */ > + tdreport = kmalloc(TDX_REPORT_LEN, GFP_KERNEL); > + if (!tdreport) { > + ret = -ENOMEM; > + goto out; > + } Perhaps you can allocate a single page, and use the first half as REPORTDATA and the second part as TDREPORT. In this case, you can save one memory allocation to simplify the code. The page will be freed anyway after this IOCTL. > + > + /* Copy REPORTDATA from the user buffer */ > + if (copy_from_user(reportdata, argp, TDX_REPORTDATA_LEN)) { > + ret = -EFAULT; > + goto out; > + } > + > + /* > + * Generate TDREPORT using "TDG.MR.REPORT" TDCALL. > + * > + * Get the TDREPORT using REPORTDATA as input. Refer to > + * section 22.3.3 TDG.MR.REPORT leaf in the TDX Module 1.0 > + * Specification for detailed information. > + */ > + ret = __tdx_module_call(TDX_GET_REPORT, virt_to_phys(tdreport), > + virt_to_phys(reportdata), 0, 0, NULL); > + if (ret) { > + pr_debug("TDREPORT TDCALL failed, status:%lx\n", ret); > + ret = -EIO; > + goto out; > + } > + > + /* Copy TDREPORT back to the user buffer */ > + if (copy_to_user(argp, tdreport, TDX_REPORT_LEN)) > + ret = -EFAULT; > + > +out: > + kfree(reportdata); > + kfree(tdreport); > + return ret; > +} > + > +static long tdx_attest_ioctl(struct file *file, unsigned int cmd, > + unsigned long arg) > +{ > + void __user *argp = (void __user *)arg; > + long ret = -EINVAL; > + > + switch (cmd) { > + case TDX_CMD_GET_REPORT: > + ret = tdx_get_report(argp); > + break; > + default: > + pr_debug("cmd %d not supported\n", cmd); > + break; > + } > + > + return ret; > +} > + > +static const struct file_operations tdx_attest_fops = { > + .owner = THIS_MODULE, > + .unlocked_ioctl = tdx_attest_ioctl, > + .llseek = no_llseek, > +}; > + > +static int __init tdx_attestation_init(void) > +{ > + long ret; > + > + /* Make sure we are in a valid TDX platform */ > + if (!cpu_feature_enabled(X86_FEATURE_TDX_GUEST)) > + return -EIO; > + > + miscdev.name = DRIVER_NAME; > + miscdev.minor = MISC_DYNAMIC_MINOR; > + miscdev.fops = &tdx_attest_fops; > + > + ret = misc_register(&miscdev); > + if (ret) { > + pr_err("misc device registration failed\n"); > + return ret; > + } > + > + return 0; > +} > +device_initcall(tdx_attestation_init) > diff --git a/arch/x86/include/uapi/asm/tdx.h b/arch/x86/include/uapi/asm/tdx.h > new file mode 100644 > index 000000000000..e06da56058a1 > --- /dev/null > +++ b/arch/x86/include/uapi/asm/tdx.h > @@ -0,0 +1,42 @@ > +/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */ > +#ifndef _UAPI_ASM_X86_TDX_H > +#define _UAPI_ASM_X86_TDX_H > + > +#include > +#include > + > +/* Length of the REPORTDATA used in TDG.MR.REPORT TDCALL */ > +#define TDX_REPORTDATA_LEN 64 > + > +/* Length of TDREPORT used in TDG.MR.REPORT TDCALL */ > +#define TDX_REPORT_LEN 1024 > + > +/** > + * struct tdx_report_req: Get TDREPORT using REPORTDATA as input. > + * > + * @reportdata : User-defined 64-Byte REPORTDATA to be included into > + * TDREPORT. Typically it can be some nonce provided by > + * attestation software so the generated TDREPORT can be "attestation software" -> "attestation service" or "verification service". Or just remove the second sentence. Anyway, it's user-defined. > + * uniquely verified. > + * @tdreport : TDREPORT output from TDCALL[TDG.MR.REPORT] of size > + * TDX_REPORT_LEN. > + * > + * Used in TDX_CMD_GET_REPORT IOCTL request. > + */ > +struct tdx_report_req { > + union { > + __u8 reportdata[TDX_REPORTDATA_LEN]; > + __u8 tdreport[TDX_REPORT_LEN]; > + }; > +}; As a userspace ABI, one concern is this doesn't provide any space for future extension. But probably it's OK since I don't see any possible additional input for now. And although TDREPORT may have additional information in future generation of TDX but the spec says the size is 1024 so perhaps this won't change even in the future. Anyway will leave to others. > + > +/* > + * TDX_CMD_GET_REPORT - Get TDREPORT using TDCALL[TDG.MR.REPORT) TDCALL[TDG.MR.REPORT) -> TDCALL[TDG.MR.REPORT] > + * > + * Return 0 on success, -EIO on TDCALL execution failure, and > + * standard errno on other general error cases. > + * > + */ > +#define TDX_CMD_GET_REPORT _IOWR('T', 0x01, struct tdx_report_req) > + > +#endif /* _UAPI_ASM_X86_TDX_H */