DOCSLIB.ORG

Altivec Technology Programming Interface Manual MOTOROLA

Total Page：16

File Type：pdf, Size：1020Kb

Download full-text PDF Read full-text
Altivec Technology Programming Interface Manual MOTOROLA Freescale Semiconductor, Inc. ALTIVECPIM/D 6/1999 Rev. 0 . c n ™ I AltiVec Technology , r o Programming Interface Manual t c u d n o c i m e S e l a c s e e r F ™ For More Information On This Product, Go to: www.freescale.com Freescale Semiconductor, Inc. c n I , r o t c u d DigitalDNA and Mfax are trademarks of Motorola, Inc. n The PowerPC name and the PowerPC logotype are trademarks of International Business Machines Corporation used by Motorola under license from o International Business Machines Corporation. c i This document contains information on a new product under development. Motorola reserves the right to change or discontinue this product without notice. m Information in this document is provided solely to enable system and software implementers to use PowerPC microprocessors. There are no express or implied copyright licenses granted hereunder to design or fabricate PowerPC integrated circuits or integrated circuits based on the information in this e document. S Motorola reserves the right to make changes without further notice to any products herein. Motorola makes no warranty, representation or guarantee regarding the suitability of its products for any particular purpose, nor does Motorola assume any liability arising out of the application or use of any product or circuit, and specifically disclaims any and all liability, including without limitation consequential or incidental damages. “Typical” parameters can and do e vary in different applications. All operating parameters, including “Typicals” must be validated for each customer application by customer’s technical l experts. Motorola does not convey any license under its patent rights nor the rights of others. Motorola products are not designed, intended, or authorized a for use as components in systems intended for surgical implant into the body, or other applications intended to support or sustain life, or for any other application in which the failure of the Motorola product could create a situation where personal injury or death may occur. Should Buyer purchase or use c Motorola products for any such unintended or unauthorized application, Buyer shall indemnify and hold Motorola and its officers, employees, subsidiaries, affiliates, and distributors harmless against all claims, costs, damages, and expenses, and reasonable attorney fees arising out of, directly or indirectly, s any claim of personal injury or death associated with such unintended or unauthorized use, even if such claim alleges that Motorola was negligent e regarding the design or manufacture of the part. Motorola and are registered trademarks of Motorola, Inc. Motorola, Inc. is an Equal Opportunity/ Affirmative Action Employer. e r Motorola Literature Distribution Centers: USA/EUROPE: Motorola Literature Distribution; P.O. Box 5405; Denver, Colorado 80217; Tel.: 1-800-441-2447 or 1-303-675-2140/ F JAPAN: Nippon Motorola Ltd SPD, Strategic Planning Office 4-32-1, Nishi-Gotanda Shinagawa-ku, Tokyo 141, Japan Tel.: 81-3-5487-8488 ASIA/PACIFC: Motorola Semiconductors H.K. Ltd.; 8B Tai Ping Industrial Park, 51 Ting Kok Road, Tai Po, N.T., Hong Kong; Tel.: 852-26629298 Mfaxª: [email protected]; TOUCHTONE 1-602-244-6609; US & Canada ONLY (800) 774-1848; World Wide Web Address: http://sps.motorola.com/mfax INTERNET: http://motorola.com/sps Technical Information: Motorola Inc. SPS Customer Support Center 1-800-521-6274; electronic mail address: [email protected]. Document Comments: FAX (512) 895-2638, Attn: RISC Applications Engineering. World Wide Web Addresses: http://www.mot.com/PowerPC http://www.mot.com/netcomm http://www.mot.com/HPESD © Motorola Inc. 1999. All rights reserved. For More Information On This Product, Go to: www.freescale.com Freescale Semiconductor, Inc. Overview 1 High-Level Language Interface 2 Application Binary Interface 3 . c AltiVec Operations and Predicates 4 n I , r o t c AltiVec Instruction Set/Operations/Predicates Cross-Reference A u d n Glossary of Terms and Abbreviations GLO o c i m Index IND e S e l a c s e e r F For More Information On This Product, Go to: www.freescale.com Freescale Semiconductor, Inc. 1 Overview 2 High-Level Language Interface 3 Application Binary Interface . 4 AltiVec Operations and Predicates c n I , r o t A AltiVec Instruction Set/Operations/Predicates Cross-Reference c u d GLO Glossary of Terms and Abbreviations n o c i IND Index m e S e l a c s e e r F For More Information On This Product, Go to: www.freescale.com Freescale Semiconductor, Inc. CONTENTS Paragraph Page Title Number Number Audience .............................................................................................................. xvi Organization......................................................................................................... xvi Suggested Reading.............................................................................................. xvii PowerPC Documentation................................................................................ xvii General Information....................................................................................... xviii . Chapter 1 . Overview c n 1.1 High-Level Language Interface ........................................................................... 1-1 I 1.2 Application Binary Interface (ABI)..................................................................... 1-2 , r o Chapter 2 t High-Level Language Interface c u 2.1 Data Types ........................................................................................................... 2-1 d 2.2 New Keywords..................................................................................................... 2-2 2.2.1 The Keyword and Predefine Method............................................................... 2-2 n 2.2.2 The Context Sensitive Keyword Method......................................................... 2-3 o 2.3 Alignment ............................................................................................................ 2-3 c i 2.3.1 Alignment of Vector Types ............................................................................. 2-3 2.3.2 Alignment of Non-Vector Types ..................................................................... 2-3 m 2.3.3 Alignment of Aggregates and Unions Containing Vector Types.................... 2-3 e 2.4 Extensions of C/C++ Operators for the New Types............................................ 2-4 S 2.4.1 sizeof() ............................................................................................................. 2-4 2.4.2 Assignment ...................................................................................................... 2-4 e l 2.4.3 Address Operator ............................................................................................. 2-4 a 2.4.4 Pointer Arithmetic............................................................................................ 2-4 c 2.4.5 Pointer Dereferencing...................................................................................... 2-4 s 2.4.6 Type Casting.................................................................................................... 2-5 e 2.5 New Operators ..................................................................................................... 2-5 e 2.5.1 Vector Literals ................................................................................................. 2-5 r 2.5.2 Vector Literals and Casts................................................................................. 2-6 F 2.5.3 Value for Adjusting Pointers ........................................................................... 2-7 2.5.4 New Operators Representing AltiVec Operations........................................... 2-7 2.6 Programming Interface ........................................................................................ 2-8 Chapter 3 Application Binary Interface (ABI) 3.1 Data Representation............................................................................................. 3-1 3.2 Register Usage Conventions................................................................................ 3-1 MOTOROLA Contents v For More Information On This Product, Go to: www.freescale.com Freescale Semiconductor, Inc. CONTENTS Paragraph Page Title Number Number 3.3 The Stack Frame .................................................................................................. 3-2 3.3.1 SVR4 ABI and EABI Stack Frame.................................................................. 3-3 3.3.2 Apple Macintosh ABI and AIX ABI Stack Frame.......................................... 3-5 3.3.3 Vector Register Saving and Restoring Functions............................................ 3-7 3.4 Function Calls ...................................................................................................... 3-9 3.4.1 SVR4 ABI and EABI Parameter Passing and Varargs.................................... 3-9 3.4.2 Apple Macintosh ABI and AIX ABI Parameter Passing without Varargs...... 3-9 . 3.4.3 Apple Macintosh ABI and AIX ABI Parameter Passing with Varargs......... 3-10 . 3.5 malloc(), vec_malloc(), and new ....................................................................... 3-10 c 3.6 setjmp() and longjmp() ...................................................................................... 3-11 n 3.7 Debugging Information...................................................................................... 3-11 I 3.8 printf() and scanf() Control Strings...................................................................
