Trusted Execution Environment

Running secure user services in a hostile environment TROOS – Trusted Open OS, the basis for a trustzone based open TEE Assaf Rosenbuam, Eli Biham and Sara Bitan Department of Computer Science, Technion Do you trust your OS? We don’t... Current TEE technologies Our contribution – TROOS Secure world interface The normal world OS utilizes the secure world in- Our goal is to provide a TEE for user services (much Intel software guard extension (SGX) terface in order to create and execute trustlets ac- like Intel’s SGX). We use Genode as the basis for cording to its needs via SMCs. As part of our efforts 60 SGX enables to run secure services in an isolated our trusted OS. On top of the native Core and Init 40 to keep the system attack surface as small as possi- execution environment (called enclave), embedded components, which are the system kernel and first 20 ble, we keep the number of SMCs at the bare min- in the process virtual memory space. user process respectively, we added a few more com- 0 imum. The most significant SMCs are described in ponents to enable trusted user services – trustlets – the following table: loading and execution. Critical vulnerabiliteis found SMC Description Jan–Apr 2017 Tcreate create a new (empty) trustlet Tload load and measure the trustlet code How to protect critical assets Tinit mark trustlet as ready to run from a compromised OS? Tstart start executing a trustlet Tresume resume trustlet run after it was stopped Exploitation of an OS vulnerability might leave the Tdestroy stop the trustlet run and free its resources system resources completely exposed to attacks. Hence, we need a mechanism (entity) that can be trusted even when the OS is breached. ARM TrsutZone Future work TrustZone provides a technology to run two virtual TZ VMM: Intercepts SMCs and handles the switch Our next step is to complete the system implemen- worlds – a Normal World (NW) and Secure World between the two worlds. tation with all of the designed components. We (SW) – on the same CPU. This mechanism is an Loader: Responsible of loading and verifying the then plan to deliver an elaborated security analysis infrastructure on which we can create a Trusted trustlets code from the normal world. of the system and a comparison to existing TEEs Execution Environment. Trustlet Manager: Acts as a container for the (TrustZone based and others). The preference im- trustlets. Manages the trustlets resources pact on the normal world will be tested as well. allocation. Due to time constrains, there are aspects that we Crypto: Will provide each trustlet with unique do not plan to address at this stage, even though keys, accessible only to it and only while they surely are beneficial to TROOS. For example: running on the secure world. • Integrating a secure element. Trustlet TCB on TROOS • Extending system abilities with secure IO. • Utilizing on chip memory to better protect trustlets code and data against probing. Trusted Execution Environment Trustlet1 Trustlet2 Trustlet3 (TEE) TZ VMM Trustlets Manager Loader Crypto Contact Information We need an environment in which we can safely The Genode OS framework store critical data and perform critical processing Init Genode is an open source operat- ing system framework designed for Core User Space increased security. Genode main- Application TCB OS tains strong application isolation Notice that in TROOS, a trustlet must not trust the TEE in Genode properties and is designed to keep other trustlets in the system. A trustlet doesn’t each application Trusted Computing Base (TCB) even need to trust the TROOS services, unless it as small as possible. wishes to use them..

Copy Link

Recommended publications

Inspecting Data from the Safety of Your Trusted Execution Environment
Inspecting data from the safety of your trusted execution environment John Williams johnwwil [at] u.washington.edu June, 2015 Abstract: This paper presents a proof of concept that uses ARM TrustZone to perform introspection of a Linux kernel running in the normal world from within a secure-world system. Techniques from existing volatile-memory analysis applications are repurposed for application in real-time through asynchronous introspection of the normal world. The solution presented here leverages open-source software that is actively being developed to support additional architectures. 1 INTRODUCTION As technology becomes more prevalent in our day-to-day lives, we increasingly rely on mobile devices for both business and personal purposes. Emerging security threats, such as rootkits that can scan memory for protected PCI information, have been responsible for large-scale breaches. It has been predicted that targeted exploit kits and attack services will soon exist for mobile platforms, which could enable the exploitation of mobile devices on an unprecedented scale [1]. Techniques for mitigating such threats to mobile devices have largely evolved in parallel with industry demand for them. While some companies have indeed made progress through continued addition of product features that increase the security of devices, other companies have simply pivoted their marketing efforts to emphasize or recast certain aspects of their existing capabilities. In order to select the right products and features for protecting sensitive data and activities, effective segmentation is key. While segmentation within devices has traditionally been implemented at the operating system level, in recent years architectural modification at the chip level has emerged as a way to provide hard segmentation between execution environments.
