Turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Mute
- Printer Friendly Page

Highlighted
##

Wu__Yongli

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-28-2019
02:54 AM

55 Views

Calling Pardiso failed after enabling OpenMP in Visual studio

Hi,

Pardiso in my **Fortran** program works well if I do not use OpenMP.

However, after I change the setting in Visual studio for **enabling OpenMP**, by:

** Project > Configuration Properties > Fortran > Language > Process OpenMP Directives > Generate Parallel Code (/Qopenmp)**

The error comes at **calling Pardiso**, showing:

** forrtl: severe (157): Program Exception - access violation**

Does anyone know how to solve this problem?

Thanks a lot!

Yongli

7 Replies

Highlighted
##

MariaZh

Employee

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-29-2019
02:41 PM

55 Views

Can you please provide more details about your MKL version and reproducer for the case if possible?

Best regards,

Maria

Highlighted
##

Wu__Yongli

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-29-2019
04:36 PM

55 Views

Hi Maria,

Thank you very much for your reply.

My MKL version: Intel® MKL 2018

Besides, I attached the code for reproducer. The code was downloaded from (http://www.pardiso-project.org) with minor modifications.

If I set Project > Configuration Properties > Fortran > Language > Process OpenMP Directives > Generate Sequential Code (/Qopenmp_stubs), the program works.

However, the problem (as previously described) comes when the above setting is changed as " ....> Process OpenMP Directives > Generate Parallel Code (/Qopenmp)".

Thanks and best regards,

Yongli

C---------------------------------------------------------------------- C Example program to show the use of the "PARDISO" routine C for symmetric linear systems C -------------------------------------------------------------------- C This program can be downloaded from the following site: C http://www.pardiso-project.org C C (C) Olaf Schenk, Institute of Computational Science C Universita della Svizzera italiana, Lugano, Switzerland. C Email: olaf.schenk@usi.ch C -------------------------------------------------------------------- PROGRAM pardiso_sym use omp_lib IMPLICIT NONE C.. Internal solver memory pointer INTEGER*8 pt(64) C.. All other variables INTEGER maxfct, mnum, mtype, phase, n, nrhs, error, msglvl INTEGER iparm(64) INTEGER ia(9) INTEGER ja(18) REAL*8 dparm(64) REAL*8 a(18) REAL*8 b(8) REAL*8 x(8) REAL*8 y(8) INTEGER i, j, idum, solver REAL*8 waltime1, waltime2, ddum, normb, normr C.. Fill all arrays containing matrix data. DATA n /8/, nrhs /1/, maxfct /1/, mnum /1/ DATA ia /1,5,8,10,12,15,17,18,19/ DATA ja 1 /1, 3, 6, 7, 2 2, 3, 5, 3 3, 8, 4 4, 7, 5 5, 6, 7, 6 6, 8, 7 7, 8 8/ DATA a 1 /7.d0, 1.d0, 2.d0, 7.d0, 2 -4.d0, 8.d0, 2.d0, 3 1.d0, 5.d0, 4 7.d0, 9.d0, 5 5.d0, 1.d0, 5.d0, 6 0.d0, 5.d0, 7 11.d0, 8 5.d0/ C .. set right hand side do i = 1, n b(i) = 1.d0 end do C C .. Setup Pardiso control parameters und initialize the solvers C internal adress pointers. This is only necessary for the FIRST C call of the PARDISO solver. C mtype = -2 ! unsymmetric matrix symmetric, indefinite solver = 10 ! use sparse direct method C.. Reordering and Symbolic Factorization, This step also allocates C all memory that is necessary for the factorization phase = 11 ! only reordering and symbolic factorization msglvl = 1 ! with statistical information iparm(33) = 1 ! compute determinant !write(*,*) "proc nums:", omp_get_num_procs() call mkl_set_dynamic(0) call omp_set_num_threads(2) call mkl_set_num_threads(2) IPARM(3) = 2 CALL pardiso (pt, maxfct, mnum, mtype, phase, n, a, ia, ja, 1 idum, nrhs, iparm, msglvl, ddum, ddum, error, dparm) WRITE(*,*) 'Reordering completed ... ' IF (error .NE. 0) THEN WRITE(*,*) 'The following ERROR was detected: ', error STOP END IF WRITE(*,*) 'Number of nonzeros in factors = ',iparm(18) WRITE(*,*) 'Number of factorization MFLOPS = ',iparm(19) C.. Factorization. phase = 22 ! only factorization CALL pardiso (pt, maxfct, mnum, mtype, phase, n, a, ia, ja, 1 idum, nrhs, iparm, msglvl, ddum, ddum, error, dparm) IF (iparm(33).EQ.1) THEN write(*,*) 'Log of determinant is ', dparm(33) ENDIF WRITE(*,*) 'Factorization completed ... ' IF (error .NE. 0) THEN WRITE(*,*) 'The following ERROR was detected: ', error STOP ENDIF C.. Back substitution and iterative refinement iparm(8) = 1 ! max numbers of iterative refinement steps phase = 33 ! only solve CALL pardiso (pt, maxfct, mnum, mtype, phase, n, a, ia, ja, 1 idum, nrhs, iparm, msglvl, b, x, error, dparm) WRITE(*,*) 'Solve completed ... ' WRITE(*,*) 'The solution of the system is ' DO i = 1, n WRITE(*,*) ' x(',i,') = ', x(i) END DO call sleep(3) END

Highlighted
##

MariaZh

Employee

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-29-2019
04:54 PM

55 Views

First thing that I've noticed: the Intel(R) MKL PARDISO interface is a little bit different, there is no dparm parameter, for instance.

Please refer to the https://software.intel.com/en-us/mkl-developer-reference-fortran-pardiso for the detailed description.

