NWChem 6.1.1 RedHat 5.4, Intel 11.1, OpenMPI 1.3.4

最新推荐文章于 2023-04-09 22:51:12 发布
最新推荐文章于 2023-04-09 22:51:12 发布
阅读量1.4k 收藏
点赞数
分类专栏： NWChem 文章标签： redhat include build variables thread linux
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/tengh/article/details/7971254
版权

NWChem 6.1.1 RedHat 5.4, Intel 11.1, OpenMPI 1.3.4

Viewed 162 times, With a total of 4 Posts
Forum >> NWChem's corner >> Compiling NWChem

   
Just Got Here
Threads 1
Posts 3
6:18:59 AM PDT - Mon, Jul 16th 2012  
Hello,

Can't see a thread with the same errors as mine, so hope someone can help.

Using following environment:

  1. !/bin/bash
    1. set environment for NWChem build
export NWCHEM_TOP=/home/cczcb/NWChem/nwchem-6.1.1-src
export NWCHEM_MODULES=all
export NWCHEM_TARGET=LINUX64
export MPI_INCLUDE=/opt/openmpi-1.3.4-1/intel/include
export LIBMPI="-lmpi_f90 -lmpi_f77 -lmpi -lpthread"
export USE_MPIF=y
export MPI_LIB=/opt/openmpi-1.3.4-1/intel/lib64
export USE_MPI=y

Assuming openmpi will look after Infiniband, so don't set IB related variables.

Build with:

make nwchem_config

make FC=ifort> make.log 2>&1 &


make.log has text below at end. For some reason, some variables aren't defined properly. Can anyone suggest why this is, and how to fix ? Would like to use MKL libraries for final build, but thought I'd get it going with inbuilt libraries first.

Really grateful for any assistance.


ifort -c -i8 -align -w -g -vec-report1 -O3 -prefetch -unroll -tpp7 -ip -xP -no-prec-div -I. -I/opt/openmpi-1.3.4-1/intel/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include -DEXT_INT -DLINUX -DLINUX64 -DIFCV8 -DIFCLINUX -DPARALLEL_DIAG -DMPI DMatrix.F
ifort -c -i8 -align -w -g -vec-report1 -O3 -prefetch -unroll -tpp7 -ip -xP -no-prec-div -I. -I/opt/openmpi-1.3.4-1/intel/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include -DEXT_INT -DLINUX -DLINUX64 -DIFCV8 -DIFCLINUX -DPARALLEL_DIAG -DMPI DMatrix-mpi.F
DMatrix-mpi.F(1363): error #6404: This name does not have a type, and must have an explicit type. [STUPID_MSGLEN]
     stupid_msglen = 7*7
^
DMatrix-mpi.F(1365): error #6404: This name does not have a type, and must have an explicit type. [STUPID_DOUBLE]
    >                      stupid_double,stupid_sum,
^
DMatrix-mpi.F(1365): error #6404: This name does not have a type, and must have an explicit type. [STUPID_SUM]
    >                      stupid_double,stupid_sum,
^
DMatrix-mpi.F(1366): error #6404: This name does not have a type, and must have an explicit type. [STUPID_WORLD]
    >                      stupid_world,stupid_ierr)
^
DMatrix-mpi.F(1366): error #6404: This name does not have a type, and must have an explicit type. [STUPID_IERR]
    >                      stupid_world,stupid_ierr)
^
DMatrix-mpi.F(2277): error #6404: This name does not have a type, and must have an explicit type. [STUPID_MSGLEN]
              stupid_msglen = kk-1
^
DMatrix-mpi.F(2278): error #6404: This name does not have a type, and must have an explicit type. [STUPID_TASKID]
              stupid_taskid = icur
^
DMatrix-mpi.F(2279): error #6404: This name does not have a type, and must have an explicit type. [STUPID_DOUBLE]
              call MPI_Bcast(work1,stupid_msglen,stupid_double,
^
DMatrix-mpi.F(2280): error #6404: This name does not have a type, and must have an explicit type. [STUPID_COMM_I]
    >               stupid_taskid,stupid_comm_i,stupid_ierr)
^
DMatrix-mpi.F(2280): error #6404: This name does not have a type, and must have an explicit type. [STUPID_IERR]
    >               stupid_taskid,stupid_comm_i,stupid_ierr)
^
DMatrix-mpi.F(2333): error #6404: This name does not have a type, and must have an explicit type. [STUPID_SUM]
    >                      stupid_double,stupid_sum,
^
DMatrix-mpi.F(2334): error #6404: This name does not have a type, and must have an explicit type. [STUPID_WORLD]
    >                      stupid_world,stupid_ierr)
^
DMatrix-mpi.F(82): (col. 7) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(86): (col. 7) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(108): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(270): (col. 7) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(286): (col. 11) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(330): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(617): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(750): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(902): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(1111): (col. 19) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(1117): (col. 19) remark: LOOP WAS VECTORIZED.
compilation aborted for DMatrix-mpi.F (code 1)
make[4]: *** [/home/cczcb/NWChem/nwchem-6.1.1-src/lib/LINUX64/libnwpwlib.a(DMatrix-mpi.o)] Error 1
make[3]: *** [optimized] Error 2
make[2]: *** [subdirs] Error 1
make[1]: *** [subdirs] Error 1
make: *** [libraries] Error 1
 

