Example Computer Architecture Research Project

Total Page：16

File Type：pdf, Size：1020Kb

Example Computer Architecture Research Project Simon Moore Trustworthy Processor Design • Mo>vaon – Security/Trustworthiness is increasingly important • need the hardware to help enforce policies • Hypothesis – Capsicum (next slide) demonstrated that capability based protec>on is good for fine-grained sandboxing of applicaons but with a performance cost – Hardware based capabili>es will allow more security but with less performance overhead Capsicum USNIX Security Best paper 2010 Observaons from Capsicum • SoRware designs that employ the principle of least privilege are neither easily nor efficiently represented in current hardware • Kernels and programming language run>mes (TCBs) building directly on hardware in C are enormous and unsound • SoRware TCB implementaons embody ar>facts of security policies rather than design principles CAP Computer (1970s) Checkered History of Capability Machines • 1966: Dennis & Van Horn invent the term • 1972: Plessey System 250 use hardware capabili>es commercially • 1976: Cambridge CAP Computer • 1979: IBM System/38 • 1981: Intel iAPX 432 – embodiment of CISC • 1999: EROS uses soRware capabili>es • 2010: Capsicum capability security model A RISC approach to capabili>es: CHERI • Base system - our own 64 bit MIPS style core (BERI) + extensive regression test suite • Implemented in Bluespec targe>ng FPGAs • Running FreeBSD – complete UNIX setup • Now adding capability mechanisms to hardware and OS CHERI tablet demo plaorm SoRware compartmentalisaon Conventional "fetch" program Compartmentalized "fetch" program Conventional Capability mode process main UNIX process loop vulnerable vulnerable main HTTP fetch HTTP fetch loop logic logic Kernel Kernel • SoRware compartmentalisaon decomposes applicaons into many isolated components • Each running with only the rights required to perform its func>on • This implements the principle of least privilege Capability Register Model Protection is a )*+*,-.#/0,123*# 7-1-85.569#!*4536*,3# first-class primitive !*4536*,3# /7# 1*,:3# 2691*#7/7# 8-3*# .*+46;# !"#$$#"# 1*,:3#7!"#<4*+*,-.#10,123*#=-1-85.569>#2691*# 8-3*# .*+46;# !%# 1*,:3# 2691*#7!%# 8-3*# .*+46;# !&# 1*,:3# 2691*#7!&# 8-3*# .*+46;# Compiler manages '# Capabilities '# capability registers as it describe data, '# '# does general purpose code & objects '# '# registers !(%# 1*,:3# 2691*#7!(%#8-3*# .*+46;# Memory Access 8$/6,* +&"'()"* !"#$%&'()"** .,",%/0* 9/4/:707$;* +,$'-* 1/$/* 1/$/* 2'',##* 2'',##* .,",%/0* 3&%4)#,* !"# 56# 56# 5,67#$,%#* 9/4/:707$;* < =#=-):# 5,67#$,%* 9/4/:707$;* =/#,* "!"# "57# "56# 5,67#$,%#* >*5/"6,*?*3,%@*9-,'A* 9.4:;'0#8**4)--# Capabilities use virtual 3/6,*</:0,* <%/"#0/()"* $%&# $%&# $%&# addresses !+,-./'0#8**4)--# Memory allocation & legacy support $'(()*#!+,-./'0#1)234,# Summary • CHERI = MIPS + capabili>es • Aiming to show that hardware-based fine- grained protec>on is a real winner for real applicaons Conclusions • There’s lots of open research ques>ons in computer architecture • FPGAs provide an efficient “sand pit” for computer architecture research Ph.D. posi>ons, Part II projects, etc. • lots of opportuni>es available for bright well mo>vated individuals to join the team over the next few years • Need people interested in: – computer architecture – operang systems/run->me systems – security – compilaon techniques – etc… .