Load more

Recommended publications
  • Vxworks Architecture Supplement, 6.2
    VxWorks Architecture Supplement VxWorks® ARCHITECTURE SUPPLEMENT 6.2 Copyright © 2005 Wind River Systems, Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means without the prior written permission of Wind River Systems, Inc. Wind River, the Wind River logo, Tornado, and VxWorks are registered trademarks of Wind River Systems, Inc. Any third-party trademarks referenced are the property of their respective owners. For further information regarding Wind River trademarks, please see: http://www.windriver.com/company/terms/trademark.html This product may include software licensed to Wind River by third parties. Relevant notices (if any) are provided in your product installation at the following location: installDir/product_name/3rd_party_licensor_notice.pdf. Wind River may refer to third-party documentation by listing publications or providing links to third-party Web sites for informational purposes. Wind River accepts no responsibility for the information provided in such third-party documentation. Corporate Headquarters Wind River Systems, Inc. 500 Wind River Way Alameda, CA 94501-1153 U.S.A. toll free (U.S.): (800) 545-WIND telephone: (510) 748-4100 facsimile: (510) 749-2010 For additional contact information, please visit the Wind River URL: http://www.windriver.com For information on how to contact Customer Support, please visit the following URL: http://www.windriver.com/support VxWorks Architecture Supplement, 6.2 11 Oct 05 Part #: DOC-15660-ND-00 Contents 1 Introduction
    [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]
  • Multi-Platform Auto-Vectorization
    H-0236 (H0512-002) November 30, 2005 Computer Science IBM Research Report Multi-Platform Auto-Vectorization Dorit Naishlos, Richard Henderson* IBM Research Division Haifa Research Laboratory Mt. Carmel 31905 Haifa, Israel *Red Hat Research Division Almaden - Austin - Beijing - Haifa - India - T. J. Watson - Tokyo - Zurich LIMITED DISTRIBUTION NOTICE: This report has been submitted for publication outside of IBM and will probably be copyrighted if accepted for publication. I thas been issued as a Research Report for early dissemination of its contents. In view of the transfer of copyright to the outside publisher, its distribution outside of IBM prior to publication should be limited to peer communications and specific requests. After outside publication, requests should be filled only by reprints or legally obtained copies of the article (e.g ,. payment of royalties). Copies may be requested from IBM T. J. Watson Research Center , P. O. Box 218, Yorktown Heights, NY 10598 USA (email: [email protected]). Some reports are available on the internet at http://domino.watson.ibm.com/library/CyberDig.nsf/home . Multi-Platform Auto-Vectorization Dorit Naishlos Richard Henderson IBM Haifa Labs Red Hat [email protected] [email protected] Abstract. The recent proliferation of the Single Instruction Multiple Data (SIMD) model has lead to a wide variety of implementations. These have been incorporated into many platforms, from gaming machines and em- bedded DSPs to general purpose architectures. In this paper we present an automatic vectorizer as implemented in GCC - the most multi-targetable compiler available today. We discuss the considerations that are involved in developing a multi-platform vectorization technology, and demonstrate how our vectorization scheme is suited to a variety of SIMD architectures.
    [Show full text]
  • A Bibliography of Publications in IEEE Micro
    A Bibliography of Publications in IEEE Micro Nelson H. F. Beebe University of Utah Department of Mathematics, 110 LCB 155 S 1400 E RM 233 Salt Lake City, UT 84112-0090 USA Tel: +1 801 581 5254 FAX: +1 801 581 4148 E-mail: [email protected], [email protected], [email protected] (Internet) WWW URL: http://www.math.utah.edu/~beebe/ 16 September 2021 Version 2.108 Title word cross-reference -Core [MAT+18]. -Cubes [YW94]. -D [ASX19, BWMS19, DDG+19, Joh19c, PZB+19, ZSS+19]. -nm [ABG+16, KBN16, TKI+14]. #1 [Kah93i]. 0.18-Micron [HBd+99]. 0.9-micron + [Ano02d]. 000-fps [KII09]. 000-Processor $1 [Ano17-58, Ano17-59]. 12 [MAT 18]. 16 + + [ABG+16]. 2 [DTH+95]. 21=2 [Ste00a]. 28 [BSP 17]. 024-Core [JJK 11]. [KBN16]. 3 [ASX19, Alt14e, Ano96o, + AOYS95, BWMS19, CMAS11, DDG+19, 1 [Ano98s, BH15, Bre10, PFC 02a, Ste02a, + + Ste14a]. 1-GHz [Ano98s]. 1-terabits DFG 13, Joh19c, LXB07, LX10, MKT 13, + MAS+07, PMM15, PZB+19, SYW+14, [MIM 97]. 10 [Loc03]. 10-Gigabit SCSR93, VPV12, WLF+08, ZSS+19]. 60 [Gad07, HcF04]. 100 [TKI+14]. < [BMM15]. > [BMM15]. 2 [Kir84a, Pat84, PSW91, YSMH91, ZACM14]. [WHCK18]. 3 [KBW95]. II [BAH+05]. ∆ 100-Mops [PSW91]. 1000 [ES84]. 11- + [Lyl04]. 11/780 [Abr83]. 115 [JBF94]. [MKG 20]. k [Eng00j]. µ + [AT93, Dia95c, TS95]. N [YW94]. x 11FO4 [ASD 05]. 12 [And82a]. [DTB01, Dur96, SS05]. 12-DSP [Dur96]. 1284 [Dia94b]. 1284-1994 [Dia94b]. 13 * [CCD+82]. [KW02]. 1394 [SB00]. 1394-1955 [Dia96d]. 1 2 14 [WD03]. 15 [FD04]. 15-Billion-Dollar [KR19a].