  • Bert Forum:Admin, Forum:Mod, NWChemDeveloper, bureaucrat, sysop
Forum Vet
Threads 3
Posts 391
10:51:42 AM PDT - Mon, Jul 16th 2012  
Add 
  

   export USE_MPIF4=y

This should eliminate your issues. 


Bert


Quote:Colinbann  Jul 16th 1:18 pm
Hello,

Can't see a thread with the same errors as mine, so hope someone can help.

Using following environment:

  1. !/bin/bash
    1. set environment for NWChem build
export NWCHEM_TOP=/home/cczcb/NWChem/nwchem-6.1.1-src
export NWCHEM_MODULES=all
export NWCHEM_TARGET=LINUX64
export MPI_INCLUDE=/opt/openmpi-1.3.4-1/intel/include
export LIBMPI="-lmpi_f90 -lmpi_f77 -lmpi -lpthread"
export USE_MPIF=y
export MPI_LIB=/opt/openmpi-1.3.4-1/intel/lib64
export USE_MPI=y

Assuming openmpi will look after Infiniband, so don't set IB related variables.

Build with: 

make nwchem_config

make FC=ifort> make.log 2>&1 &


make.log has text below at end. For some reason, some variables aren't defined properly. Can anyone suggest why this is, and how to fix ? Would like to use MKL libraries for final build, but thought I'd get it going with inbuilt libraries first.

Really grateful for any assistance.


