Embedded Systems Design Europe - August/September 2008 - (Page 10)

news ARM compiler boosts Freescale i.MX31 New certiﬁed benchmarks from Embedded Microprocessor Benchmark Consortium (EEMBC) indicate a twofold improvement in performance on the ARM processor-based Freescale i.MX31 multimedia applications processor, when tested with the ARM RealView RVCT 3.1 build 559 compiler. The 532-MHz Freescale i.MX31 was tested in an out-of-the-box environment against four EEMBC benchmark suites - AutoBench 1.1, ConsumerBench 1.1, OABench 1.1, and TeleBench 1.1 – yielding extensive data on the processor’s expected performance in automotive/industrial, digital imaging, ofﬁce automation, and ﬁxed telecom applications. The new results supersede a previous set of certiﬁed scores using a GCC version 3.43 compiler and limited - O2 optimizations. Original scores were certiﬁed running under the Linux OS; the new scores were certiﬁed running without an OS. “The ARM compiler helped the i.MX31 achieve up to a 2x increase in performance with one benchmark, and a 43 percent increase in overall performance with all four benchmarks over an earlier version of the GCC compiler,” said EEMBC President Markus Levy. “These new benchmark scores demonstrate the beneﬁts of ensuring that developers take advantage of the latest in compiler optimizations and technologies.” Freescale’s i.MX31 processor, with a vector ﬂoating point coprocessor and L2 cache, is designed for wireless devices, such as portable media players and portable navigation devices, running computationally intensive multimedia applications. Based on an ARM1136JF-STM processor, its target devices include feature rich smartphones, digital video recorders, digital cameras, mobile gaming consoles, mobile multimedia players, and many other mobile wireless applications. The ARM compiler, part of the RealView Development Suite 3.1, is designed to enable developers to get the best performance out of their ARM processor-based devices. ARM develops its compiler technology in conjunction with its processor technology to offer superior performance at the lowest possible memory cost. ■ On page 20 Markus Levy explains why the embedded systems industry needs is a standard way to compare hypervisors . LabVIEW updated for multicore and wireless LabVIEW version 8.6 from National Instruments Corp. adds techniques to improve design with multicore processors, ﬁeld-programmable gate arrays (FPGAs) and wireless communication. LabVIEW now provides engineers with a single platform to increase test and control system throughput with multicore processors, reduce the development time of high-performance FPGA-based advanced control and embedded prototyping applications and more easily create distributed measurement systems to acquire data remotely. To increase performance, LabVIEW 8.6 includes more than 1,200 newlyoptimized advanced analysis functions that offer faster, enhanced math and signal processing on multicore systems for control and test applications. Vision applications also can beneﬁt from multicore systems by using innovative image processing functions included in the NI Vision Development Module for LabVIEW 8.6 that automatically dis10 tributes data sets across multiple cores. Using multicore features, test engineers can develop applications to test wireless devices up to four times faster with the latest version of the NI Modulation Toolkit for LabVIEW and control system engineers can execute simulation models in parallel up to ﬁve times faster with the LabVIEW 8.6 Control Design and Simulation Module. Identiﬁcation of parallel sections of code are eased through a feature that reorganises LabVIEW diagrams. The LabVIEW FPGA Module can be used with FPGA-based commercial off-the-shelf hardware such as NI Compact RIO to customize measurement and control systems for increased performance in applications such as semiconductor validation and advanced machine control. LabVIEW 8.6 enables the programing of CompactRIO programmable automation controllers (PACs) directly without having to separately program the FPGA. Additional simulation features reduce the number of time-consuming compilations by validating an FPGA application on the desktop. Version 8.6 adds IP development and integration features including the a fast Fourier transform (FFT) IP core to ofﬂoad spectral analysis functions which increases the performance in applications such as machine condition monitoring and RF test and the new Component-Level IP (CLIP) Node to ease import of existing or third-party IP into LabVIEW FPGA. With support for the latest wireless data acquisition devices and drivers for more than 20 third-party wireless sensors, LabVIEW 8.6 simpliﬁes programming of distributed measurement systems with a single software platform. Engineers can now conﬁgure data acquisition applications to use NI Wi-Fi data acquisition hardware without making code changes in LabVIEW 8.6. AUGUST – SEPTEMBER 2008 | embedded systems design europe | www.embedded.com/europe 010_ESDE.indd 10 28/08/08 17:37:34 http://www.embedded.com/europe

Table of Contents Feed for the Digital Edition of Embedded Systems Design Europe - August/September 2008

Embedded Systems Design Europe - August/September 2008 Contents TI Overhauls DSP Lineup, Adds 15 Processors QNX Publishes Source Code for File System Congatec to Take on Proprietary Market Swiss Multicore Project Wins Microsoft Grant OpenCores Bundles Development Tool ARM Compiler Boosts Freescale i.MX31 LabVIEW Updated for Multicore and Wireless Cover Feature: Interactive Tool Supports Multiprocessor SoC Design Wanted: Benchmaking for Embedded VMM Hypervisors Graphical Design Empowers Spider Robots Building a Power Supply for Discontinuous Transmission Wireless Networks RTOS Selection & Best Practices Achieving Cache Coherence in a MIPS32 Multicore Design New Products Advertising Contacts

Embedded Systems Design Europe - August/September 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.