    [Show full text]
  • Compiler-Based Data Prefetching and Streaming Non-Temporal Store Generation for the Intel R Xeon Phitm Coprocessor
    Compiler-based Data Prefetching and Streaming Non-temporal Store Generation for the Intel R Xeon PhiTM Coprocessor Rakesh Krishnaiyer†, Emre K¨ult¨ursay†‡, Pankaj Chawla†, Serguei Preis†, Anatoly Zvezdin†, and Hideki Saito† † Intel Corporation ‡ The Pennsylvania State University TM Abstract—The Intel R Xeon Phi coprocessor has software used, and generating them. In this paper, we: TM prefetching instructions to hide memory latencies and spe- • Present how the Intel R Xeon Phi coprocessor soft- cial store instructions to save bandwidth on streaming non- ware prefetch and non-temporal streaming store instructions temporal store operations. In this work, we provide details on R compiler-based generation of these instructions and evaluate are generated by the Intel Composer XE 2013, TM their impact on the performance of the Intel R Xeon Phi • Evaluate the impact of these mechanisms on the overall coprocessor using a wide range of parallel applications with performance of the coprocessor using a variety of parallel different characteristics. Our results show that the Intel R applications with different characteristics. Composer XE 2013 compiler can make effective use of these Our experimental results demonstrate that (i) a large mechanisms to achieve significant performance improvements. number of applications benefit significantly from software prefetching instructions (on top of hardware prefetching) that I. INTRODUCTION are generated automatically by the compiler for the Intel R TM TM The Intel R Xeon Phi coprocessor based on Intel R Xeon Phi coprocessor, (ii) some benchmarks can further Many Integrated Core Architecture (Intel R MIC Architec- improve when compiler options that control prefetching ture) is a many-core processor with long vector (SIMD) units behavior are used (e.g., to enable indirect prefetching), targeted for highly parallel workloads in the High Perfor- and (iii) many applications benefit from compiler generated mance Computing (HPC) segment.
    [Show full text]
  • Vybrid Controllers Technical Overview
    TM June 2013 Freescale, the Freescale logo, AltiVec, C-5, CodeTEST, CodeWarrior, ColdFire, ColdFire+, C- Ware, the Energy Efficient Solutions logo, Kinetis, mobileGT, PEG, PowerQUICC, Processor Expert, QorIQ, Qorivva, SafeAssure, the SafeAssure logo, StarCore, Symphony and VortiQa are trademarks of Freescale Semiconductor, Inc., Reg. U.S. Pat. & Tm. Off. Airfast, BeeKit, BeeStack, CoreNet, Flexis, Layerscape, MagniV, MXC, Platform in a Package, QorIQ Qonverge, QUICC Engine, Ready Play, SMARTMOS, Tower, TurboLink, Vybrid and Xtrinsic are trademarks of Freescale Semiconductor, Inc. All other product or service names are the property of their respective owners. © 2013 Freescale Semiconductor, Inc. • Overview of Vybrid Family • Vybrid Tower Board • Vybrid System Modules • QuadSPI Flash • Vybrid