background image

Intel® Xeon Phi™ Coprocessor D

EVELOPER

Q

UICK 

S

TART 

G

UIDE

 

 

31 

 

Notices 

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.  

 

A "Mission Critical Application" is any application in which failure of the Intel Product could result, directly or 

indirectly, in personal injury or death. SHOULD YOU PURCHASE OR USE INTEL'S PRODUCTS FOR ANY 

SUCH MISSION CRITICAL APPLICATION, YOU SHALL INDEMNIFY AND HOLD INTEL AND ITS 

SUBSIDIARIES, SUBCONTRACTORS AND AFFILIATES, AND THE DIRECTORS, OFFICERS, AND 

EMPLOYEES OF EACH, HARMLESS AGAINST ALL CLAIMS COSTS, DAMAGES, AND EXPENSES AND 

REASONABLE ATTORNEYS' FEES ARISING OUT OF, DIRECTLY OR INDIRECTLY, ANY CLAIM OF 

PRODUCT LIABILITY, PERSONAL INJURY, OR DEATH ARISING IN ANY WAY OUT OF SUCH MISSION 

CRITICAL APPLICATION, WHETHER OR NOT INTEL OR ITS SUBCONTRACTOR WAS NEGLIGENT IN 

THE DESIGN, MANUFACTURE, OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS.  

 

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. The information here is subject to change without notice. 

Do not finalize a design with this information.  

 

The products described in this document 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.  

 

Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing 

your product order.

 

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 go 
to:  

http://www.intel.com/design/literature.htm

 

Intel, the Intel logo, Cilk, Xeon and Intel Xeon Phi are trademarks of Intel Corporation in the U.S. 
and other countries. 

*Other names and brands may be claimed as the property of others 

Copyright© 2013 Intel Corporation. All rights reserved. 

 

 

Содержание Xeon Phi

Страница 1: ...White Paper Intel Xeon Phi Coprocessor DEVELOPER S QUICK START GUIDE Version 1 7...

Страница 2: ...Intel Xeon Phi Coprocessor If It Hangs 11 Monitoring the Intel Xeon Phi Coprocessor 12 Running an Intel Xeon Phi Coprocessor program from the host system 12 Working directly with the uOS Environment I...

Страница 3: ...Programming on the Intel Xeon Phi Coprocessor OpenMP Intel Cilk Plus Extended Array Notation 23 Parallel Programming on the Intel Xeon Phi Coprocessor Intel Cilk Plus 24 Parallel Programming on Intel...

Страница 4: ...known methods BKMs developed by users at Intel This document does not 1 Cover each tool in detail Please refer to the user guides for the individual tools 2 Provide in depth training Terminology Host...

Страница 5: ...e with the Intel Xeon Phi Coprocessor SCI Symmetric Communications Interface the mechanism for inter node communication within a single platform where an node is a Intel Xeon Phi Coprocessor or an Int...

Страница 6: ...ty such as loading and unloading executables onto the Intel Xeon Phi Coprocessor invoking functions from the executables on the card and providing a two way notification mechanism between host and car...

Страница 7: ...U is a key feature of the Intel MIC Architecture based cores Fully utilizing the vector unit is critical for best Intel Xeon Phi Coprocessor performance It is important to note that Intel MIC Architec...

Страница 8: ...rnel 2 6 32 431 SUSE Linux Enterprise Server SLES 11 SP2 kernel 3 0 13 0 27 default or SUSE Linux Enterprise Server SLES 11 SP3 kernel 3 0 76 0 11 default Section 2 1 in readme txt Be sure to install...

Страница 9: ...If you acquired a serial number for Intel tools go to the Intel Registration Center IRC at http registrationcenter intel com to register and download the products Click the button Register Product wil...

Страница 10: ...MPSS gets started it loads the data collection driver automatically But for some reason if it fails to load the data collection driver you can manually load the driver by going to opt intel vtune_amp...

Страница 11: ...Access to the Intel Xeon Phi Coprocessor after Reboot The Intel Xeon Phi Coprocessor will not start when the host system reboots You will need to manually start the Intel Xeon Phi Coprocessor and then...

Страница 12: ...o Monitoring the Intel Xeon Phi Coprocessor If you want to monitor the load on your coprocessor its temperature etc run the System Management and Configuration SMC utility See section 8 3 of the MPSS...