[Show full text]
Leveraging ARM Trustzone and Verifiable Computing to Provide Auditable Mobile Functions Nuno O
Leveraging ARM TrustZone and Verifiable Computing to Provide Auditable Mobile Functions Nuno O. Duarte Sileshi Demesie Yalew INESC-ID / IST, University of Lisbon KTH Royal Institute of Technology [email protected] [email protected] Nuno Santos Miguel Correia INESC-ID / IST, University of Lisbon INESC-ID / IST, University of Lisbon [email protected] [email protected] ABSTRACT 1 INTRODUCTION The increase of personal data on mobile devices has been fol- For over 30 years [5], an important problem that has fasci- lowed by legislation that forces service providers to process nated the research community is about trusting the result of and maintain users’ data under strict data protection policies. computations performed on untrusted platforms. Since the In this paper, we propose a new primitive for mobile applica- advent of cloud computing, this interest has further grown tions called auditable mobile function (AMF) to help service in obtaining proofs of code execution on untrusted clouds. providers enforcing such policies by enabling them to process An approach to address this problem that has garnered sensitive data within users’ devices and collecting proofs of particular attention is verifiable computing. At a high-level, function execution integrity. We present SafeChecker, a com- verifiable computing consists of studying how a prover can putation verification system that provides mobile application convince a verifier of a mathematical assertion. The main support for AMFs, and evaluate the practicality of different obstacle faced by this field has been practicality, as the costs usage scenario AMFs on TrustZone-enabled hardware. of computing and verifying proofs over general computa- tions tend to be very high [21].
[Show full text]
Trusted Computing Serving an Anonymity Service
Trusted Computing Serving an Anonymity Service Alexander Böttcher Bernhard Kauer Hermann Härtig Technische Universität Dresden Department of Computer Science Operating Systems Group {boettcher, kauer, haertig}@os.inf.tu-dresden.de Abstract [15] in the authors’ group, will show. An unknown crim- inal to be investigated was supposed to use an anonymity We leveraged trusted computing technology to service. The police and later on the German Federal Bu- counteract certain insider attacks. Furthermore, we reau of Criminal Investigation (FBCI) required to investi- show with one of the rare server based scenarios that gate a criminal that uses a known URL. In order to enable an anonymity service can proﬁt from trusted comput- criminal prosecution by a given warrant the software was ing. We based our design on the Nizza Architecture extended to log connection data [6, 5]. This function of [14] with its small kernel and minimal multi-server the anonymity software [3] was used only with a warrant OS. We even avoided Nizza’s legacy container and but without directly notifying the users. After the warrant got a much smaller, robust and hopefully more secure was reversed the FBCI showed up with a delivery warrant system, since we believe that minimizing the trusted for a log record (which were later on reversed illegal [4]), computing base is an essential requirement for trust ﬁrst at the institute and later on at the institute’s directors into software. private home. This incident shows a whole class of attacks on the 1 Introduction anonymity and more general every computing service. Vendor and providers under constraint can reveal the ser- Anonymity while using the Internet is widely considered vice because of insider knowledge and physical access a legitimate and - for many use cases - essential require- to those services.
[Show full text]
The Nizza Secure-System Architecture
Appears in the proceedings of CollaborateCom 2005, San Jose, CA, USA The Nizza Secure-System Architecture Hermann Härtig Michael Hohmuth Norman Feske Christian Helmuth Adam Lackorzynski Frank Mehnert Michael Peter Technische Universität Dresden Institute for System Architecture D-01062 Dresden, Germany [email protected] Abstract rely on a standard OS (including the kernel) to assure their security properties. The trusted computing bases (TCBs) of applications run- To address the conﬂicting requirements of complete ning on today’s commodity operating systems have become functionality and the protection of security-sensitive data, extremely large. This paper presents an architecture that researchers have devised system architectures that reduce allows to build applications with a much smaller TCB. It the system’s TCB by running kernels in untrusted mode is based on a kernelized architecture and on the reuse of in a secure compartment on top of a small security kernel; legacy software using trusted wrappers. We discuss the de- security-sensitive services run alongside the OS in isolated sign principles, the architecture and some components, and compartments of their own. This architecture is widely re- a number of usage examples. ferred to as kernelized standard OS or kernelized system. In this paper, we describe Nizza, a new kernelized- system architecture. In the design of Nizza, we set out to answer the question of how small the TCB can be made. 1 Introduction We have argued in previous work that the (hardware and software) technologies needed to build small secure-system Desktop and hand-held computers are used for many platforms have become much more mature since earlier at- functions, often in parallel, some of which are security tempts [8].
