DOCSLIB.ORG

Intel(R) Software Guard Extensions Developer Reference for Linux* OS

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Intel(R) Software Guard Extensions Developer Reference for Linux* OS Intel® Software Guard Extensions (Intel® SGX) SDK for Linux* OS Developer Reference Intel® Software Guard Extensions Developer Reference for Linux* OS Legal Information No license (express or implied, by estoppel or otherwise) to any intellectual prop- erty rights is granted by this document. Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non- infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade. This document contains information on products, services and/or processes in development. All information provided here is subject to change without notice. Contact your Intel representative to obtain the latest forecast, schedule, spe- cifications and roadmaps. The products and services described may contain defects or errors known as errata which may cause deviations from published specifications. Current characterized errata are available on request. Intel technologies features and benefits depend on system configuration and may require enabled hardware, software or service activation. Learn more at Intel.com, or from the OEM or retailer. Copies of documents which have an order number and are referenced in this doc- ument may be obtained by calling 1-800-548-4725 or by visiting www.in- tel.com/design/literature.htm. Intel, the Intel logo, VTune, Xeon, and Xeon Phi are trademarks of Intel Corporation in the U.S. and/or other countries. Optimization Notice Intel's compilers may or may not optimize to the same degree for non-Intel micro- processors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2, SSE3, and SSSE3 instruction sets and other optim- izations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on microprocessors not manufactured by Intel. Microprocessor- dependent optimizations in this product are intended for use with Intel micro- processors. Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and Reference Guides for more information regarding the specific instruction sets covered by this notice. Notice revision #20110804 * Other names and brands may be claimed as the property of others. Copyright 2014-2019 Intel Corporation. - 2 - Intel® Software Guard Extensions Developer Reference for Linux* OS This software and the related documents are Intel copyrighted materials, and your use of them is governed by the express license under which they were provided to you (License). Unless the License provides otherwise, you may not use, modify, copy, publish, distribute, disclose or transmit this software or the related documents without Intel's prior written permission. This software and the related documents are provided as is, with no express or implied warranties, other than those that are expressly stated in the License. - 3 - Intel® Software Guard Extensions Developer Reference for Linux* OS Revision History Revision Number Description Revision Date 1.5 Intel® SGX Linux 1.5 release May 2016 1.6 Intel® SGX Linux 1.6 release September 2016 1.7 Intel® SGX Linux 1.7 release December 2016 1.8 Intel® SGX Linux 1.8 release March 2017 1.9 Intel® SGX Linux 1.9 release July 2017 2.0 Intel® SGX Linux 2.0 release November 2017 2.1 Intel® SGX Linux 2.1 release December 2017 2.1.1 Intel® SGX Linux 2.1.1 release March 2018 2.1.2 Intel® SGX Linux 2.1.2 release March 2018 2.1.3 Intel® SGX Linux 2.1.3 release April 2018 2.2 Intel® SGX Linux 2.2 release July 2018 2.3 Intel® SGX Linux 2.3 release September 2018 2.4 Intel® SGX Linux 2.4 release November 2018 2.5 Intel® SGX Linux 2.5 release March 2019 2.6 Intel® SGX Linux 2.6 release June 2019 2.7 Intel® SGX Linux 2.7 release September 2019 2.7.1 Intel® SGX Linux 2.7.1 release November 2019 2.8 Intel® SGX Linux 2.8 release January 2020 - 4 - Intel® Software Guard Extensions Developer Reference for Linux* OS Introduction Intel provides the Intel® Software Guard Extensions (Intel® SGX) SDK Developer Reference for software developers who wish to harden their applic- ation security using the Intel Software Guard Extensions technology. This document covers an overview of the technology, tutorials, tools, sample code as well as an API reference. Intel® Software Guard Extensions SDK is a collection of APIs, sample source code, libraries, and tools that enable the software developer to write and debug Intel® Software Guard Extensions applications in C/C++ programming language. NOTE Intel® Software Guard Extensions (Intel® SGX) technology is only available on the 6th Generation Intel® Core(TM) Processor or newer. Intel® Software Guard Extensions Technology Overview Intel® Software Guard Extensions is an Intel technology whose objective is to enable a high-level protection of secrets. It operates by allocating hardware- protected memory where code and data reside. The protected memory area is called an enclave. Data within the enclave memory can only be accessed by the code that also resides within the enclave memory space. Enclave code can be invoked via special instructions. An enclave can be built and loaded as a shared object on Linux* OS. NOTE: The enclave file can be disassembled, so the algorithms used by the enclave developer will not remain secret. Intel® Software Guard Extensions technology has a hard limit on the protected memory size, typically 64 MB or 128 MB. As a result, the number of active enclaves (in memory) is limited. Depending on the memory footprint of each enclave, use cases suggest that 5-20 enclaves can reside in memory sim- ultaneously. Linux*, however, can increase the protected memory size through paging. Intel® Software Guard Extensions Security Properties l Intel designs the Intel® Software Guard Extensions to protect