Intel® Math Kernel Library 11.2 Update 4 is now available

Gennady_F_Intel — Sun, 08 Feb 2015 19:36:24 GMT

Intel® Math Kernel Library (Intel® MKL) is a highly optimized, extensively threaded, and thread-safe library of mathematical functions for engineering, scientific, and financial applications that require maximum performance. Intel MKL 11.2 Update 4 packages are now ready for download. Intel MKL is available as part of the Intel® Parallel Studio XE 2015 . Please visit the Intel® Software Evaluation Center to evaluate this product.

Intel® MKL 11.2 Update 4 Bug fixes

What's New in Intel® MKL 11.2 Update 4 : Release Notes

Hi Gennady,

Wei_Z_Intel — Mon, 09 Feb 2015 03:41:49 GMT

Hi Gennady,

Thank you for the MKL 11.2 update, I am using it, but meet an issue, I tried many suggestions from posts, but none of them worked, so ask for help here.

Below is the issue I meet. I'm trying to use MKL Vector Mathematical Functions like vcMulByConj etc as below as an example(did not do input buffer initialization here， just an example).

void mklFuncTest()
{

MKL_Complex8 inputPtr1[100];
MKL_Complex8 inputPtr2[100];
MKL_Complex8 outputPtr[100];
MKL_INT numData = 100;
//call mkl function
vcMulByConj(numData, inputPtr1, inputPtr2, outputPtr);

return;
}

When I run the code, for win32 platform, it builds and runs both fine, but for x64 platform, when it goes to vcMulByConj, it fly, by going to dissembly, I see it fly when running below code

000000013FB74A6F call mkl_vml_serv_load_vml_dll (013FB761F0h)

It looks to me that it could not run well with lib/dll 'mkl_vml_serv_load_vml_dll' on x64 platform. Building is ok, looks that lib/dll is there.

And I see, some of other MKL functions can run well like FFT(DftiComputeForward) etc on x64 application.

I am using below version of tools:

parallel_studio_xe_2015

w_mkl_11.2.1.148

w_ipp_8.2.1.148

Visual Studio professional version 2013

Per my understanding, I used automatical link to MKL via property->Intel Performance Library->use intel MKL->Parallel, in property->Linker->Advanced->Target Machine, I set to MachineX64 (/MACHINE:X64). For Linker->Input->Additional Dependance, I keep it unchanged.

My PC info is as below:

Processor: Intel core i5-3320M CPU

System Type: 64-bit Operation system

John

topic Hi Gennady, in Intel® oneAPI Math Kernel Library

Intel® Math Kernel Library 11.2 Update 4 is now available

Hi Gennady,