error while using MKL library

[Tai_Q_]

Dear all,

I am running a program that has been running many times in a cluster.

Maybe because the cluster has been through software upgrade, there are errors while running the executable file a.out.

There is no problem for compiling and linking. Just error will show up while run the program halfway..

forrtl: error (65): floating invalid

Now we are using intel/17.0.4, impi/17.0.3.

      call ZGETRF( N_LEN_2, N_LEN_2, BQ , N_LEN_2, IPIV , INFO )
 
      call ZGETRI( N_LEN_2, BQ, N_LEN_2, IPIV, WORK, N_LEN_2, INFO )

The first subroutine

ZGETRF

is fine. But when it comes to the second function

ZGETRI. There is always a floating invalid error. 

I just do not understand. Because the input of ZGETRI are just the output of ZGTRF.

This program has been used for such a long time!

[mecej4]

 But when it comes to the second function ZGETRI, there is always a floating invalid error.

The MKL examples include a test program zgetrix.f (and the data file zgetrix.d), which makes calls to ZGETRF and ZGETRI. I tested this program using the 16.0.4, 17.0.4 and 18.0.0 compilers on Windows. No errors, all give identical results.

I suspect that the problem lies elsewhere in your program, but the symptom (Floating Invalid) is only occurring and getting noticed later in the program. As with most of the BLAS/Lapack routines, problems can occur only for certain matrix sizes and the values passed, with other sizes and values producing correct results.

Please submit a working example (code, data, instructions to build and run, CPU/OS information).

MODERATORS: Please consider moving this thread to the MKL forum.