against soft- ware attacks: - 5 - Intel® Software Guard Extensions Developer Reference for Linux* OS o The enclave memory cannot be read or written from outside the enclave regardless of current privilege level and CPU mode (ring3/user-mode, ring0/kernel-mode, SMM, VMM, or another enclave). The abort page is returned in such conditions. o An enclave can be created with a debug attribute that allows a spe- cial debugger (Intel® Software Guard Extensions debugger) to view its content like a standard debugger. Production enclaves (non- debug) cannot be debugged by software or hardware debuggers. o The enclave environment cannot be entered via classic function calls, jumps, register manipulation or stack manipulation. The only way to call an enclave function is via a new instruction that per- forms several protect checks. Classic function calls initiated by enclave code to functions inside the enclave are allowed. o CPU mode can only be 32 or 64 bit when executing enclave code. Other CPU modes are not supported. An exception is raised in such conditions. l Intel designs the Intel® Software Guard Extensions to protect against known hardware attacks: o The enclave memory is encrypted using industry-standard encryp- tion algorithms with replay protection. o Tapping the memory or connecting the DRAM modules to another system will only give access to encrypted data. o The memory encryption key changes every power cycle randomly (for example, boot/sleep/hibernate). The key is stored within the CPU and it is not accessible. o Intel® Software Guard Extensions is not designed to handle side channel attacks or reverse engineering. It is up to the Intel® SGX developers to build enclaves that are protected against these types of attack. Intel® Software Guard Extensions uses strong industry-standard algorithms for signing enclaves. The signature of an enclave characterizes the content and the layout of the enclave at build time. If the enclave’s content and layout are not correct per the signature, then the enclave will fail to be initialized and, hence, will not be executed. If an enclave is initialized, it should be identical to the original enclave and will not be modified at runtime. - 6 - Intel® Software Guard Extensions Developer Reference for Linux* OS Application Design Considerations An Intel® Software Guard Extensions application design is different from non- Intel® SGX application as it requires dividing the application into two logical components: l Trusted component. The code that accesses the secret resides here. This component is also called an enclave. More than one enclave can exist in an application. l Untrusted component. The rest of the application including all its mod- ules.1 The application writer should make the trusted part as small as possible. It is suggested that enclave functionality should be limited to operate on the secret data. A large enclave statistically has more bugs and (user created) security holes than a small enclave. The enclave code can leave the protected memory region and call functions in the untrusted zone (by a special instruction). Reducing the enclave depend- ency on untrusted code will also strengthen its protection against possible attacks. Embracing the above design considerations will improve protection as the attack surface is minimized. The application designer, as the first step to harnessing Intel® Software Guard Extensions SDK in the application, must redesign or refactor the application to fit these guidelines. This is accomplished by isolating the code module(s) that access any secrets and then moving these modules to a separate pack- age/library. The details of how to create such an enclave are detailed in the tutorials section. You can also see the demonstrations on creating an enclave in the sample code that are shipped with the Intel® Software Guard Extensions SDK. Terminology and Acronyms AE Architectural enclaves. Enclaves that are part of the Intel® Soft- ware Guard Extensions framework. They include the quoting 1From an enclave standpoint, the operating system and VMM are not trusted components, either.
Load more

Recommended publications
  • Fencing Cyberspace: Drawing Borders in a Virtual World Maureen A
    University of Minnesota Law School Scholarship Repository Minnesota Law Review 1998 Fencing Cyberspace: Drawing Borders in a Virtual World Maureen A. O'Rourke Follow this and additional works at: https://scholarship.law.umn.edu/mlr Part of the Law Commons Recommended Citation O'Rourke, Maureen A., "Fencing Cyberspace: Drawing Borders in a Virtual World" (1998). Minnesota Law Review. 1923. https://scholarship.law.umn.edu/mlr/1923 This Article is brought to you for free and open access by the University of Minnesota Law School. It has been accepted for inclusion in Minnesota Law Review collection by an authorized administrator of the Scholarship Repository. For more information, please contact [email protected]. Fencing Cyberspace: Drawing Borders in a Virtual World Maureen A. O'Rourke* Introduction ............................................................................... 610 I. An Introduction to the Internet and the World W ide W eb .................................................................. 615 A. Origins of the Internet ................................................... 615 B. Development of the World Wide Web ........................... 619 C. Emergence of the Internet as a Commercial Marketplace ............................................... 624 1. Development of the Marketplace .............................. 624 2. Web Business Models ................................................ 625 a. Advertising-Based Models ................................... 626 b. Subscription-Based Models .................................