[Show full text]
Reducing TCB Size by Using Untrusted Components — Small Kernels Versus Virtual-Machine Monitors
To be published in Proceedings of the 11th ACM SIGOPS European Workshop, Leuven, Belgium, 2004 Reducing TCB size by using untrusted components — small kernels versus virtual-machine monitors Michael Hohmuth Michael Peter Hermann Hartig¨ Jonathan S. Shapiro Technische Universitat¨ Dresden Johns Hopkins University Department of Computer Science Department of Computer Science Operating Systems Group Systems Research Laboratory hohmuth,peter,haertig ¡ @os.inf.tu-dresden.de [email protected] Abstract tems, such as message passing and memory sharing, the overall size of the TCB (which includes all components an Secure systems are best built on top of a small trusted oper- application relies on) can actually be reduced for a large class ating system: The smaller the operating system, the easier it of applications. can be assured or veriﬁed for correctness. Another assumption we address in this paper is that all In this paper, we oppose the view that virtual-machine components on which an application has operational depen- monitors (VMMs) are the smallest systems that provide se- dencies must be in this application’s TCB. This presumption cure isolation because they have been speciﬁcally designed leads to the unnecessary inclusion of many (protocol and de- to provide little more than this property. The problem with vice) drivers into the TCB. this assertion is that VMMs typically do not support inter- The basic idea for reducing TCB size is to extract sys- process communication, complicating the use of untrusted tem components from the TCB and consider them as un- components inside a secure systems. trusted without violating the security requirements of user We propose extending traditional VMMs with features for applications.
[Show full text]
Trusted Computer System Evaluation Criteria
DoD 5200.28-STD Supersedes CSC-STD-00l-83, dtd l5 Aug 83 Library No. S225,7ll DEPARTMENT OF DEFENSE STANDARD DEPARTMENT OF DEFENSE TRUSTED COMPUTER SYSTEM EVALUATION CRITERIA DECEMBER l985 December 26, l985 Page 1 FOREWORD This publication, DoD 5200.28-STD, "Department of Defense Trusted Computer System Evaluation Criteria," is issued under the authority of an in accordance with DoD Directive 5200.28, "Security Requirements for Automatic Data Processing (ADP) Systems," and in furtherance of responsibilities assigned by DoD Directive 52l5.l, "Computer Security Evaluation Center." Its purpose is to provide technical hardware/firmware/software security criteria and associated technical evaluation methodologies in support of the overall ADP system security policy, evaluation and approval/accreditation responsibilities promulgated by DoD Directive 5200.28. The provisions of this document apply to the Office of the Secretary of Defense (ASD), the Military Departments, the Organization of the Joint Chiefs of Staff, the Unified and Specified Commands, the Defense Agencies and activities administratively supported by OSD (hereafter called "DoD Components"). This publication is effective immediately and is mandatory for use by all DoD Components in carrying out ADP system technical security evaluation activities applicable to the processing and storage of classified and other sensitive DoD information and applications as set forth herein. Recommendations for revisions to this publication are encouraged and will be reviewed biannually by the National Computer Security Center through a formal review process. Address all proposals for revision through appropriate channels to: National Computer Security Center, Attention: Chief, Computer Security Standards. DoD Components may obtain copies of this publication through their own publications channels.
[Show full text]
Secure Bootstrap Is Not Enough: Shoring up the Trusted Computing Base James Hendricks Leendert Van Doorn Carnegie Mellon University IBM T.J
Proceedings of the Eleventh SIGOPS European Workshop, ACM SIGOPS, Leuven, Belgium, September 2004. Secure Bootstrap is Not Enough: Shoring up the Trusted Computing Base James Hendricks Leendert van Doorn Carnegie Mellon University IBM T.J. Watson Research Center 5000 Forbes Ave 19 Skyline Drive Pittsburgh, PA Hawthorne, NY [email protected] [email protected] Abstract for each bootstrap step before it is executed. For exam- We propose augmenting secure boot with a mechanism ple, the BIOS could verify a public-key signature of the to protect against compromises to field-upgradeable de- disk’s boot sector to ensure its authenticity; the boot sector vices. In particular, secure boot standards should verify could then verify the public-key signature of the OS boot- the firmware of all devices in the computer, not just de- strap code, which could likewise verify the privileged OS vices that are accessible by the host CPU. Modern comput- processes and drivers. Though such an approach would ob- ers contain many autonomous processing elements, such viously not guarantee the security of the OS code, it would as disk controllers, disks, network adapters, and coproces- at least guarantee the authenticity. sors, that all have field-upgradeable firmware and are an A weakness to this approach is that the BIOS in most essential component of the computer system’s trust model. personal computers is writable. One solution is to store Ignoring these devices opens the system to attacks similar the BIOS on a ROM. However, a ROM-based approach to those secure boot was engineered to defeat.
