Документ взят из кэша поисковой машины. Адрес оригинального документа : http://www.naic.edu/~phil/software/amd/CompilerOptQuickRef-62004200.pdf Дата изменения: Sat Feb 16 21:03:31 2013 Дата индексирования: Sat Mar 1 02:48:06 2014 Кодировка:

AMD OpteronTM 6200 Series processors AMD OpteronTM 4200 Series processors Compiler Options Quick Reference Guide
Open64
Latest release: 4.5.1, December 2011 http://developer.amd.com/open64
Architecture Generate instructions specific to Interlagos Generate instructions for the local machine Optimization Levels Disable all optimizations Local optimizations Global optimizations (default) Additional aggressive optimizations Maximize performance Additional Optimizations Autoparallelization Feedback directed optimization Huge pages Interprocedural Analysis and Optimizations Link to ACML 5.0 Loop nest optimizations, vectorization, prefetch, fission, fusion Multicore scalability OpenMP Prefetch (disabled by default) Floating point accuracy Floating point accuracy -fp-accuracy -apo -fb-create -fb-opt -HP -ipa -L/opt/acml5.0.0/ open64_64_fma4/ lib -lacml -LNO:fission=n -LNO:fusion=n -mso -mp -LNO:prefetch -LNO:prefetch_ahead -O0 -O1 -O2 -O3 -Ofast -march=bdver1 -march=auto

gcc
Latest release: 4.7, March 2012 http://gcc.gnu.org
Architecture Generate instructions specific to Interlagos Optimization Levels Disable all optimizations (default) Local optimizations Global optimizations Additional aggressive optimizations Maximize performance Additional Optimizations AVX tuning Enable unrolling Generate prefetch instructions for loops Inline string operations Link to ACML 5.0 OpenMP Profile guided optimization Turn off partial redundancy elimination Vectorization Floating point accuracy Enable generation of code that follows IEEE arithmetic -mieee-fp -mprefer-avx128 -funroll-all-loops -fprefetch-loop-arrays --param prefetch-latency=300 (300-700) -minline-all-stringops -L/opt/acml5.0.0/ gfortran64_fma4/ lib -lacml -fopenmp -fprofile-generate -fprofile-use -fno-tree-pre -ftree-vectorize -O0 -O1 -O2 -O3 -Ofast -march=bdver1

Enable faster, less precise math -ffast-math operations

For more information, visit http://developer.amd.com
Advanced Micro Devices One AMD Place P.O. Box 3453 Sunnyvale, CA 94088-3453 © 2011 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, AMD Opteron, and combinations thereof are trademarks of Advanced Micro Devices, Inc. Linux is a registered trademark of Linus Torvalds. Other names are for informational purposes only and may be trademarks of their respective owners.


AMD OpteronTM 6200 Series processors AMD OpteronTM 4200 Series processors Compiler Options Quick Reference Guide
ICC
Latest release: 12.1, January 2012 http://software.intel.com
Architecture Generate instructions that run on -msse3 (avoid ­mAVX) Interlagos Optimization Levels Disable all optimizations -O0

PGI
Latest release: 12.3, March 2012 http://www.pgroup.com
Architecture Generate instructions that run on Interlagos Optimization Levels Disable all optimizations Local optimization Global optimization Aggressive global optimization -O3 -fast Hoist guarded invariant floating point expressions Maximize performance -unroll-aggressive -no-prec-div -simd -ipo ­L/opt/acml5.0.0/ifort64_fma4/ lib ­lacml -openmp -opt-prefetch -prof-gen -prof-use ­use-intel-optimized-headers Additional Optimizations Huge pages Autoparallelize loops Enable vectorization Interprocedural Optimization Link to ACML to 5.0 OpenMP Prefetch instructions Profile guided optimization Unroll loops Floating point accuracy -fp-model Generate relaxed precision code Perform floating point operations in conformance with IEEE standard -Mfprelaxed -Kieee -Msmartalloc=huge -Mconcur -Mvect -Mipa=fast,inline -L/opt/acml5.0.0/ pgi64_fma4/lib -lacml -mp -Mvect=prefetch -Mpfi -Mpfo -Munroll -O0 -O1 -O2 -O3 -O4 -fast -tp bulldozer

Speed optimization without code -O1 growth Enable optimization including vectorization Aggressive optimization Maximize performance Additional Optimizations Aggressive unrolling Disable improved precision floating divides Enable vectorization Interprocedural Optimization Link to ACML to 5.0 OpenMP Prefetch optimization Profile generated optimization Use optimized header definitions Floating point accuracy Floating point accuracy -O2

For more information, visit http://developer.amd.com
Advanced Micro Devices One AMD Place P.O. Box 3453 Sunnyvale, CA 94088-3453 © 2011 Advanced Micro Devices, Inc. All rights reserved. AMD, the AMD Arrow logo, AMD Opteron, and combinations thereof are trademarks of Advanced Micro Devices, Inc. Linux is a registered trademark of Linus Torvalds. Other names are for informational purposes only and may be trademarks of their respective owners.