    [Show full text]
  • Intel® IA-64 Architecture Software Developer's Manual
    Intel® IA-64 Architecture Software Developer’s Manual Volume 1: IA-64 Application Architecture Revision 1.1 July 2000 Document Number: 245317-002 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, SPECIFICATION OR SAMPLE. Information in this document is provided in connection with Intel products. No license, express or implied, by estoppel or otherwise, to any intellectual property rights is granted by this document. Except as provided in Intel's Terms and Conditions of Sale for such products, Intel assumes no liability whatsoever, and Intel disclaims any express or implied warranty, relating to sale and/or use of Intel products including liability or warranties relating to fitness for a particular purpose, merchantability, or infringement of any patent, copyright or other intellectual property right. Intel products are not intended for use in medical, life saving, or life sustaining applications. Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined." Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. Intel® IA-64 processors may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request. Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling 1-800- 548-4725, or by visiting Intel’s website at http://developer.intel.com/design/litcentr.
    [Show full text]
  • Survey of Methodologies, Approaches, and Challenges in Parallel Programming Using High-Performance Computing Systems
    Hindawi Scientific Programming Volume 2020, Article ID 4176794, 19 pages https://doi.org/10.1155/2020/4176794 Review Article Survey of Methodologies, Approaches, and Challenges in Parallel Programming Using High-Performance Computing Systems Paweł Czarnul ,1 Jerzy Proficz,2 and Krzysztof Drypczewski 2 1Dept. of Computer Architecture, Faculty of Electronics, Telecommunications and Informatics, Gdansk University of Technology, Gda´nsk, Poland 2Centre of Informatics–Tricity Academic Supercomputer & Network (CI TASK), Gdansk University of Technology, Gdan´sk, Poland Correspondence should be addressed to Paweł Czarnul; [email protected] Received 11 October 2019; Accepted 30 December 2019; Published 29 January 2020 Guest Editor: Pedro Valero-Lara Copyright © 2020 Paweł Czarnul et al. ,is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. ,is paper provides a review of contemporary methodologies and APIs for parallel programming, with representative tech- nologies selected in terms of target system type (shared memory, distributed, and hybrid), communication patterns (one-sided and two-sided), and programming abstraction level. We analyze representatives in terms of many aspects including programming model, languages, supported platforms, license, optimization goals, ease of programming, debugging, deployment, portability, level of parallelism, constructs enabling parallelism and synchronization, features introduced in recent versions indicating trends, support for hybridity in parallel execution, and disadvantages. Such detailed analysis has led us to the identification of trends in high-performance computing and of the challenges to be addressed in the near future. It can help to shape future versions of programming standards, select technologies best matching programmers’ needs, and avoid potential difficulties while using high- performance computing systems.
    [Show full text]
  • SIMD Extensions
    SIMD Extensions PDF generated using the open source mwlib toolkit. See http://code.pediapress.com/ for more information. PDF generated at: Sat, 12 May 2012 17:14:46 UTC Contents Articles SIMD 1 MMX (instruction set) 6 3DNow! 8 Streaming SIMD Extensions 12 SSE2 16 SSE3 18 SSSE3 20 SSE4 22 SSE5 26 Advanced Vector Extensions 28 CVT16 instruction set 31 XOP instruction set 31 References Article Sources and Contributors 33 Image Sources, Licenses and Contributors 34 Article Licenses License 35 SIMD 1 SIMD Single instruction Multiple instruction Single data SISD MISD Multiple data SIMD MIMD Single instruction, multiple data (SIMD), is a class of parallel computers in Flynn's taxonomy. It describes computers with multiple processing elements that perform the same operation on multiple data simultaneously. Thus, such machines exploit data level parallelism. History The first use of SIMD instructions was in vector supercomputers of the early 1970s such as the CDC Star-100 and the Texas Instruments ASC, which could operate on a vector of data with a single instruction. Vector processing was especially popularized by Cray in the 1970s and 1980s. Vector-processing architectures are now considered separate from SIMD machines, based on the fact that vector machines processed the vectors one word at a time through pipelined processors (though still based on a single instruction), whereas modern SIMD machines process all elements of the vector simultaneously.[1] The first era of modern SIMD machines was characterized by massively parallel processing-style supercomputers such as the Thinking Machines CM-1 and CM-2. These machines had many limited-functionality processors that would work in parallel.