ifort -c -i8 -align -w -g -vec-report1 -O3 -prefetch -unroll -tpp7 -ip -xP -no-prec-div -I. -I/opt/openmpi-1.3.4-1/intel/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include -DEXT_INT -DLINUX -DLINUX64 -DIFCV8 -DIFCLINUX -DPARALLEL_DIAG -DMPI DMatrix.F
ifort -c -i8 -align -w -g -vec-report1 -O3 -prefetch -unroll -tpp7 -ip -xP -no-prec-div -I. -I/opt/openmpi-1.3.4-1/intel/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/include -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include -DEXT_INT -DLINUX -DLINUX64 -DIFCV8 -DIFCLINUX -DPARALLEL_DIAG -DMPI DMatrix-mpi.F
DMatrix-mpi.F(1363): error #6404: This name does not have a type, and must have an explicit type. [STUPID_MSGLEN] 
     stupid_msglen = 7*7
^
DMatrix-mpi.F(1365): error #6404: This name does not have a type, and must have an explicit type. [STUPID_DOUBLE] 
    >                      stupid_double,stupid_sum,
^
DMatrix-mpi.F(1365): error #6404: This name does not have a type, and must have an explicit type. [STUPID_SUM] 
    >                      stupid_double,stupid_sum,
^
DMatrix-mpi.F(1366): error #6404: This name does not have a type, and must have an explicit type. [STUPID_WORLD] 
    >                      stupid_world,stupid_ierr)
^
DMatrix-mpi.F(1366): error #6404: This name does not have a type, and must have an explicit type. [STUPID_IERR] 
    >                      stupid_world,stupid_ierr)
^
DMatrix-mpi.F(2277): error #6404: This name does not have a type, and must have an explicit type. [STUPID_MSGLEN] 
              stupid_msglen = kk-1
^
DMatrix-mpi.F(2278): error #6404: This name does not have a type, and must have an explicit type. [STUPID_TASKID] 
              stupid_taskid = icur
^
DMatrix-mpi.F(2279): error #6404: This name does not have a type, and must have an explicit type. [STUPID_DOUBLE] 
              call MPI_Bcast(work1,stupid_msglen,stupid_double,
^
DMatrix-mpi.F(2280): error #6404: This name does not have a type, and must have an explicit type. [STUPID_COMM_I] 
    >               stupid_taskid,stupid_comm_i,stupid_ierr)
^
DMatrix-mpi.F(2280): error #6404: This name does not have a type, and must have an explicit type. [STUPID_IERR] 
    >               stupid_taskid,stupid_comm_i,stupid_ierr)
^
DMatrix-mpi.F(2333): error #6404: This name does not have a type, and must have an explicit type. [STUPID_SUM] 
    >                      stupid_double,stupid_sum,
^
DMatrix-mpi.F(2334): error #6404: This name does not have a type, and must have an explicit type. [STUPID_WORLD] 
    >                      stupid_world,stupid_ierr)
^
DMatrix-mpi.F(82): (col. 7) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(86): (col. 7) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(108): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(270): (col. 7) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(286): (col. 11) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(330): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(617): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(750): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(902): (col. 10) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(1111): (col. 19) remark: LOOP WAS VECTORIZED.
DMatrix-mpi.F(1117): (col. 19) remark: LOOP WAS VECTORIZED.
compilation aborted for DMatrix-mpi.F (code 1)
make[4]: *** [/home/cczcb/NWChem/nwchem-6.1.1-src/lib/LINUX64/libnwpwlib.a(DMatrix-mpi.o)] Error 1
make[3]: *** [optimized] Error 2
make[2]: *** [subdirs] Error 1
make[1]: *** [subdirs] Error 1
make: *** [libraries] Error 1
 

Just Got Here
Threads 1
Posts 3
3:05:18 AM PDT - Wed, Jul 18th 2012  
Thanks for suggestion Bert.

Adding that environment variable has changed things, but still bit getting build to complete. Now get below in last lines of make.log:

/home/cczcb/NWChem/nwchem-6.1.1-src/bin/LINUX64/depend.x -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include > dependencies
make[2]: warning: -jN forced in submake: disabling jobserver mode.
touch include_stamp
Making include_stamp in /home/cczcb/NWChem/nwchem-6.1.1-src/src/dangchang
make[2]: warning: -jN forced in submake: disabling jobserver mode.
/home/cczcb/NWChem/nwchem-6.1.1-src/bin/LINUX64/depend.x -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include > dependencies
make[2]: warning: -jN forced in submake: disabling jobserver mode.
touch include_stamp
Making include_stamp in /home/cczcb/NWChem/nwchem-6.1.1-src/src/ccca
make[2]: warning: -jN forced in submake: disabling jobserver mode.
/home/cczcb/NWChem/nwchem-6.1.1-src/bin/LINUX64/depend.x -I../basis -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include > dependencies
make[2]: warning: -jN forced in submake: disabling jobserver mode.
touch include_stamp
Making libraries in basis
make[1]: warning: -jN forced in submake: disabling jobserver mode.
make[1]: *** No rule to make target `/home/cczcb/NWChem/nwchem-6.1.1-src/src/include/stdio.fh', needed by `/home/cczcb/NWChem/nwchem-6.1.1-src/lib/LINUX64/libnwcutil.a(basis.o)'. Stop.
make: *** [libraries] Error 1

Any ideas ?
thanks,
Colin
 

  • Bert Forum:Admin, Forum:Mod, NWChemDeveloper, bureaucrat, sysop
Forum Vet
Threads 3
Posts 391
12:51:34 PM PDT - Wed, Jul 18th 2012  
Try copying src/util/*fh src/include/. and hit make again.

Bert
Quote:Colinbann  Jul 18th 10:05 am
Thanks for suggestion Bert.

