I was wondering if there is (or there are plans for) a feature built in to the Intel compilers that would combine compiler optimization reports with source code listings, as with the VecAnalysis script provided by Intel. That script is really handy, but unfortunately it does not seem to work with anything later than the 13.1 compiler release. This script makes the vectorization and other reports much more understandable.
In an application I am opening different files residing in different folders. To perform the assignment, I try to force Windows to display the correct folder. in the way that if I want to open an .INP file, the FileOpen dialog should list the content of the INP directory aso. However, the FileOpen dialog box lists the file content of the most recent directory. In the example below, the files in the "InputDir" should be listed and this happens. But next, when I will open a PLT file (not shown here), the dialog box list the files in the InputDir, and vice versa...
IVF XE16 Update 1 works fine with my Visual Studio 2013.
But with Visual Studio 2015 I can not link (I can only compile).
Error message --> LINK : fatal error LNK1104: cannot open file "ï»¿/OUT:Release\Console1.exe"
VS 2013 and VS 2015 are installed on the same computer (parallel).
I have also IVF XE16 Update 1 deinstalled and new installed. Still the same problem.
Any ideas what I can do ?
So, I downloaded the latest installer image <m_fcompxe_2016.1.043.dmg>. Installed the latest Xcode (7.1.1) and also the command line developer tools. Then opened the Intel Fortran installer image and launched the installer application <m_fcompxe_2016.1.043.app>. I was logged in as Administrator, and am running OS X El Capitan 10.11.1. I have a valid product serial number for Intel® Parallel Studio XE Composer Edition for Fortran OS X*. The installer application provided a screen to select between two options: 1. Install as Administrator. 2. Install as root.
I have tried compiling a large code under both Visual Studio and Cygwin environments using this FORTRAN compiler. It is too slow to be usable. Whereas ifort 12.1(8) took 1-3 seconds per routine this version takes ~5 minutes per routine. I stopped the build after a few routines. There is something very wrong with this behavior. The install went normally after installing VS 2016 and Cygwin.
I was trying to learn how to add a dialog box to my 25 year old Fortran code. I was trying to follow along with this guide:
I have this version of Visual Studio:
Microsoft Visual Studio 2010
Version 10.0.40219.1 SP1Rel
Microsoft .NET Framework
Version 4.5.51209 SP1Rel
Installed Version: IDE Standard
Microsoft Visual Web Developer 2010 01011-532-2002361-70772
Microsoft Visual Web Developer 2010
I am trying to compile CP2K using the Intel(R) Fortran 64 Compiler XE compiler. To optimize my code I have been playing with some of the compilation flags. When I add the flags
-xSSE2 -no-ipo -fpp -free
My compilation fails with an error
I came across the following awkward behaviour:
an argument being used in ACOS function displayed in the debugger shows a value 1.00000000000000 but it resulted an a Nan result.
When displayed as hexadecimal, the argument shows a value #3FF0000000000001.
A PARAMETER initialised to 1.0d+00 displays as 1.00000000000000 and when displayed as hexadecimal shows #3FF0000000000000.
I want to look at a module variable, so following the documented technique for doing this, in the Watch window I type
The debugger replies that the variable is not recognised, even though the variable is defined as a PARAMETER in the module and calculations up to that point use its value.
The module is in a separate Fortran file. Is that the problem?
I am using VS 2012 Pro.