Copy Link

Recommended publications

Operating Systems and Virtualisation Security Knowledge Area (Draft for Comment)
OPERATING SYSTEMS AND VIRTUALISATION SECURITY KNOWLEDGE AREA (DRAFT FOR COMMENT) AUTHOR: Herbert Bos – Vrije Universiteit Amsterdam EDITOR: Andrew Martin – Oxford University REVIEWERS: Chris Dalton – Hewlett Packard David Lie – University of Toronto Gernot Heiser – University of New South Wales Mathias Payer – École Polytechnique Fédérale de Lausanne © Crown Copyright, The National Cyber Security Centre 2019. Following wide community consultation with both academia and industry, 19 Knowledge Areas (KAs) have been identified to form the scope of the CyBOK (see diagram below). The Scope document provides an overview of these top-level KAs and the sub-topics that should be covered under each and can be found on the project website: https://www.cybok.org/. We are seeking comments within the scope of the individual KA; readers should note that important related subjects such as risk or human factors have their own knowledge areas. It should be noted that a fully-collated CyBOK document which includes issue 1.0 of all 19 Knowledge Areas is anticipated to be released by the end of July 2019. This will likely include updated page layout and formatting of the individual Knowledge Areas. Operating Systems and Virtualisation Security Herbert Bos Vrije Universiteit Amsterdam April 2019 INTRODUCTION In this knowledge area, we introduce the principles, primitives and practices for ensuring security at the operating system and hypervisor levels. We shall see that the challenges related to operating system security have evolved over the past few decades, even if the principles have stayed mostly the same. For instance, when few people had their own computers and most computing was done on multiuser (often mainframe-based) computer systems with limited connectivity, security was mostly focused on isolating users or classes of users from each other1.
[Show full text]
CHERI Instruction-Set Architecture
UCAM-CL-TR-876 Technical Report ISSN 1476-2986 Number 876 Computer Laboratory Capability Hardware Enhanced RISC Instructions: CHERI Instruction-Set Architecture Robert N. M. Watson, Peter G. Neumann, Jonathan Woodruff, Michael Roe, Jonathan Anderson, David Chisnall, Brooks Davis, Alexandre Joannou, Ben Laurie, Simon W. Moore, Steven J. Murdoch, Robert Norton, Stacey Son September 2015 15 JJ Thomson Avenue Cambridge CB3 0FD United Kingdom phone +44 1223 763500 http://www.cl.cam.ac.uk/ c 2015 Robert N. M. Watson, Peter G. Neumann, Jonathan Woodruff, Michael Roe, Jonathan Anderson, David Chisnall, Brooks Davis, Alexandre Joannou, Ben Laurie, Simon W. Moore, Steven J. Murdoch, Robert Norton, Stacey Son, SRI International Approved for public release; distribution is unlimited. Sponsored by the Defense Advanced Research Projects Agency (DARPA) and the Air Force Research Laboratory (AFRL), under contracts FA8750-10-C-0237 (“CTSRD”) and FA8750-11-C-0249 (“MRC2”) as part of the DARPA CRASH and DARPA MRC research programs. The views, opinions, and/or ﬁndings contained in this report are those of the authors and should not be interpreted as representing the ofﬁcial views or policies, either expressed or implied, of the Department of Defense or the U.S. Government. Additional support was received from St John’s College Cambridge, the SOAAP Google Focused Research Award, the RCUK’s Horizon Digital Economy Research Hub Grant (EP/G065802/1), the EPSRC REMS Programme Grant (EP/K008528/1), the Isaac Newton Trust, the UK Higher Education Innovation Fund (HEIF), and Thales E-Security. Technical reports published by the University of Cambridge Computer Laboratory are freely available via the Internet: http://www.cl.cam.ac.uk/techreports/ ISSN 1476-2986 Abstract This technical report describes CHERI ISAv4, the fourth version of the Capability Hardware Enhanced RISC Instructions (CHERI) Instruction-Set Architecture (ISA)1 being developed by SRI International and the University of Cambridge.