    [Show full text]
  • The Microarchitecture of the Pentium 4 Processor
    The Microarchitecture of the Pentium 4 Processor Glenn Hinton, Desktop Platforms Group, Intel Corp. Dave Sager, Desktop Platforms Group, Intel Corp. Mike Upton, Desktop Platforms Group, Intel Corp. Darrell Boggs, Desktop Platforms Group, Intel Corp. Doug Carmean, Desktop Platforms Group, Intel Corp. Alan Kyker, Desktop Platforms Group, Intel Corp. Patrice Roussel, Desktop Platforms Group, Intel Corp. Index words: Pentium® 4 processor, NetBurst™ microarchitecture, Trace Cache, double-pumped ALU, deep pipelining provides an in-depth examination of the features and ABSTRACT functions of the Intel NetBurst microarchitecture. This paper describes the Intel® NetBurst™ ® The Pentium 4 processor is designed to deliver microarchitecture of Intel’s new flagship Pentium 4 performance across applications where end users can truly processor. This microarchitecture is the basis of a new appreciate and experience its performance. For example, family of processors from Intel starting with the Pentium it allows a much better user experience in areas such as 4 processor. The Pentium 4 processor provides a Internet audio and streaming video, image processing, substantial performance gain for many key application video content creation, speech recognition, 3D areas where the end user can truly appreciate the applications and games, multi-media, and multi-tasking difference. user environments. The Pentium 4 processor enables real- In this paper we describe the main features and functions time MPEG2 video encoding and near real-time MPEG4 of the NetBurst microarchitecture. We present the front- encoding, allowing efficient video editing and video end of the machine, including its new form of instruction conferencing. It delivers world-class performance on 3D cache called the Execution Trace Cache.
    [Show full text]
  • SAP Solutions on Vmware Vsphere Guidelines Summary and Best Practices
    SAP On VMware Best Practices Version 1.1 December 2015 © 2015 VMware, Inc. All rights reserved. Page 1 of 61 SAP on VMware Best Practices © 2015 VMware, Inc. All rights reserved. This product is protected by U.S. and international copyright and intellectual property laws. This product is covered by one or more patents listed at http://www.vmware.com/download/patents.html. VMware is a registered trademark or trademark of VMware, Inc. in the United States and/or other jurisdictions. All other marks and names mentioned herein may be trademarks of their respective companies. VMware, Inc. 3401 Hillview Ave Palo Alto, CA 94304 www.vmware.com © 2015 VMware, Inc. All rights reserved. Page 2 of 61 SAP on VMware Best Practices Contents 1. Overview .......................................................................................... 5 2. Introduction ...................................................................................... 6 2.1 Support ........................................................................................................................... 6 2.2 SAP ................................................................................................................................ 6 2.3 vCloud Suite ................................................................................................................... 9 2.4 Deploying SAP with vCloud Suite ................................................................................ 11 3. Virtualization Overview ..................................................................
    [Show full text]
  • Oracle Solaris and Oracle SPARC Systems—Integrated and Optimized for Mission Critical Computing
    An Oracle White Paper September 2010 Oracle Solaris and Oracle SPARC Servers— Integrated and Optimized for Mission Critical Computing Oracle Solaris and Oracle SPARC Systems—Integrated and Optimized for Mission Critical Computing Executive Overview ............................................................................. 1 Introduction—Oracle Datacenter Integration ....................................... 1 Overview ............................................................................................. 3 The Oracle Solaris Ecosystem ........................................................ 3 SPARC Processors ......................................................................... 4 Architected for Reliability ..................................................................... 7 Oracle Solaris Predictive Self Healing ............................................ 7 Highly Reliable Memory Subsystems .............................................. 9 Oracle Solaris ZFS for Reliable Data ............................................ 10 Reliable Networking ...................................................................... 10 Oracle Solaris Cluster ................................................................... 11 Scalable Performance ....................................................................... 14 World Record Performance ........................................................... 16 Sun FlashFire Storage .................................................................. 19 Network Performance ..................................................................
