3.1.7.11.3. OpenMP

The following is an example of compiling a Fortran program of a single node job (OpenMP).

  1. Prepare source program.
    Prepare sample program as /home/system/sample/Fortran/openmp/sample_omp.f.
    (Download sample code)
 1      parameter(n=11776)
 2      real*8 a(n+1,n),b(n+1,n),c(n+1,n)
 3      real*8 t0,t1,second
 4!
 5!  print message
 6!
 7      write(6,500)
 8      write(6,510) n,n
 9 500  format(" *** matrix multiply *** ")
10 510  format(' (NPROW,NPCOL) : (',I6,',',I6,')')
11
12!
13!  initialize matrix
14!
15!$omp parallel do default(private) shared(a,b,c)
16      do j=1,n
17        do i=1,n
18          a(i,j)=0.1d0
19          b(j,i)=0.3d0
20          c(i,j)=0.0d0
21        end do
22      end do
23!$omp end parallel do
24!
25!  do matrix multiply
26!
27      t0 = second()
28!$omp parallel do default(private) shared(a,b,c)
29      do j=1,n
30        do i=1,n
31          do k=1,n
32            c(i,j) = c(i,j) + a(i,k) * b(k,j)
33          end do
34        end do
35      end do
36!$omp end parallel do
37      t1 = second()
38!
39!  print result
40!
41      write(6,520) c(1,1)
42      write(6,530) t1-t0
43  520 format(" c(1,1)=",g10.4)
44  530 format(" time :",f10.2," sec")
45      end
46!----
47      function second()
48      real*8 second,t
49      call gettod(t)
50      second=t*1.e-6
51      end

  1. Compile sample program.

[_LNlogin]$ frtpx -V -Kfast,openmp,optmsg=2 -Nlst=t -o sample_omp sample_omp.f
frtpx: Fujitsu Fortran Compiler 4.1.0 tcsds-1.2.24
jwd_fortpx: Fujitsu Fortran Compiler 4.1.0 (Feb 26 2020 07:41:18)
Fortran diagnostic messages: program name(main)
  jwd6002s-i  "sample_omp.f", line 17: SIMD conversion is applied to DO loop.
  jwd8204o-i  "sample_omp.f", line 17: This loop is software pipelined.
  jwd8205o-i  "sample_omp.f", line 17: The software-pipelined loop is chosen at run time when the iteration count is greater than or equal to 48.
  jwd8220o-i  "sample_omp.f", line 29: Optimizations is performed in this program unit with possibility of side effects. See informational messages below to determine which such optimizations have been performed.
  jwd8331o-i  "sample_omp.f", line 29: This DO loop was changed to the library call(matmul).
GNU assembler version 2.30 (aarch64-linux-gnu) using BFD version version 2.30-49.el7
GNU ld version 2.30-49.el7
  Supported emulations:
   aarch64linux
   aarch64elf
   aarch64elf32
   aarch64elf32b
   aarch64elfb
   armelf
   armelfb
   aarch64linuxb
   aarch64linux32
   aarch64linux32b
   armelfb_linux_eabi
   armelf_linux_eabi
   i386pep
   i386pe
flistpx: Fujitsu Listing Processor 4.1.0 (Jan  9 2020 14:46:36)
  1. Prepare job script.
    Job script sample is prepared as /home/system/sample/Fortran/openmp/job_openmp.sh.
    (Download sample code)
#!/bin/sh
#PJM -L "node=1"
#PJM -L "rscgrp=small"
#PJM -L "elapse=10:00"
#PJM -x PJM_LLIO_GFSCACHE=/vol000N
#PJM -g groupname
#PJM -s

# execute job
export OMP_NUM_THREADS=12
./sample_omp

  1. Submit a job with pjsub command.

[_LNlogin]$ pjsub job_openmp.sh
[INFO] PJM 0000 pjsub Job 113 submitted.
  1. Check execution result.
    The standard output is output as Job name.Job ID.out.
[_LNlogin]$ cat job_openmp.sh.113.out
 *** matrix multiply ***
  (NPROW,NPCOL) : ( 11776, 11776)
  c(1,1)= 353.3
  time :     30.55 sec