Clock System • Vybrid Power System • Vybrid Boot Operation • High Assurance Boot • Vybrid Trusted Execution • LinuxLink and MQX Embedded Software • DS-5 compiler TM Freescale, the Freescale logo, AltiVec, C-5, CodeTEST, CodeWarrior, ColdFire, ColdFire+, C-Ware, the Energy Efficient Solutions logo, Kinetis, mobileGT, PEG, PowerQUICC, Processor Expert, QorIQ, Qorivva, SafeAssure, the SafeAssure logo, StarCore, Symphony and VortiQa are trademarks of Freescale Semiconductor, Inc., Reg. U.S. Pat. & Tm. Off. 2 Airfast, BeeKit, BeeStack, CoreNet, Flexis, Layerscape, MagniV, MXC, Platform in a Package, QorIQ Qonverge, QUICC Engine, Ready Play, SMARTMOS, Tower, TurboLink, Vybrid and Xtrinsic are trademarks of Freescale Semiconductor, Inc. All other product or service names are the property of their respective owners. © 2013 Freescale Semiconductor, Inc. TM Freescale, the Freescale logo, AltiVec, C-5, CodeTEST, CodeWarrior, ColdFire, ColdFire+, C- Ware, the Energy Efficient Solutions logo, Kinetis, mobileGT, PEG, PowerQUICC, Processor Expert, QorIQ, Qorivva, SafeAssure, the SafeAssure logo, StarCore, Symphony and VortiQa are trademarks of Freescale Semiconductor, Inc., Reg.
    [Show full text]
  • POWER Block Course Assignment 3: SIMD with Altivec
    POWER Block Course Assignment 3: SIMD with AltiVec Hasso Plattner Institute Agenda 1. SIMD & AltiVec 2. Programming AltiVec in C 3. Example programs 4. Assignment 3 5. Performance measurement 6. Useful resources Assignment 3 What is this assignment about? In this assignment you will: □ Get to know AltiVec □ Use AltiVec instructions to speed up computation □ Do performance comparison on Linux using perf Submission Deadline: 04.11.2016 by any issues: [email protected] SIMD & AltiVec Sven Köhler, 29.09.2016 Chart 3 SIMD & AltiVec Sven Köhler, SIMD & AltiVec 29.09.2016 1 Chart 4 1. SIMD & AltiVec AltiVec is SIMD on POWER SIMD ::= Single Instruction Multiple Data The same instruction is performed simultaneously on multiple data points (data-level parallelism). Many architectures provide SIMD instruction set extensions. Intel: MMX, SSE, AVX ARM: NEON SIMD & AltiVec Sven Köhler, POWER: AltiVec (VMX), VSX 29.09.2016 Chart 5 Multiprocessor: Flynn‘s Taxonomy 1. (1966)SIMD & AltiVec Flynn’s Taxonomy on Multiprocessors (1966) 20 ■ Classify multiprocessor architectures among instruction and data processing dimension Single Instruction, Single Data (SISD) Single Instruction, Multiple Data (SIMD) (C) Blaise Barney Barney (C) Blaise SIMD & AltiVec Sven Köhler, 29.09.2016 Single Data (MISD) Multiple Instruction, Multiple Instruction, Multiple Data (MIMD) ) Chart 6 ( Single Instruction, Single Data (SISD) 1. SIMD & AltiVec Scalar vs. SIMD How many instructions are needed to add four numbers from memory? scalar 4 element SIMD + A0 B0 = C0 A0 B0 C0 A + B = C 1 1 1 A1 B1 C1 + = A B C A2 + B2 = C2 2 2 2 A B C + 3 3 3 A3 B3 = C3 SIMD & AltiVec 4 additions 1 addition Sven Köhler, 29.09.2016 8 loads 2 loads Chart 7 4 stores 1 store 1.