    [Show full text]
  • Cluster Suite Overview
    Red Hat Enterprise Linux 4 Cluster Suite Overview Red Hat Cluster Suite for Red Hat Enterprise Linux Edition 1.0 Last Updated: 2020-03-08 Red Hat Enterprise Linux 4 Cluster Suite Overview Red Hat Cluster Suite for Red Hat Enterprise Linux Edition 1.0 Landmann [email protected] Legal Notice Copyright © 2009 Red Hat, Inc. This document is licensed by Red Hat under the Creative Commons Attribution-ShareAlike 3.0 Unported License. If you distribute this document, or a modified version of it, you must provide attribution to Red Hat, Inc. and provide a link to the original. If the document is modified, all Red Hat trademarks must be removed. Red Hat, as the licensor of this document, waives the right to enforce, and agrees not to assert, Section 4d of CC-BY-SA to the fullest extent permitted by applicable law. Red Hat, Red Hat Enterprise Linux, the Shadowman logo, the Red Hat logo, JBoss, OpenShift, Fedora, the Infinity logo, and RHCE are trademarks of Red Hat, Inc., registered in the United States and other countries. Linux ® is the registered trademark of Linus Torvalds in the United States and other countries. Java ® is a registered trademark of Oracle and/or its affiliates. XFS ® is a trademark of Silicon Graphics International Corp. or its subsidiaries in the United States and/or other countries. MySQL ® is a registered trademark of MySQL AB in the United States, the European Union and other countries. Node.js ® is an official trademark of Joyent. Red Hat is not formally related to or endorsed by the official Joyent Node.js open source or commercial project.
    [Show full text]
  • 6Th Gen Intel® Core™ Processors
    6th Generation Intel® Processor Family Specification Update Supporting the Intel® Pentium® Processor Family based on the U-Processor Supporting the 6th Generation Intel® Core™ Processor Family based on the Y-Processor September 2015 Version 1.0 Order Number: 332994-001EN Preface You may not use or facilitate the use of this document in connection with any infringement or other legal analysis concerning Intel products described herein. You agree to grant Intel a non-exclusive, royalty-free license to any patent claim thereafter drafted which includes subject matter disclosed herein. No license (express or implied, by estoppel or otherwise) to any intellectual property rights is granted by this document. Intel technologies’ features and benefits depend on system configuration and may require enabled hardware, software or service activation. Performance varies depending on system configuration. No computer system can be absolutely secure. Check with your system manufacturer or retailer or learn more at intel.com. Intel technologies may require enabled hardware, specific software, or services activation. Check with your system manufacturer or retailer. The products described may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request. Intel disclaims all express and implied warranties, including without limitation, the implied warranties of merchantability, fitness for a particular purpose, and non-infringement, as well as any warranty arising from course of performance, course of dealing, or usage in trade. All information provided here is subject to change without notice. Contact your Intel representative to obtain the latest Intel product specifications and roadmaps Copies of documents which have an order number and are referenced in this document may be obtained by calling 1-800-548-4725 or visit www.intel.com/design/literature.htm.
    [Show full text]
  • Arxiv:1312.1411V2 [Cs.LO] 9 Jun 2014 Ocretporm R Adt Einadipeet Seilyw Especially Implement, Buffering and Implement Design Multiprocessors to Tectures
    Don’t sit on the fence A static analysis approach to automatic fence insertion Jade Alglave1, Daniel Kroening2, Vincent Nimal2, and Daniel Poetzl2 1 University College London, UK 2University of Oxford, UK Abstract Modern architectures rely on memory fences to prevent undesired weakenings of memory consistency. As the fences’ semantics may be subtle, the automation of their placement is highly desirable. But precise methods for restoring consistency do not scale to deployed systems code. We choose to trade some precision for genuine scalability: our technique is suitable for large code bases. We implement it in our new musketeer tool, and detail experiments on more than 350 executables of packages found in Debian Linux 7.1, e.g. memcached (about 10000 LoC). 1 Introduction Concurrent programs are hard to design and implement, especially when running on multiprocessor archi- tectures. Multiprocessors implement weak memory models, which feature e.g. instruction reordering, store buffering (both appearing on x86), or store atomicity relaxation (a particularity of Power and ARM). Hence, multiprocessors allow more behaviours than Lamport’s Sequential Consistency (SC) [Lam79], a theoretical model where the execution of a program corresponds to an interleaving of the different threads. This has a dramatic effect on programmers, most of whom learned to program with SC. Fortunately, architectures provide special fence (or barrier) instructions to prevent certain behaviours. Yet both the questions of where and how to insert fences are contentious, as fences are architecture-specific and expensive. Attempts at automatically placing fences include Visual Studio 2013, which offers an option to guarantee acquire/release semantics (we study the performance impact of this policy in Sec.
