Intel® Visual Fortran Compiler for Windows*

Specification expressions, DIMENSION ordering and a mischievous user

With ifort 16.0 Build 20151021 (that's update 1, I think) this program prints `1 2`.  That doesn't make much sense - I think a strict reading of the standard suggests the program is conforming (noting that the marked line does not result in a reference to either the value of an element of `a` or depend on an array bound of `a`) and the output should be `2 2` (which is what current gfortran does). 

Failing that it might be standard interp territory.  Either way... enjoy!

extra under line generate when convert from cvf to xe2015

hello !

i recently install  xe2015  in vs 2012.

when i convert code that written in compaq visual fortran (right click on program name in solution then convert to compaq...), an extra underline added at the end of my program name that Clicked. i delete underline and then run correctly but i must do this for my all converted code at each time because of an error of extra underline given. any one know the solution?

with regard.

ERROR: entry point _erfq could not be located in libifcoremdd.dll

Question: Has a windows 8.1 update recently (like last 3 days?) somehow adversely affected the Dynamic Link for debug compilations? I have a large solution with 4 projects, (EXE & 3 DLLs) compiling/running in Win8.1., Ia32, Visual Studio 2010, w_fcompxe_2015.5.280 all latest greatest. (i have and use 2016 and may see if it occurs in that but this project already distributed w/2010 so delaying having to update redistro libraries)

Unresolved contained procedures at link in XE2016 Update 1

I have an abstract type and 3 concrete types derived from it. The abstract type declares some deferred internal procedures and the derived types implement them. The project compiles fine under Visual Studio 2015 but link fails with unresolved references in a subroutine in a separate source. But that routine does not call any of the unresolved references.

The link error is only related to a small percentage of similar derived type structures in my project.

It is a clean build before anyone suggests this.

"Find all references" makes VS2015 stall


"Find all references" on any symbols in the VS2015 code editor makes VS stop running and inactive.

My environment is as below:

Windows 7 x64

Microsoft Visual Studio Professional 2015 Version 14.0.23107.0 D14REL
Intel® Parallel Studio XE 2016 Update 1 Composer Edition for Fortran Windows*     Package ID: w_comp_lib_2016.1.146

New comment for Intel® Visual Fortran Compiler for Windows* Forum topic: ........


My in-box is filled up by messages "New comment for Intel® Visual Fortran Compiler for Windows* Forum topic:....."  I used to get a message when I got comments to issues I had raised, however I now receive notification for all the others as well. Have I by accident changed settings on my dashboard? How do I get rid of it ?


MKL in Windows / Linux

Dear all,

I have developer a program in ifort for windows. The program uses cholesky decomposition (dpotrf), and for that I am using MKL. Recently I had the need to use my program on Linux so I compiled it for Linux and carried out some testing. I observed that I got a substantial difference in results and I have traced this back to the dpotrf routine. 

Initialization of local variables in subroutines

In Fortran 77 is it possible to globally ensure that the local variables in a subroutine are not preserved (which seems to be the default), i.e. without initializing all variables and arrays individually?

My problem is that calling a subroutine repeatedly results in different answers.
For example, if I send it an input value x, it returns the correct value y.
However, if I put it inside a loop and send it values x-1 and then x, it returns the correct output for x-1, but for x it now returns an incorrect value different to y.

Intel® Visual Fortran Compiler for Windows* abonnieren