Adding that environment variable has changed things, but still bit getting build to complete. Now get below in last lines of make.log:

/home/cczcb/NWChem/nwchem-6.1.1-src/bin/LINUX64/depend.x -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include > dependencies
make[2]: warning: -jN forced in submake: disabling jobserver mode.
touch include_stamp
Making include_stamp in /home/cczcb/NWChem/nwchem-6.1.1-src/src/dangchang
make[2]: warning: -jN forced in submake: disabling jobserver mode.
/home/cczcb/NWChem/nwchem-6.1.1-src/bin/LINUX64/depend.x -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include > dependencies
make[2]: warning: -jN forced in submake: disabling jobserver mode.
touch include_stamp
Making include_stamp in /home/cczcb/NWChem/nwchem-6.1.1-src/src/ccca
make[2]: warning: -jN forced in submake: disabling jobserver mode.
/home/cczcb/NWChem/nwchem-6.1.1-src/bin/LINUX64/depend.x -I../basis -I/home/cczcb/NWChem/nwchem-6.1.1-src/src/tools/install/include > dependencies
make[2]: warning: -jN forced in submake: disabling jobserver mode.
touch include_stamp
Making libraries in basis
make[1]: warning: -jN forced in submake: disabling jobserver mode.
make[1]: *** No rule to make target `/home/cczcb/NWChem/nwchem-6.1.1-src/src/include/stdio.fh', needed by `/home/cczcb/NWChem/nwchem-6.1.1-src/lib/LINUX64/libnwcutil.a(basis.o)'. Stop.
make: *** [libraries] Error 1

Any ideas ?
thanks,
Colin
 

Just Got Here
Threads 1
Posts 3
6:43:59 AM PDT - Fri, Jul 20th 2012  
Thanks again, now build works!  

Now to do some testing, and try build using optimised Blas (usually use MKL).