Can you please adjust your code accordingly and double check if the issue is still there?

Best regards,

Maria

Highlighted
##

Wu__Yongli

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-29-2019
05:07 PM

55 Views

Hi Maria,

Thank you very much for your quick reply and suggestions.

Then, I will check it as your suggestion first.

Thanks and best regards,

Yongli

Zhukova, Maria (Intel) wrote:Hi Yongli,

First thing that I've noticed: the Intel(R) MKL PARDISO interface is a little bit different, there is no dparm parameter, for instance.

Please refer to the https://software.intel.com/en-us/mkl-developer-reference-fortran-pardiso for the detailed description.

Can you please adjust your code accordingly and double check if the issue is still there?Best regards,

Maria

Highlighted
##

Wu__Yongli

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-29-2019
07:20 PM

55 Views

Hi Maria,

The interface has been adjusted to be the same as the attached code.

However, the problem remains.

Your further suggestions will be highly appreciated.

Best regards,

Yongli

C---------------------------------------------------------------------- C Example program to show the use of the "PARDISO" routine C for symmetric linear systems C -------------------------------------------------------------------- C This program can be downloaded from the following site: C http://www.pardiso-project.org C C (C) Olaf Schenk, Institute of Computational Science C Universita della Svizzera italiana, Lugano, Switzerland. C Email: olaf.schenk@usi.ch C -------------------------------------------------------------------- PROGRAM pardiso_sym use omp_lib IMPLICIT NONE C.. Internal solver memory pointer INTEGER*8 pt(64) C.. All other variables INTEGER maxfct, mnum, mtype, phase, n, nrhs, error, msglvl INTEGER iparm(64) INTEGER ia(9) INTEGER ja(18) REAL*8 dparm(64) REAL*8 a(18) REAL*8 b(8) REAL*8 x(8) REAL*8 y(8) INTEGER i, j, perm, solver REAL*8 waltime1, waltime2, ddum, normb, normr C.. Fill all arrays containing matrix data. DATA n /8/, nrhs /1/, maxfct /1/, mnum /1/ DATA ia /1,5,8,10,12,15,17,18,19/ DATA ja 1 /1, 3, 6, 7, 2 2, 3, 5, 3 3, 8, 4 4, 7, 5 5, 6, 7, 6 6, 8, 7 7, 8 8/ DATA a 1 /7.d0, 1.d0, 2.d0, 7.d0, 2 -4.d0, 8.d0, 2.d0, 3 1.d0, 5.d0, 4 7.d0, 9.d0, 5 5.d0, 1.d0, 5.d0, 6 0.d0, 5.d0, 7 11.d0, 8 5.d0/ C .. set right hand side do i = 1, n b(i) = 1.d0 end do C C .. Setup Pardiso control parameters und initialize the solvers C internal adress pointers. This is only necessary for the FIRST C call of the PARDISO solver. C mtype = -2 ! unsymmetric matrix symmetric, indefinite solver = 10 ! use sparse direct method C.. Reordering and Symbolic Factorization, This step also allocates C all memory that is necessary for the factorization phase = 11 ! only reordering and symbolic factorization msglvl = 1 ! with statistical information iparm(33) = 1 ! compute determinant !write(*,*) "proc nums:", omp_get_num_procs() call mkl_set_dynamic(0) call omp_set_num_threads(2) call mkl_set_num_threads(2) IPARM(3) = 2 call pardiso (pt, maxfct, mnum, mtype, phase, n, a, ia, ja, 1 perm, nrhs, iparm, msglvl, b, x, error) WRITE(*,*) 'Reordering completed ... ' IF (error .NE. 0) THEN WRITE(*,*) 'The following ERROR was detected: ', error STOP END IF WRITE(*,*) 'Number of nonzeros in factors = ',iparm(18) WRITE(*,*) 'Number of factorization MFLOPS = ',iparm(19) C.. Factorization. phase = 22 ! only factorization call pardiso (pt, maxfct, mnum, mtype, phase, n, a, ia, ja, 1 perm, nrhs, iparm, msglvl, b, x, error) IF (iparm(33).EQ.1) THEN write(*,*) 'Log of determinant is ', dparm(33) ENDIF WRITE(*,*) 'Factorization completed ... ' IF (error .NE. 0) THEN WRITE(*,*) 'The following ERROR was detected: ', error STOP ENDIF C.. Back substitution and iterative refinement iparm(8) = 1 ! max numbers of iterative refinement steps phase = 33 ! only solve call pardiso (pt, maxfct, mnum, mtype, phase, n, a, ia, ja, 1 perm, nrhs, iparm, msglvl, b, x, error) WRITE(*,*) 'Solve completed ... ' WRITE(*,*) 'The solution of the system is ' DO i = 1, n WRITE(*,*) ' x(',i,') = ', x(i) END DO call sleep(3) END

Zhukova, Maria (Intel) wrote:

Hi Yongli,

First thing that I've noticed: the Intel(R) MKL PARDISO interface is a little bit different, there is no dparm parameter, for instance.

Please refer to the https://software.intel.com/en-us/mkl-developer-reference-fortran-pardiso for the detailed description.

Can you please adjust your code accordingly and double check if the issue is still there?Best regards,

Maria

Highlighted
##

Gennady_F_Intel

Moderator

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-29-2019
09:54 PM

55 Views

Highlighted
##

Wu__Yongli

Beginner

- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Email to a Friend
- Report Inappropriate Content

04-30-2019
02:03 AM

55 Views

Dear Maria and Gennady,

Much appreciated for your kind replies and suggestions.

Now, the code works even the version is 2018. I just checked and adjusted the previous code according to the examples in the folder of the Intel MKL installation directory: • examples/solverf/source.

Best regards,

Yongli

For more complete information about compiler optimizations, see our Optimization Notice.