[Show full text]
A VLSI Architecture for Enhancing Software Reliability Kanad Ghose Iowa State University
Iowa State University Capstones, Theses and Retrospective Theses and Dissertations Dissertations 1988 A VLSI architecture for enhancing software reliability Kanad Ghose Iowa State University Follow this and additional works at: https://lib.dr.iastate.edu/rtd Part of the Computer Sciences Commons Recommended Citation Ghose, Kanad, "A VLSI architecture for enhancing software reliability " (1988). Retrospective Theses and Dissertations. 9345. https://lib.dr.iastate.edu/rtd/9345 This Dissertation is brought to you for free and open access by the Iowa State University Capstones, Theses and Dissertations at Iowa State University Digital Repository. It has been accepted for inclusion in Retrospective Theses and Dissertations by an authorized administrator of Iowa State University Digital Repository. For more information, please contact [email protected]. INFORMATION TO USERS The most advanced technology has been used to photo graph and reproduce this manuscript from the microfilm master. UMI films the original text directly from the copy submitted. Thus, some dissertation copies are in typewriter face, while others may be from a computer printer. In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyrighted material had to be removed, a note will indicate the deletion. Oversize materials (e.g., maps, drawings, charts) are re produced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps. Each oversize page is available as one exposure on a standard 35 mm slide or as a 17" x 23" black and white photographic print for an additional charge.
[Show full text]
Capabilities
CS 261, Fall 2015 Scribe notes September 8: Capabilities Scribe: Riyaz Faizullabhoy 1 Motivation for Capabilities In The Confused Deputy, the compiler had two sets of rights: one from the user to access user files such as source code to compile, and another set granted to the compiler intended to access a special statistics file that would collect information about language feature usage. The compiler was installed in the SYSX directory, where it was granted access to write to any file in this directory. Naturally, the statistics file SYSX/STAT was also located in this directory such that the compiler (SYSX/FORT) could add language feature information. To achieve this, the compiler's privileges were set with a home files license that was allowed by the operating system to write to any file in the SYSX directory. A billing information file SYSX/BILL was also stored in SYSX { due to its sensitive nature, this billing file was not directly accessible to users. However, since the compiler was granted access to the entire SYSX directory, users could subvert their access restrictions on the billing information file by using the compiler like so: SYSX/FORT foo.f -o SYSX/BILL The above user-invoked command to the compiler would allow the compiler to first read the user's source code in foo.f, but then the compiler's privilege to SYSX would cause the SYSX/BILL file to be overwritten with the user's binary. In effect, the compiler is the confused deputy having to reconcile both its own and the user's set of privileges, but unfortunately allows for unintended behavior.
[Show full text]
SCALABLE CAPABILITY-BASED AUTHORIZATION for HIGH- PERFORMANCE PARALLEL FILE SYSTEMS Nicholas Mills Clemson University, [email protected]
Clemson University TigerPrints All Theses Theses 5-2011 SCALABLE CAPABILITY-BASED AUTHORIZATION FOR HIGH- PERFORMANCE PARALLEL FILE SYSTEMS Nicholas Mills Clemson University, [email protected] Follow this and additional works at: https://tigerprints.clemson.edu/all_theses Part of the Computer Engineering Commons Recommended Citation Mills, Nicholas, "SCALABLE CAPABILITY-BASED AUTHORIZATION FOR HIGH-PERFORMANCE PARALLEL FILE SYSTEMS" (2011). All Theses. 1131. https://tigerprints.clemson.edu/all_theses/1131 This Thesis is brought to you for free and open access by the Theses at TigerPrints. It has been accepted for inclusion in All Theses by an authorized administrator of TigerPrints. For more information, please contact [email protected]. SCALABLE CAPABILITY-BASED AUTHORIZATION FOR HIGH-PERFORMANCE PARALLEL FILE SYSTEMS A Thesis Presented to the Graduate School of Clemson University In Partial Fulfillment of the Requirements for the Degree Master of Science Computer Engineering by Nicholas L. Mills May 2011 Accepted by: Dr. Walter B. Ligon III, Committee Chair Dr. Richard R. Brooks Dr. Adam W. Hoover Abstract As the size and scale of supercomputers continues to increase at an exponential rate the number of users on a given supercomputer will only grow larger. A larger number of users on a supercomputer places a greater importance on the strength of information security. Nowhere is this requirement for security more apparent than the file system, as users expect their data to be protected from accidental or deliberate modification. In spite of the ever-increasing demand for more secure file system access the majority of parallel file systems do not implement a robust security protocol for fear it will negatively impact the performance and scalability of the file system.