help much appreciated,

Colin
 


Forum >> NWChem's corner >> Compiling NWChem
tengh
关注
  • 0
    点赞
  • 踩
  • 0
    收藏
    觉得还不错? 一键收藏
  • 0
    评论
专栏目录
错误代码 missing-method 错误原因: 缺少方法名参数_Makefile中常见的错误信息
weixin_29007809的博客
01-22 2757
make 执行过程中所产生错误并不都是致命的，特别是在命令行之前存在 "-"、或者 make 使用 "-k" 选项执行时。make 执行过程的致命错误都带有前缀字符串 "***"。错误信息都有前缀，一种是执行程序名作为错误前缀(通常是 "make")；另外一种是当 Makefile 本身存在语法错误无法被 make 解析并执行时，前缀包含了 Makefile 文件名和出现错误的行号。在下...
构建JDK8报错了，哪位大神能帮忙看下什么原因 ？
WoZhiShuoLiangJuHua的博客
08-23 3361
构建JDK8报错了，哪位大神能帮忙看下什么原因 ？
make的常见错误信息
u012849539的博客
10-02 8142
错误信息都有前缀，一种是执行程序名作为错误前缀（通常是“make”）；另外一种是当Makefile本身存在语法错误无法被make解析并执行时，前缀包含了makefile文件名和出现错误的行号。 在下述的错误列表中，省略了普通前缀： [FOO] Error NN [FOO] signal description 这类错误并不是make的真正错误。它表示make检测到make所调用的作为执行命令的程序返回一个非零状态（Error NN），或者此命令程序以非正常方式退出（携带某种信号）。 如果错误信息中没有附加“
Makefile （12）— Makefile 常见的错误信息
stephenbruce的博客
04-09 3502
make执行过程中所产生错误并不都是致命的，特别是在命令行之前存在、或者make使用-k选项执行时。make 执行过程的致命错误都带有前缀字符串***。错误信息都有前缀，一种是执行程序名作为错误前缀（通常是make另外一种是当Makefile本身存在语法错误无法被make解析并执行时，前缀包含了Makefile文件名和出现错误的行号。这类错误并不是make的真正错误。它表示make检测到make所调用的作为执行命令的程序返回一个非零状态（Error NN。
Nwchem-6.6.revision27746-src.2015-10-20.tar.gz
01-20
Nwchem (Linux)
nwchempy:使用python解析nwchem输出
05-26
nwchempy是一个基于Python的开源库，专门设计用于解析nwchem（Northwest Computational Chemistry Package）软件产生的输出文件。nwchem是一款强大的分子模拟软件，广泛应用于量子化学计算，它提供了多种方法来解决...
nwchemgit.github.io：NWChem网站
02-16
nwchemgit.github.io：NWChem网站
chemPackage：创建用于操作NWChem输入和输出文件的Python包
02-28
创建Python包以处理NWChem输入和输出文件 依存关系 截至目前，像numpy这样的常规包之外的唯一依赖项是natsort 。 可以通过pip或conda找到并安装此软件包 附加条款 该软件包是高度实验性的。 甚至在我从python2快速...
DFT的matlab源代码-aiida-nwchem:NWChem的AiiDA插件
05-26
AiiDA-NWChem是NWChem代码的AiiDA插件。 该插件为以下计算任务提供支持： 活力 优化 频率 ...使用以下理论类型： 科幻小说 dft NWPW乐队 西北太平洋水文计划 传统文化表现形式 该插件的2.x +版本仅与aiida-core v1.0...
Fortran笔记，函数
qq_21091051的博客
01-26 1112
参考资料：彭国论，《Fortran 95程序设计》，2002年 1. 子程序的使用 把经常使用、具有特定功能的程序代码独立出来，封装成子程序（SUBROUTINE）。 通过CALL命令即可调用这段代码； 子程序之间也可相互调用； 子程序可位于主程序（PROGRAM）之前或之后； 子程序以SUBROUTINE开头，并附子程序的名称，以END或END SUBROUTINE结束； 子程序的最后一个命令通常是"RETURN"，表示程序要“返回”原来调用它的地方来继续执行程序。RETURN命令也可省略，
Fortran初学笔记1：复数声明的错误及改正
qq_33037143的博客
12-03 3981
初学Fortran,编写一个求解二元一次方程的代码，用到复数类型，使用 complex(-b/(2*a),sqrt(-deta)/(2*a)) 格式出现错误，改为 cmplx(-b/(2*a),sqrt(-deta)/(2*a)) 没有错误，原因目前未知。 原代码： program main implicit none !定义变量 real(8) :: a,b,c,x,...
suse gfotran 安装备忘
ooshanshui的专栏
04-29 1164
rand（）随机数的产生在fortran中不是标准库的函数，ifort编译含rand（）的f90文件会提示  error #6404: This name does not have a type, and must have an explicit type.   [RAND] random_number是标准库的一个函数，可以在源码中使用自定义的rand（），则可通过ifort编译
Intel Fortran Compiler
资源备忘
08-20 7987
http://wiki.seas.harvard.edu/geos-chem/index.php/Intel_Fortran_Compiler Intel Fortran Compiler This page contains information about the Intel Fortran Compiler (aka "IFORT" compiler). The I
在组内集群编译nwchem
jslove1997的博客
02-20 1845
1. 编译openblas 这是nwchem编译需要使用的库，从官网上下载源代码。 #编译 make #安装到指定路径 make install PREFIX=/home/jrf/tools/openblas 2.编译ucx 要使用的OpenMPI为4.0.1版本 所以使用ucx连接。 最开始下载和编译ucx通讯模块。 此处参考 https://github.com/openucx/ucx/wi...
超算中心用的linux代码,正确使用超算中心，5分钟搞定vasp5.4.4编译
weixin_39673601的博客
05-12 1650
各位老师，请问在中科院“元”超算平台上编译出现这样的错误，如何解决呢？mkdir build/std ; \cp src/makefile src/.objects makefile.include build/std ; \make -C build/std VERSION=std allmkdir: cannot create directory `build/std': File exist...
(转)make的常见错误信息
活到老、学到老
08-26 2663
(转)make的常见错误信息 make执行过程中所产生错误并不都是致命的；特别是在命令行之前存在“-”、或者make使用“-k”选项执行时。make执行过程的致命错误都带有前缀字符串“***”。错误信息都有前缀，一种是执行程序名作为错误前缀（通常是“make”）；另外一种是当Makefile本身存在语法错误无法被make解析并执行时，前缀包含了make
量子化学计算python
最新发布
07-28
NWChem可以在大多数计算平台上使用，它使用标准量子力学描述电子波函或密度，可以计算分子和周期性系统的特性，还可以进行经典分子动力学和自由能模拟。\[2\] 这些量子化学计算框架提供了丰富的功能和灵活性，可以...

“相关推荐”对你有帮助么？

  • 非常没帮助
  • 没帮助
  • 一般
  • 有帮助
  • 非常有帮助
提交
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值