[Show full text]
Trusted Computing Base
Microkernels and L4 Introduction COMP9242 2006/S2 Week 1 cse/UNSW/NICTA WHY MICROKERNELS? MONOLITHIC KERNEL WHY MICROKERNELS? MONOLITHIC KERNEL: • Kernel has access to everything § all optimisations possible § all techniques/mechanisms/concepts implementable WHY MICROKERNELS? MONOLITHIC KERNEL: • Kernel has access to everything § all optimisations possible § all techniques/mechanisms/concepts implementable • Can be extended by simply adding code WHY MICROKERNELS? MONOLITHIC KERNEL: • Kernel has access to everything § all optimisations possible § all techniques/mechanisms/concepts implementable • Can be extended by simply adding code • Cost: Complexity § growing size § limited maintainability cse/UNSW/NICTA COMP9242 2006/S2 W1 P1 MICROKERNEL:IDEA • Small kernel providing core functionality § only code running in privileged mode • Most OS services provided by user-level servers • Applications communicate with servers via message-passing IPC Application syscall user mode VFS IPC, file system Application UNIX Device File Server Driver Server Scheduler, virtual memory kernel IPC mode Device drivers, Dispatcher, ... IPC, virtual memory Hardware Hardware cse/UNSW/NICTA COMP9242 2006/S2 W1 P2 TRUSTED COMPUTING BASE The part of the system which must be trusted to operate correctly TRUSTED COMPUTING BASE The part of the system which must be trusted to operate correctly Application Service Hardware System: traditional embedded TRUSTED COMPUTING BASE The part of the system which must be trusted to operate correctly Application Service Hardware System:
[Show full text]
TC) and Next Generation Secured Computing Base (NGSCB
Trusted Computing (TC) and Next Generation Secured Computing Base (NGSCB) “For years, Bill Gate has dreamed of finding a way to make the Chinese pay for software: TC looks like being the answer to his prayer.” – Ross Anderson [Reference 7] The Trusted Computing (TC) and Next Generation Secured Computing Base (NGSCB) Version 1.0 Spring 2005 Computer Networking Management By: Jianji (Joseph) Yu Jeffrey Khuu Joseph Yu and Jeffrey Khuu Computer Network Management Page 1 of 10 Trusted Computing (TC) and Next Generation Secured Computing Base (NGSCB) Table of Contents 1. Introduction ……………………………………………………………3 Overview 1.1 The Trusted Computing: Trusted Computing Base (TCB) And Trusted Computing Group (TCG) …………………………3 1.2 Next Generation Secured Computing Base (NGSCB) ......…4 2. NGSCB ……………………………………………………………....….5 2.1 Features 2.1.1 Two operating environments ….............………………….…..5 2.1.2 Four features of NGSCB …..……………......………....…...5 2.2 Architecture 2.2.1 Two primary system components in Trusted operating environment ………………...….6 2.2.2 Nexus …………...……………………………………………..7 2.2.3 NCA …………………………..……......…………………………. 7 2.3 NGSCB Summary …………...………….......…………………….8 3. Analysis of NGSCB ……………………………………………9 3.1 The current problematic computing ……………………………9 3.2 Applications ………………………………………………….……….9 3.3 Will NGSCB be the solution? …………………………….………9 4.References …………………………………………………………....10 Joseph Yu and Jeffrey Khuu Computer Network Management Page 2 of 10 Trusted Computing (TC) and Next Generation Secured Computing Base (NGSCB) 1. Introduction Overview 1.1 The Trusted Computing ² Trusted Computing (TC) The term Trusted Computing (TC) means the computer security integrated with hardware and software security components. As Ross Anderson suggested, the original motivation of TC was intended for digital rights management (DRM); that limits the abuse of file sharing over the network and making illegal copies without the authorization from the vendors, perhaps, restrict user's computing actions.
[Show full text]
Topic 20: TCSEC and Common Criteria
Information Security CS 526 Topic 20: TCSEC and Common Criteria CS526 Topic 20: TCSEC and Common 1 Criteria Related Readings for This Lecture • Wikipedia – Trusted computing base – TCSEC – Common Criteria, – Evaluation Assurance Level CS526 Topic 20: TCSEC and Common 2 Criteria Terminology: Trusted Computing Base (TCB) • The set of all hardware, software and procedural components that enforce the security policy. – In order to break security, an attacker must subvert one or more of them. – The smaller the TCB, the more secure a system is. • What consists of the conceptual Trusted Computing Based in a Unix/Linux system? – hardware, kernel, system binaries, system configuration files, setuid root programs, etc. One approach to improve security is to reduce the size of TCB, i.e., reduce what one relies on for security. CS526 Topic 20: TCSEC and Common 3 Criteria Assurance • Assurance: “estimate of the likelihood that a system will not fail in some particular way” • Based on factors such as – Software architecture • E.g., kernelized design, – Development process – Who developed it – Technical assessment CS526 Topic 20: TCSEC and Common 4 Criteria Kernelized Design User space • Trusted Computing Base User – Hardware and software for process enforcing security rules • Reference monitor – Part of TCB – All system calls go through Reference reference monitor for security monitor checking TCB – Most OS not designed this way OS kernel Kernel space CS526 Topic 20: TCSEC and Common 5 Criteria Reference Monitor • Three required properties for reference
[Show full text]
TCG Guidance for Secure Update of Software and Firmware on F Embedded Systems E
R E TCG Guidance for Secure Update of Software and Firmware on F Embedded Systems E R Version 1.0 Revision 64 E July 18, 2019 N Contact: [email protected] C PUBLIC REVIEW E Work in Progress This document is an intermediate draft for comment only and is subject to change without notice. Readers should not design products based on this document. TCG Guidance for Secure Update of Software and Firmware on Embedded Systems | Version 1.0 | Revision 64 | 7/18/2019 | PUBLIC REVIEW © TCG 2019 TCG Guidance for Secure Update of Software and Firmware on Embedded Systems DISCLAIMERS, NOTICES, AND LICENSE TERMS THIS DOCUMENT IS PROVIDED "AS IS" WITH NO WARRANTIES WHATSOEVER, INCLUDING ANY WARRANTY OF MERCHANTABILITY, NONINFRINGEMENT, FITNESS FOR ANY PARTICULAR PURPOSE, OR ANY WARRANTY OTHERWISE ARISING OUT OF ANY PROPOSAL, DOCUMENT OR SAMPLE. Without limitation, TCG disclaims all liability, including liability for infringement of any proprietary rights, relating to use of information in this document and to the implementation of this document, and TCG disclaims all liability for cost of procurement of substitute goods or services, lost profits, loss of use, loss of data or any incidental, consequential, direct, indirect, or special damages, whether under contract, tort, warranty or otherwise, arising in any way out of use or reliance upon this document or any information herein. This document is copyrighted by Trusted Computing Group (TCG), and no license, express or implied, is granted herein other than as follows: You may not copy or reproduce the document or distribute it to others without written permission from TCG, except that you may freely do so for the purposes of (a) examining or implementing TCG documents or (b) developing, testing, or promoting information technology standards and best practices, so long as you distribute the document with these disclaimers, notices, and license terms.
[Show full text]
Georgios V. Michalakidis [email protected]
CSM27 Computer Security University of Surrey Autumn 2008 – Week 5 Exercise Georgios V. Michalakidis [email protected] Weekly Exercises 1.1 Gollmann 5.2 Can you have security without security kernels? Discuss the advantages and disadvantages of having a security kernel built into the Operating System Kernel (as opposed to the Application Layer) to form the trusted computing base (TCB). 1.2 Gollmann 5.5 Some buffer overrun attacks put the code they want to be executed on the call stack. How can the ability to distinguish between programs and data help to construct a defense against this particular type of buffer overrun attacks? Briefly describe a protection mechanism based on this distinction. 1.3 P&P 5.9 Consider time-sharing on the CPU. Explain what is necessary to provide temporal separation (with proper security). Your answer can take one of two approaches • Describe the (formal) conditions which must be met in order for two processes to be adequately separated. • Describe each action which must be taken by the CPU and OS during a context switch (i.e. when one process is swapped out and a new one in)? Give rationale for each condition/action. Response to Exercise 1.1: Security kernels are not only needed but are obligatory in order to have good protection over code or data or any other type of insecure state. In all modern Operating Systems there exist different and separated security levels inside the kernels, coping with intrusion matters or execution of malicious code. If we were only using security on the application layer, chances are we would be facing problems caused by bugs or any other types of threats.
[Show full text]