    [Show full text]
  • Cell Processor and Playstation 3
    Cell Processor and Playstation 3 Guillem Borrell i Nogueras February 24, 2009 • Cell systems • Bad news • More bad news • Good news • Q&A • No SPU Double precision improvements expected from IBM IBM Blades • QS21 • Cell BE based. • 8 SPE • 460 Gflops Float • 20 GFLops Double • QS22 • PowerXCell 8i based • 8 SPE* • 460 GFlops Float • 200 GFlops Double • Some of them already installed in BSC IBM Blades • QS21 • Cell BE based. • 8 SPE • 460 Gflops Float • 20 GFLops Double • QS22 • PowerXCell 8i based • 8 SPE* • 460 GFlops Float • 200 GFlops Double • Some of them already installed in BSC • No SPU Double precision improvements expected from IBM • 256 MB RAM Playstation 3 • Cell BE based. • 6 SPE • 460 Gflops Float • 20 GFLops Double Playstation 3 • Cell BE based. • 6 SPE • 460 Gflops Float • 20 GFLops Double • 256 MB RAM • 1 TFlop on a chip IBM Power 7 • 8 PowerPC cores • 4 threads per core (32 Threads!) • ? SPE IBM Power 7 • 8 PowerPC cores • 4 threads per core (32 Threads!) • ? SPE • 1 TFlop on a chip Cell Broadband Engine PPE PowerPC Processor Element PPU PowerPC Processor Unit EIB Element Interconnect Bus SPE Synergistic Processor Element SPU Synergistic Processor Unit MFC Memory Flow Controller DMA Direct Memory Access SIMD Single Instruction Multiple Data • I have said nothing about the data ¿How does it compute? • PPU starts a program • PPU loads an SPU context on a thread • SPU acquires the thread • SPU executes context • SPU ends the task and returns control to PPU ¿How does it compute? • PPU starts a program • PPU loads an SPU context on a thread • SPU acquires the thread • SPU executes context • SPU ends the task and returns control to PPU • I have said nothing about the data Overview • LS is regiser based • No type distinction • Data should be aligned by hand • MFC is a DMA controller • Data moved with DMA primitives.
    [Show full text]
  • Product Brief Rev
    Freescale Semiconductor Document Number: P5020PB Product Brief Rev. 1, 02/2013 P5020 QorIQ Communications Processor Product Brief This product brief provides an overview of the P5020 Contents QorIQ communications processor features as well as 1 P5020 Application Use Cases. 2 application use cases. 2 P5020 Dual-Core Processing Options . 4 3 Features . 5 The P5020 combines two Power Architecture® 4 Developer Environment. 29 5 Document Revision History. 31 processor cores with high-performance datapath acceleration logic and network and peripheral bus interfaces required for control processing in applications such as routers, switches, internet access devices, firewall and other packet filtering processors, network attached storage, storage area networks, imaging and general-purpose embedded computing. Its high level of integration offers significant performance benefits and greatly helps to simplify board design. © 2011-2013 Freescale Semiconductor, Inc. All rights reserved. P5020 Application Use Cases 1 P5020 Application Use Cases 1.1 Router Control Processor The following figure shows the P5020 in a linecard control plane application, where the linecard is part of a high-end network router. P5020 e5500 e5500 Figure 1. Control Plane Processor for a Router P5020 QorIQ Communications Processor Product Brief, Rev. 1 2 Freescale Semiconductor P5020 Application Use Cases 1.2 DSP Farm Control Processor The following figure shows a DSP farm enabled by the P5020 utilizing serial RapidIO. P5020 e5500 e5500 Figure 2. Control Plane Processor for a DSP Farm 1.3 SAN RAID 6 Controller The following figure shows a RAID-enabled Disk Array Controller in an redundant active-active system for block-oriented storage systems. The P5020 Data Path Acceleration Architecture (DPAA) accelerates RAID 5/6 calculations and low-overhead data movement while optionally supporting data-at rest encryption and Data Integrity Field support.