Страница 13: ...is directory to their default path micinfo provides information about host and coprocessor system configuration micflash updates the flash on the coprocessor saves and retrieves the version and other...

Страница 14: ...l MIC Architecture o Intel Threading Building Blocks Intel TBB o Intel Integrated Performance Primitive Intel IPP Libraries packaged separately include o Intel MPI for Linux OS including Intel Many In...

Страница 15: ...cture Information on Intel MIC Architecture intrinsics can be found in the Compiler Reference Intrinsics section under Intrinsics for Intel MIC Architecture o Release_notes_ _2013SP1_l_en pdf please r...

Страница 16: ...ad opt intel composerxe mkl examples mic_ao blasc and mic_ao blasf o The rest of the samples demonstrate use of MKL via compiler assisted offload opt intel composerxe mkl examples mic_offload Some sam...

Страница 17: ...You can visit the Forum on the Intel Xeon Phi Coprocessor to post questions It can be found at the http software intel com en us forums intel many integrated core Using the Offload Compiler Explicit M...

Страница 18: ...e programmer uses pragma offload target mic as shown in the example below to mark statements offload constructs that should execute on the Intel Xeon Phi Coprocessor The offloaded region is defined as...

Страница 19: ...ode Example 3 Vector Reduction with Offload in C C Asynchronous Offload and Data Transfer Asynchronous offload and data transfer between the host and the Intel Xeon Phi Coprocessor is available For de...

Страница 20: ...hared memory there is no hardware that maps some portion of the memory on the Intel Xeon Phi Coprocessor to the host system The memory subsystems on the coprocessor and host are completely independent...

Страница 21: ...d Reference Guides The section Restrictions on Offload Using Shared Virtual Memory in the document Intel C Compiler User and Reference Guide shows some restrictions of using this programming model Nat...

Страница 22: ...no unusual complications beyond the larger number of threads Parallel Programming on the Intel Xeon Phi Coprocessor OpenMP There is no correspondence between OpenMP threads on the host CPU and on the...

Страница 23: ...et mic in size in data length size omp parallel do reduction ret do i 1 size ret ret data i enddo omp end parallel do FTNReductionOMP ret return end function FTNReductionOMP Code Example 6 Fortran Usi...

Страница 24: ...available to an application built for the Intel MIC Architecture using Intel Cilk Plus wrap the header files with pragma offload_attribute push target mic and pragma offload_attribute pop as follows p...

Страница 25: ...d global data required on the Intel Xeon Phi Coprocessor should be appended by the special function attribute __attribute__ target mic As an example parallel_reduce recursively splits an array into su...

Страница 26: ...CReductionTBB data size return ret Code Example 13 Offloading Intel TBB Code to the coprocessor in C C NOTE Codes using Intel TBB with an offload should be compiled with tbb flag instead of ltbb Using...

Страница 27: ...offload target mic PHI_DEV in A length matrix_elements free_if 0 in B length matrix_elements free_if 0 in C length matrix_elements free_if 0 Code Example 14 Sending the Data to the Intel Xeon Phi Cop...

Страница 28: ...a A N B N beta C N Code Example 17 Controlling Threads on the Intel Xeon Phi Coprocessor Using omp_set_num_threads Intel MKL Automatic Offload Model A few of the host Intel MKL functions are Automatic...

Страница 29: ...ugging Intel MIC Architecture applications under the Debugging with the Intel Debugger on Eclipse and Debugging on the Command Line sections of opt intel composerxe Documentation en_US debugger debugg...

Страница 30: ...efore joining Intel Charles was a consulting software engineer for Oracle Corporation where he concentrated on parallelism and 64 bit support in Windows NT and OpenVMS versions of the Oracle RDBMS on...

Страница 31: ...SUCH MISSION CRITICAL APPLICATION WHETHER OR NOT INTEL OR ITS SUBCONTRACTOR WAS NEGLIGENT IN THE DESIGN MANUFACTURE OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS Intel may make changes to specif...

Страница 32: ...ions include SSE2 SSE3 and SSE3 instruction sets and other optimizations Intel does not guarantee the availability functionality or effectiveness of any optimization on microprocessors not manufacture...

Отзывы: