Intel® C++ Compiler

Speed Up shown in the optimization report but no speedup is shown while execution.


I had a for loop which had some branches due to which the loop was not a candidate of vectorization which I confirmed from the optimization report. I removed this branches using masking, and now the same for loop satisfies all the necessary requirement for a loop to be vectorized, like

1. No Branches or Jumps.

2. No Dependencies. e.t.c

shared libraries "" error


I am tryoing to compile and run helloflops2 from "example from "Intel Xeon Phi Coprocessor High-Performance Programming". I could compile the code using this command line:

   > icc -openmp -mmic -qopt-report-phase=vec -qopt-report=3  -O3 helloflops2.c -o helloflops2

Then, I scp "helloflops2" file to mic0:/root and try to run it using: "./helloflops2"  and I got this error:

./helloflops2: error while loading shared libraries: cannot open shared object file: No such file or directory

error code -1073741819

2>xilink: : error #10014: problem during multi-file optimization compilation (code -1073741819)
2>xilink: : error #10014: problem during multi-file optimization compilation (code -1073741819)

I am using the latest stable 15 version of ICC, and MSVC 2013.

What the hell this means? Shouldn't be numbers in error code dated back to 1980's ?!


Performance issue


I just finished to run a benchmark on two different machines, a laptop equipped with an Intel(R) Core(TM) i7-4810MQ CPU @ 2.80GHz  and the other is a desktop with an Intel(R) Core(TM) i7-5960X CPU @ 3.00GHz processor. Strangely the laptop outperforms the desktop of a factor grater than 3x. I am using the same compiler version with the same compiling options on the same OS version, i.e, gcc 5.1.0, "-march=native -O3" and Ubuntu 14.04. The code is sequential. Do you have any suggestion?

Best, Gabriele

/Qipo with 2500 files -- is that a problem?

I try to compile the ngspice SPICE program with the Intel compiler on
Windows 7 64bit, using Visual Studio 2013 Community. The machine has
32GB RAM. NGSPICE is a program with over 2500 files (2400 .obj in
the final result).

It is not possible to turn on /Qipo as it results in lots of [~44]
object files that can't be found (the compiler seems to be doing some
4 passes before giving up). With /Qip the code is some 20% slower
than with the old Microsoft compiler, which makes using the Intel
compiler somewhat pointless :-)

CUDA 7.5 sample failed

If you download latest CUDA 7.5 and compile the samples with Pre-Release (Beta) Package ID: w_comp_lib_2016.0.06 most of the Projects failed with error MSB3721.

MS Compiler
147>     Bibliothek "../../bin/win64/Debug/cuSolverRf.lib" und Objekt "../../bin/win64/Debug/cuSolverRf.exp" werden erstellt.
147>  cuSolverRf_vs2012.vcxproj -> C:\ProgramData\NVIDIA Corporation\CUDA Samples\v7.5\7_CUDALibraries\cuSolverRf\../../bin/win64/Debug/cuSolverRf.exe
========== Alles neu erstellen: 147 erfolgreich, 0 fehlerhaft, 0 übersprungen ==========


I noticed from the end-of-beta survey that option protect-parens was introduced, presumably to comply with that part of C and C++ standards while permitting optimization of sum reduction when /fp:fast is set (as it is by default in ICL).

protect-parens shows excellent results in my testing of C and Cilk(tm) Plus code, with performance gains up to 40%, even though several cases of pessimization in violation of parens had already been corrected (where previously /fp:source was needed to optimize).

Compilation of the same code under Win and Linux with Intel Compiler


I'm not sure if this is the proper forum to post on, so please excuse me if my question should have been posted elsewhere.

We are developing software that depends heavily on numerical computation and we want to use intel compiler to achieve a boost in our performance. So far, the platform we have been developing on is Linux.

We consider purchasing Intel® Parallel Studio XE 2015 Composer Edition for Linux and Windows.

Suscribirse a Intel® C++ Compiler