Intel® Fortran Compiler for Linux* and Mac OS X*

Problem with Array Assignment on v9.1 Fortran Compiler

Hi all,

I came across a problem with the following statement ( simplified for clarity) in a F95 code:

A%vector(:) = A%vector(:) * value

The array component is a vector (1D array) of size 3,200,000 elements of double complex values (~50MB of data). Using the array assignment above to scale the vector values....results in a segmentation fault....whereas when I re-coded the assigment as an explicit do loop i.e.

do i=1,size(A%vector)
A%vector(i) = A%vector(i) * value
end do

Need help understanding performance of nested DO loop

The loop below is part of a much larger program. It gets compiled
-02 -xW and I run on a dual-cpu AMD Opteron....profiling with gprof
shows 70 s for the loop as shown (nvar on outer loop).
Interchanging the order of loops increases the run time to 90
s. Typical value for nvar is 7 and ib,ie range is typically

First question is why the slowdown? Shouldn't the code run faster with the inner index varying fastest?

compile and link f90 files

hi folks!

at first I have to say that I'm not an expert in programming, I am more a common user.

but sometimes the user has to adapt his tools...

anyway: the point is that I used to work with the compaq visual studio environment and now I'm supposed to compile my code for linux (ifort) to get it started on our cluster.

why does the ifort compiler complain about .f90 files, that work with the visual compiler? ain't the syntax the same? maybe someone out there's got a valid strategy for such problems?



bug in ifort?

not sure this is the right place, however
I would like to report what seems to me a bug
in ifort.

The attached code works when compiled with

ifort ib.f90 -o ib

but does not when compiled with

ifort ib.f90 -r8 -o ib

My compiler:
ifort -V
Intel Fortran Compiler for Intel EM64T-based applications, Version 9.1 Build 20061101 Package ID:
Copyright (C) 1985-2006 Intel Corporation. All rights reserved.

Thanks, Best Regards
Marco Restelli

Multiple symbol definition errors at link time


I'm using Apple Xcode 2.4.1 and Intel Fortran v9.1.034 for Mac OS X running on a MacPro, OS: 10.4.8.

I am trying to link a dylib that contains a bunch of archive files of Fortran modules (compiled by me). I need to use the Apple "-all_load" linker flag to make sure all symbols from the Fortran archives get pulled into the dylib.

forrtl: severe (174): SIGSEGV, segmentation fault occurred

Dear all,
I have a problem with my code that I would you like you help me to understand.
It compiles well, but when I execute it, if the size of the coputation is "large" it gives me the following answer:

forrtl: severe (174): SIGSEGV, segmentation fault occurred

This is not really surprising. What I find ununderstandable for me is that if I compile with the -traceback options, the compiler tells me that the error happens on a line that opens a file for writing:


Subscribe to Intel® Fortran Compiler for Linux* and Mac OS X*