[Show full text]
Privacy Engineering for Social Networks
UCAM-CL-TR-825 Technical Report ISSN 1476-2986 Number 825 Computer Laboratory Privacy engineering for social networks Jonathan Anderson December 2012 15 JJ Thomson Avenue Cambridge CB3 0FD United Kingdom phone +44 1223 763500 http://www.cl.cam.ac.uk/ c 2012 Jonathan Anderson This technical report is based on a dissertation submitted July 2012 by the author for the degree of Doctor of Philosophy to the University of Cambridge, Trinity College. Technical reports published by the University of Cambridge Computer Laboratory are freely available via the Internet: http://www.cl.cam.ac.uk/techreports/ ISSN 1476-2986 Privacy engineering for social networks Jonathan Anderson In this dissertation, I enumerate several privacy problems in online social net- works (OSNs) and describe a system called Footlights that addresses them. Foot- lights is a platform for distributed social applications that allows users to control the sharing of private information. It is designed to compete with the performance of today’s centralised OSNs, but it does not trust centralised infrastructure to en- force security properties. Based on several socio-technical scenarios, I extract concrete technical problems to be solved and show how the existing research literature does not solve them. Addressing these problems fully would fundamentally change users’ interactions with OSNs, providing real control over online sharing. I also demonstrate that today’s OSNs do not provide this control: both user data and the social graph are vulnerable to practical privacy attacks. Footlights’ storage substrate provides private, scalable, sharable storage using untrusted servers. Under realistic assumptions, the direct cost of operating this storage system is less than one US dollar per user-year.
[Show full text]
Operating Systems & Virtualisation Security Knowledge Area
Operating Systems & Virtualisation Security Knowledge Area Issue 1.0 Herbert Bos Vrije Universiteit Amsterdam EDITOR Andrew Martin Oxford University REVIEWERS Chris Dalton Hewlett Packard David Lie University of Toronto Gernot Heiser University of New South Wales Mathias Payer École Polytechnique Fédérale de Lausanne The Cyber Security Body Of Knowledge www.cybok.org COPYRIGHT © Crown Copyright, The National Cyber Security Centre 2019. This information is licensed under the Open Government Licence v3.0. To view this licence, visit: http://www.nationalarchives.gov.uk/doc/open-government-licence/ When you use this information under the Open Government Licence, you should include the following attribution: CyBOK © Crown Copyright, The National Cyber Security Centre 2018, li- censed under the Open Government Licence: http://www.nationalarchives.gov.uk/doc/open- government-licence/. The CyBOK project would like to understand how the CyBOK is being used and its uptake. The project would like organisations using, or intending to use, CyBOK for the purposes of education, training, course development, professional development etc. to contact it at con- [email protected] to let the project know how they are using CyBOK. Issue 1.0 is a stable public release of the Operating Systems & Virtualisation Security Knowl- edge Area. However, it should be noted that a fully-collated CyBOK document which includes all of the Knowledge Areas is anticipated to be released by the end of July 2019. This will likely include updated page layout and formatting of the individual Knowledge Areas KA Operating Systems & Virtualisation Security j October 2019 Page 1 The Cyber Security Body Of Knowledge www.cybok.org INTRODUCTION In this Knowledge Area, we introduce the principles, primitives and practices for ensuring se- curity at the operating system and hypervisor levels.
