24小时热门版块排行榜    

查看: 1744  |  回复: 7

love5264

木虫 (正式写手)

[求助] siesta-3.1编译成功,但有几个问题求助

ifort12.1+openmpi-1.4.5+siesta-3.1
centos 6.2
问题一
arch.make文件里
FFLAGS=?(怎么设置?)
看到网上帖子有:
FFLAGS=-O2 -mp1 -pad -ip
FFLAGS=-O2 -i-static
FFLAGS=-g -O2

请教各位前辈,这后面值应该怎么写?或者在哪能找到相关信息?
问题二

BLAS_LIBS=
LAPACK_LIBS=
BLACS_LIBS=
SCALAPACK_LIBS=

请教各位前辈,都需要填哪些库?我看大家写的都不完全一样?哪些库是必须的,顺序应该怎样?
还有-lguide -pthread -openmp 这几个应该加在哪?

谢谢!

[ 来自科研家族 Solar Cells 之家 ]
回复此楼
谋定而动
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖
回帖置顶 ( 共有1个 )

love5264

木虫 (正式写手)

引用回帖:
7楼: Originally posted by love5264 at 2012-05-02 15:51:39:
安装参照:
http://muchong.com/bbs/viewthread.php?tid=4433857&fpage=1

问题一:
FFLAGS= -O3 -m64 -prec-div -pc80 -pad -ip -i-static
问题二:
ifort安装目录:/opt/intel/composer_xe_2011_sp1.9.293/
BLAS_LIBS=-L/opt/intel/composer_xe_2011_sp1.9.293/mkl/lib/intel64 -lmkl_solver_lp64 -lmkl_intel_lp64
LAPACK_LIBS=-L/opt/intel/composer_xe_2011_sp1.9.293/mkl/lib/intel64 -lmkl_sequential -lmkl_core
BLACS_LIBS=-L/opt/intel/composer_xe_2011_sp1.9.293/mkl/lib/intel64 -lmkl_blacs_openmpi_lp64
SCALAPACK_LIBS=-L/opt/intel/composer_xe_2011_sp1.9.293/mkl/lib/intel64 -lmkl_scalapack_lp64
谋定而动
8楼2012-05-02 16:26:38
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖
普通回帖

love5264

木虫 (正式写手)

找到一个:
$ ifort -help opt
Optimization
------------

-O1       optimize for maximum speed, but disable some optimizations which
          increase code size for a small speed benefit
-O2       optimize for maximum speed (DEFAULT)
-O3       optimize for maximum speed and enable more aggressive optimizations
          that may not improve performance on some programs
-O        same as -O2
-Os       enable speed optimizations, but disable some optimizations which
          increase code size for small speed benefit
-O0       disable optimizations
-fast     enable -xHOST -O3 -ipo -no-prec-div -static
          options set by -fast cannot be overridden with the exception of
          -xHOST, list options separately to change behavior
-Ofast    enable -O3 -no-prec-div optimizations
-fno-alias
          assume no aliasing in program
-fno-fnalias
          assume no aliasing within functions, but assume aliasing across calls
-nolib-inline
          disable inline expansion of intrinsic functions
谋定而动
2楼2012-04-14 22:53:27
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

感谢参与,应助指数 +1
liliangfang: 金币-1, 违规存档, 非应助请勿选择应助回帖 2012-04-17 09:09:08
zzy870720z: 应助指数-1, 扣回应助指数,下次注意 2012-04-20 12:17:08
祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福祝福
3楼2012-04-17 08:56:04
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

lemonare

新虫 (小有名气)

【答案】应助回帖

★ ★ ★ ★ ★ ★ ★ ★ ★ ★ ★
感谢参与,应助指数 +1
fzx2008: 金币+1, 谢谢回帖交流! 2012-04-25 16:00:53
love5264: 回帖置顶 2012-04-26 20:36:19
love5264: 金币+10, ★★★很有帮助, 感谢你的回复! 2012-04-26 20:36:59
love5264: 取消置顶 2012-05-02 16:27:03
这是我在网上找到的:
FFLAGS=-O3 -ipo -xHost -ip -prec-div -prec-sqrt -opt-prefetch -mkl=parallel
Do not use -g ( you cant debug, but if you havnt had use for it, then why
use it? )
Do not use -mp, it is a deprecated option and there are other flags that
should perform better. For instance  -prec-div -prec-sqrt.
-ipo will make the compile time a much longer process when linking in the
final step, i have seen little performance gain from this.
-xHost should be used if the machine you compile on is the same as the
excution nodes. Otherwise you should use -xSSE3 (E5000 has this).
The last -mkl=parallel enables the mkl library to use parallel execution on
some of the routines (notice that this option will not affect blacs as it is
already parallelised.). I have had better runs with it! But be aware that
you have to set the OMP_NUM_THREADS=x, otherwise it will use all available cores on the machine.
每一天都是一个成长的过程!
4楼2012-04-25 15:58:57
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

guohuazhong

至尊木虫 (职业作家)


WDD880227: 金币+1, 感谢交流提示 2012-04-27 08:53:10
FFLAGS=-O3 我的编译器,优化级别设成-O3会提示被0除的情况

» 本帖已获得的红花(最新10朵)

5楼2012-04-25 17:50:13
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

love5264

木虫 (正式写手)

送鲜花一朵
引用回帖:
5楼: Originally posted by guohuazhong at 2012-04-25 17:50:13:
FFLAGS=-O3 我的编译器,优化级别设成-O3会提示被0除的情况

感谢你的回复!
谋定而动
6楼2012-04-26 20:38:34
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖

love5264

木虫 (正式写手)

★ ★
zzy870720z: 金币+2, 鼓励发表解决方案,呵呵 2012-05-02 18:02:47
谋定而动
7楼2012-05-02 15:51:39
已阅   回复此楼   关注TA 给TA发消息 送TA红花 TA的回帖
相关版块跳转 我要订阅楼主 love5264 的主题更新
信息提示
请填处理意见