    [Show full text]
  • Enhancing the Matrix Transpose Operation Using Intel Avx Instruction Set Extension
    International Journal of Computer Science & Information Technology (IJCSIT) Vol 6, No 3, June 2014 ENHANCING THE MATRIX TRANSPOSE OPERATION USING INTEL AVX INSTRUCTION SET EXTENSION Ahmed Sherif Zekri1,2 1Department of Mathematics & Computer Science, Alexandria University, Egypt 2Department of Mathematics & Computer Science, Beirut Arab University, Lebanon ABSTRACT General-purpose microprocessors are augmented with short-vector instruction extensions in order to simultaneously process more than one data element using the same operation. This type of parallelism is known as data-parallel processing. Many scientific, engineering, and signal processing applications can be formulated as matrix operations. Therefore, accelerating these kernel operations on microprocessors, which are the building blocks or large high-performance computing systems, will definitely boost the performance of the aforementioned applications. In this paper, we consider the acceleration of the matrix transpose operation using the 256-bit Intel advanced vector extension (AVX) instructions. We present a novel vector-based matrix transpose algorithm and its optimized implementation using AVX instructions. The experimental results on Intel Core i7 processor demonstrates a 2.83 speedup over the standard sequential implementation, and a maximum of 1.53 speedup over the GCC library implementation. When the transpose is combined with matrix addition to compute the matrix update, B + AT, where A and B are squared matrices, the speedup of our implementation over the sequential algorithm increased to 3.19. KEYWORDS Matrix Transpose, vector instructions, streaming and advanced vector extensions, data-parallel computations 1. INTRODUCTION Matrix transpose is a main operation in many matrix- and vector-based computations of image, video, and scientific and image/signal processing applications.
    [Show full text]
  • The X86 Is Dead. Long Live the X86!
    the x86 is dead. long live the x86! CC3.0 share-alike attribution copyright c 2013 nick black with diagrams by david kanter of http://realworldtech.com “Upon first looking into Intel’s x86” that upon which we gaze is mankind’s triumph, and we are its stewards. use it well. georgia tech ◦ summer 2013 ◦ cs4803uws ◦ nick black The x86 is dead. Long live the x86! Why study the x86? Used in a majority of servers, workstations, and laptops Receives the most focus in the kernel/toolchain Very complex processor, thus large optimization space Excellent documentation and literature Fascinating, revealing, lengthy history Do not think that x86 is all that’s gone on over the past 30 years1. That said, those who’ve chased peak on x86 can chase it anywhere. 1Commonly expressed as “All the world’s an x86.” georgia tech ◦ summer 2013 ◦ cs4803uws ◦ nick black The x86 is dead. Long live the x86! In the grim future of computing there are 10,000 ISAs Alpha + BWX/FIX/CIX/MVI SPARC V9 + VIS3a AVR32 + JVM JVMb CMS PTX/SASSc PA-RISC + MAX-2 TILE-Gxd SuperH ARM + NEONe i960 Blackfin IA64 (Itanium) PowerISA + AltiVec/VSXf MIPS + MDMX/MIPS-3D MMIX IBMHLA (s390 + z) a Most recently the “Oracle SPARC Architecture 2011”. b m68k Most recently the Java SE 7 spec, 2013-02-28. c Most recently the PTX ISA 3.1 spec, 2012-09-13. VAX + VAXVA d TILE-Gx ISA 1.2, 2013-02-26. e z80 / MOS6502 ARMv8: A64, A32, and T32, 2011-10-27. f MIX PowerISA v.2.06B, 2010-11-03.
    [Show full text]
  • Release History
    Release History TRACE32 Online Help TRACE32 Directory TRACE32 Index TRACE32 Technical Support ........................................................................................................... Release History ............................................................................................................................. 1 General Information ................................................................................................................... 4 Code 4 Release Information ................................................................................................................... 4 Software Release from 01-Feb-2021 5 Build 130863 5 Software Release from 01-Sep-2020 8 Build 125398 8 Software Release from 01-Feb-2020 11 Build 117056 11 Software Release from 01-Sep-2019 13 Build 112182 13 Software Release from 01-Feb-2019 16 Build 105499 16 Software Release from 01-Sep-2018 19 Build 100486 19 Software Release from 01-Feb-2018 24 Build 93173 24 Software Release from 01-Sep-2017 27 Build 88288 27 Software Release from 01-Feb-2017 32 Build 81148 32 Build 80996 33 Software Release from 01-Sep-2016 36 Build 76594 36 Software Release from 01-Feb-2016 39 Build 69655 39 Software Release from 01-Sep-2015 42 Build 65657 42 Software Release from 02-Feb-2015 45 Build 60219 45 Software Release from 01-Sep-2014 48 Build 56057 48 Software Release from 16-Feb-2014 51 ©1989-2021 Lauterbach GmbH Release History 1 Build 51144 51 Software Release from 16-Aug-2013 54 Build 50104 54 Software Release from 16-Feb-2013 56
    [Show full text]