[Show full text]
Object Capabilities and Isolation of Untrusted Web Applications
Object Capabilities and Isolation of Untrusted Web Applications Sergio Maffeis John C. Mitchell Ankur Taly Imperial College London Stanford University Stanford University [email protected] [email protected] [email protected] Abstract—A growing number of current web sites combine Intel iAPX423, the Amoeba operating system, and others active content (applications) from untrusted sources, as in (see [10], [11]). The main idea is that code possessing a so-called mashups. The object-capability model provides an capability, such as an unforgeable reference to a ﬁle or appealing approach for isolating untrusted content: if separate applications are provided disjoint capabilities, a sound object- system object, is allowed to access the resource by virtue of capability framework should prevent untrusted applications possessing the capability. If a system is capability safe, and from interfering with each other, without preventing interaction a process possesses only the capabilities that it is explicitly with the user or the hosting page. In developing language-based given, then isolation between two untrusted processes may foundations for isolation proofs based on object-capability con- be achieved granting them non-overlapping capabilities. cepts, we identify a more general notion of authority safety that also implies resource isolation. After proving that capability An attractive adaptation to programming language con- safety implies authority safety, we show the applicability of texts is the object-capability model [12], [13], which replaces our framework for a speciﬁc class of mashups. In addition the traditional subject-object dichotomy with programming to proving that a JavaScript subset based on Google Caja language objects that are both subjects that initiate access is capability safe, we prove that a more expressive subset of and objects (targets) of regulated actions.
[Show full text]
Runtime Data Driven Partitioning of Linux Code
This is the author's version of an article that has been published in this journal. Changes were made to this version by the publisher prior to publication. The final version of record is available at http://dx.doi.org/10.1109/TDSC.2017.2745574 1 How to Fillet a Penguin: Runtime Data Driven Partitioning of Linux Code Oliver Schwahn, Stefan Winter, Nicolas Coppik, and Neeraj Suri Abstract—In many modern operating systems (OSs), there exists no isolation between different kernel components, i.e., the failure of one component can affect the whole kernel. While microkernel OSs introduce address space separation for large parts of the OS, their improved fault isolation comes at the cost of performance. Despite signiﬁcant improvements in modern microkernels, monolithic OSs like Linux are still prevalent in many systems. To achieve fault isolation in addition to high performance and code reuse in these systems, approaches to move only fractions of kernel code into user mode have been proposed. These approaches solely rely on static code analyses for deciding which code to isolate, neglecting dynamic properties like invocation frequencies. We propose to augment static code analyses with runtime data to achieve better estimates of dynamic properties for common case operation. We assess the impact of runtime data on the decision what code to isolate and the impact of that decision on the performance of such “microkernelized” systems. We extend an existing tool chain to implement automated code partitioning for existing monolithic kernel code and validate our approach in a case study of two widely used Linux device drivers and a ﬁle system.
[Show full text]
Capsicum Slides
Capsicum Capability-Based Sandboxing David Drysdale Google UK Features Current LXC uses the following kernel features to contain processes: ● Kernel namespaces (ipc, uts, mount, pid, network and user) ● Apparmor and SELinux profiles ● Seccomp policies ● Chroots (using pivot_root) ● Kernel capabilities ● CGroups (control groups) Agenda ● Ideas ○ Privilege Separation ○ Capabilities ● Capsicum ○ Hybrid with POSIX ○ Application changes ● Linux container features ● Status/Outlook Check Your Privileges ● Drop unnecessary privileges ○ Just because a process starts as root, doesn't have to stay that way Check Your Privileges ● Drop unnecessary privileges ○ Just because a process starts as root, doesn't have to stay that way ● Divide up software according to what privileges are needed ○ E.g. separate media processing from credentials processing Check Your Privileges ● Drop unnecessary privileges ○ Just because a process starts as root, doesn't have to stay that way ● Divide up software according to what privileges are needed ○ E.g. separate media processing from credentials processing ● Examples: ○ OpenSSH: credential checking process ○ Chrome: renderer processes ● Design impact: ○ Do privileged operations first ○ Pass resources down a privilege gradient Capability-Based Security ● Make the privileges that a process holds more explicit Capability-Based Security ● Make the privileges that a process holds more explicit ● Access objects via unforgeable token: the capability ○ Identifies the object ○ Accompanying rights give allowed operations ○ Can only reduce, not increase rights ○ Can pass capabilities around Capability-Based Security ● Make the privileges that a process holds more explicit ● Access objects via unforgeable token: the capability ○ Identifies the object ○ Accompanying rights give allowed operations ○ Can only reduce, not increase rights ○ Can pass capabilities around ● Remove other ways of accessing objects ○ No access by name, i.e.
