Intel® Advisor XE

Advisor 2015 XE not working with Visual Studio 2015

I had installed Parallel Studio XE 2015 Update 5 on a Windows 10 Pro x64 machine.  Integration into Microsoft Visual Studio 2013 worked with Inspector, Amplifier, and Advisor showing up in the VS menu bar.  Integration into Microsoft Visual Studio 2015 worked only for Amplifier. How to fix Intel Inspector 2015 XE and Advisor 2015 XE with Visual Studio 2015 please help.
Thank you.

How to Change the Target type drop-down to Correctness Analysis


This is my first post to the forum.
I am testing advisor_xe_2016.1.0.423501 on a E5-2690 v3 cpu using the nqueens tutorial.
According to the html doc "Predict Parallel Data Sharing Problems", i should be able to:

Change Intel Advisor Project Properties:
In the Intel Advisor GUI, choose File > Project Properties....
In the Analysis Target tab:
Change the Target type drop-down to Correctness Analysis.

installing Intel Paralle Studio XE Composer Edition for Fortran and C++ Windows

Dell model:T7610 I have 64bit windows 7 We've been trying to install Intel Paralle Studio XE Composer Edition for Fortran and C++ Windows. The trouble we're having is that it will not complete the install. After the certificate install. The Intel splash screen will come up and look like it was about install but then the splash screen disappears.

MKL in C#


I'm testing MKL in a C# application.

I have to solve a linear system of complex matrix, so first of all I want to understand how the libraries works. 

I'm not expert of c++, and Marshaling too so I have tried to define the dll as best as I could. I have used the library installed with Parallel Studio XE.

 The example that I have done, doesn't work is there any one that can help me to understand why? 

advisor use with msvc++ or gcc/gfortran/g++ Windows

Advisor (beta update 2) seems to report reliably on gcc/gfortran/g++ loop vectorization and use of AVX/FMA, even on Windows, but doesn't display source code on Windows.

With MSVC++, Advisor displays source code correctly, but doesn't report vectorization (which occurs on about 50% of for loops which ICL or g++ can vectorize effectively).  In my examples, MSVC++ invariably chooses AVX-128 (possibly to avoid alignment analysis).  It does occasionally vectorize with AVX where it would not with SSE2.

inefficient remainder report

I looked into some of these reports.  It seems that it means simply that at least 25% of the samples fell in the remainder loop (for an Intel compilation with vectorization).   I looked into one with VTune and that sampling looked more believable, with a few samples falling in the vector remainder loop and none in the scalar remainder where Advisor complained.  Is a method needed to give Advisor better granularity for short runs or runs with time distributed over many vectorized loops?

broken advisor beta installation

If I try to install advisor beta update 1 or 2, the install fails and rolls back, saying that advisor version is already installed.  I don't know where to look for documentation on what is required to install it.  I believe all components of psxe will install if I remove advisor from the list.

When I last had Advisor installed, it reported as update 2 but had the reported faults of update 1 (inability to recognize vectorization in beta compilation).

订阅 Intel® Advisor XE