    [Show full text]
  • High Performance Virtual Machine Recovery in the Cloud
    High Performance Virtual Machine Recovery in the Cloud Valentina Salapura1 and Richard Harper2 1IBM T. J. Watson Research Center, 1101 Kitchawan Rd, NY, Yorktown Heights, U.S.A. 2IBM T. J. Watson Research Center, Research Triangle Park, NC, U.S.A. Keywords: Cloud Computing, High Availability, Virtualization, Automation, Enterprise Class. Abstract: In this paper, we outline and illustrate concepts that are essential to achieve fast, highly scalable virtual machine planning and failover at the Virtual Machine (VM) level in a data center containing a large number of servers, VMs, and disks. To illustrate the concepts a solution is implemented and analyzed for IBM’s Cloud Managed Services enterprise cloud. The solution enables at-failover-time planning, and keeps the recovery time within tight service level agreement (SLA) allowed time budgets via parallelization of recovery activities. The initial serial failover time was reduced for an order of magnitude due to parallel VM restart, and to parallel VM restart combined with parallel storage device remapping. 1 INTRODUCTION originally designed for smaller managed environments, and do not scale well as the system Cloud computing is being rapidly adopted across the size and complexity increases. Detecting failures, IT industry as a platform for increasingly more determining appropriate failover targets, re-mapping demanding workloads, both traditional and a new storage to those failover targets, and restarting the generation of mobile, social and analytics virtual workload have to be carefully designed and applications. In the cloud, customers are being led parallelized in order to meet the service level to expect levels of availability that until recently agreement (SLA) for large systems.
    [Show full text]
  • Floating-Point on X86-64
    Floating-Point on x86-64 Sixteen registers: %xmm0 through %xmm15 • float or double arguments in %xmm0 – %xmm7 • float or double result in %xmm0 • %xmm8 – %xmm15 are temporaries (caller-saved) Two operand sizes: • single-precision = 32 bits = float • double-precision = 64 bits = double ��� Arithmetic Instructions addsx source, dest subsx source, dest mulsx source, dest divsx source, dest x is either s or d Add doubles addsd %xmm0, %xmm1 Multiply floats mulss %xmm0, %xmm1 3 Conversion cvtsx2sx source, dest cvttsx2sx source, dest x is either s, d, or i With i, add an extra extension for l or q Convert a long to a double cvtsi2sdq %rdi, %xmm0 Convert a float to a int cvttss2sil %xmm0, %eax 4 Example Floating-Point Compilation double scale(double a, int b) { return b * a; } cvtsi2sdl %edi, %xmm1 mulsd %xmm1, %xmm0 ret 5 SIMD Instructions addpx source, dest subpx source, dest mulpx source, dest divpx source, dest Combine pairs of doubles or floats ... because registers are actually 128 bits wide Add two pairs of doubles addpd %xmm0, %xmm1 Multiply four pairs of floats mulps %xmm0, %xmm1 6 Auto-Vectorization void mult_all(double a[4], double b[4]) { a[0] = a[0] * b[0]; a[1] = a[1] * b[1]; a[2] = a[2] * b[2]; a[3] = a[3] * b[3]; } • What if a and b are alises? • What if a or b is not 16-byte aligned? ��� Auto-Vectorization void mult_all(double * __restrict__ ai, double * __restrict__ bi) { double *a = __builtin_assume_aligned(ai, 16); double *b = __builtin_assume_aligned(bi, 16); a[0] = a[0] * b[0]; a[1] = a[1] * b[1]; a[2] = a[2] * b[2]; a[3] = a[3] * b[3]; } movapd 16(%rdi), %xmm0 movapd (%rdi), %xmm1 mulpd 16(%rsi), %xmm0 -O3 mulpd (%rsi), %xmm1 movapd %xmm0, 16(%rdi) gcc movapd %xmm1, (%rdi) ret ���� History: Floating-Point Support in x86 8086 • No foating-point hardware • Software can implement IEEE arithmatic by manipulating bits, but that’s slow 8087 (a.k.a.
    [Show full text]