Intel® Fortran Compiler

FPP for non-Fortran text files

Hi

Are there any pitfalls when using FPP to manipulate non-Fortran text files?

We have some text data files. We want to build multiple versions of the data files from one master copy, using FPP and #ifdef #else #endif. I've tried this and it does work. But I am aware that FPP was not designed to do this, but only to preprocess Fortran code. I am worried that there could be pitfalls: certain text sequences that will cause behaviour that I don't want. 

LIBCMT.lib(_fpint_.obj) : fatal error LNK1112: module machine type 'x64' conflicts with target machine type 'IA64'

Hello All,

I have been compiling FOTRAN code with with a 32-bit compiler successfully for a finite element software called LS-DYNA for years, but because of some problems with large models I have decided to switch to a 64-bit version of LS-DYNA. I am now running into a compile error.

My machine is: Processor: Intel i7-6700HQ. OS: Windows 10 64-bit, x64-based processor.

I am using the following:

Conflict between IMSL and MPI

I am trying to divide my fortran code into several parts and I want to parallelize each part by using MPI. For each part, I use IMSL library to solve an optimization problem (use BCONF). However, I find that IMSL library has its own subroutines about MPI and it does not allow me to call the standard MPI start subroutine "Call MPI_INIT(ierror)". It just gives me an fatal error and ends the program.

I give two examples to illustrate the issue. 

 

Example 1, print "Hello World " from each node: 

program main
   use mpi

   implicit none

DPD200247916 (debug reformat display type) followup

Hi

 

I saw an old post (earlier this year) that referenced an existing problem, DPD200247916.  This relates to not being able to use the debugger (i.e., VS2015) with Intel Fortran XE2016 to view variables in alternative formats - for example REAL*8 MyVar displayed in debugger as string using MyVar,s.

 

First is there an update on this, is it working again and if so which update has the fix?

 

Second, can I do this with arrays as well?  So similar to the above example, can I have REAL*8 MyVar(100) and can I show all array contents as strings?

pointers, arrays of different dimensions, and derived types

Hello! Right now I'm hoping to point data that is inside a derived type to an array (target) that is of different dimensions and is passed to the rest of the world (e.g. Python) via iso_c_binding.  The application is a ray trace program, where each derived type is passed to and computed in a separate thread via openMP, and all the results are aggregated into the c type arrays.  I'd like to point 1D arrays inside the derived types to subsets of higher-D (more than 3D) arrays.

initialize undefined variables

Having uninitialized variables is, of course, poor practice but inevitable.  Using options like Qzero is not much better.  Zero is too forgiving.  On legacy hardware (e.g., CDC 7600) I would always preset memory to infinite or indefinite values because the use of such values in ANY arithmetic operation immediately threw a MODE error.  The question is then twofold:

(1) is there a way to set undefined variables scalars and arrays to a characteristic value that if used in an arithmetic operation will cause an immediate error, and

Inverse matrix with Intel MKL

Dear All,

I am trying to migrate my code from Compaq Visual Fortran 6.5 to Visual Studio 2015 with Intel Fortran Compiler. I understand that IMSL used to be built in CVF. Now however, IMSL is an additional add-on for Visual Studio. I am having problems trying to change the program for inversing matrix using MKL.

The original code was:

Creating a dll. file

Hello,

I am trying to create a dll file for my library in TRNSYS with the intel visual compiler and microsoft visual studio. When compiling I receive two errors "Compilation Aborted (code1)" and "error #6580: Name in only-list does not exist.", which can be see from the build log.

TRNSYS Version 17.01.0028 and Microsoft Visual Studio Community 2015 Version 14.0.25431.01 update 3.

 Any help with this issue would be highly appreciated. The source file is also included.

Segmentation fault when coarray attribute omitted for the dummy argument

Hi,

The following code gives a segmentation fault.

This error is triggered when I call a procedure (ProceVersion2) passing a coarray actual argument but the associated dummy argument has NOT the coarray attribute.

If the same procedure is written with the coarray attribute for the dummy argument (ProceVersion1), then there is no error.

Surprisingly, if both procedures are called, there is also no error.

Also, if the allocatable attribute is removed, there is no error

Thanks for looking at this issue.

Subscribe to Intel® Fortran Compiler