[Show full text]
On Access Checking in Capability-Based Systems1
On Access Checking in Capability-Based Systems1 Richard Y. Kain Carl E. Landwehr University of Minnesota Naval Research Laboratory ABSTRACT Public descriptions of capability-based system designs often do not clarify the necessary details concerning the propagation of access rights within the sys- tems. A casual reader may assume that it is adequate for capabilities to be passed in accordance with the rules for data copying. A system using such a rule cannot enforce either the military security policy or the Bell and LaPadula rules. The paper shows why this problem arises and provides a taxonomy of capability-based designs. Within the space of design options de®ned by the tax- onomy we identify a class of designs that cannot enforce the Bell-LaPadula rules and two designs that do allow their enforcement. Index Terms--Access control, capabilities, capability-based architectures, secu- rity policy, *-property, taxonomy. 1. Introduction Capability systems were ®rst described in the literature in the mid-1960's. Their informal descriptions are typically based upon the notion that a capability is equivalent to a ``ticket,'' in the sense that possession of the ticket allows the possessing process access to the object described in the capability, provided that the access mode is compatible with the ``access rights'' stored within the capability. Several systems using the capability concept have been marketed (IBM System 38, CAP, i432, Plessey S250) [1]. Whether a computer system based upon capabilities can provably enforce the DoD security policy [2] has been a matter of discussion for some time. Boebert [3] has argued that an ``unmodi®ed'' capability machine must be incapable of enforcing the *-property de®ned by Bell and LaPadula [4].
[Show full text]
Semperos: a Distributed Capability System
SemperOS: A Distributed Capability System Matthias Hille† Nils Asmussen† ∗ Pramod Bhatotia‡ Hermann Härtig† ∗ †Technische Universität Dresden ‡The University of Edinburgh ∗ Barkhausen Institut Abstract systems have received renewed attention recently to provide Capabilities provide an efficient and secure mechanism for an efficient and secure mechanism for resource management fine-grained resource management and protection. However, in modern hardware architectures [5, 24, 30, 36, 44, 64, 67]. as the modern hardware architectures continue to evolve Today the main improvements in compute capacity with large numbers of non-coherent and heterogeneous cores, are achieved by either adding more cores or integrating we focus on the following research question: can capability accelerators into the system. However, the increasing core systems scale to modern hardware architectures? counts exacerbate the hardware complexity required for global In this work, we present a scalable capability system to drive cache coherence. While on-chip cache coherence is likely to future systems with many non-coherent heterogeneous cores. remain a feature of future hardware architectures [45], we see More specifically, we have designed a distributed capability characteristics of distributed systems added to the hardware system based on a HW/SW co-designed capability system. by giving up on global cache coherence across a whole We analyzed the pitfalls of distributed capability operations machine [6, 28]. Additionally, various kinds of accelerators running concurrently and built the protocols in accordance are added like the Xeon Phi Processor, the Matrix-2000 accel- with the insights. We have incorporated these distributed erator, GPUs, FPGAs, or ASICs, which are used in numerous capability management protocols in a new microkernel-based application fields [4,21,32,34,43,60].
[Show full text]