Informatique parallèle

Using IPP in a library

This question may have been asked before, but it's kinda hard to find something using the combination of IPP and Library...

I need to build a Windows library (.lib) for someone and I'm using XE2015 with IPP 8.2. I have built the library here, with 'Using Intel IPP' set to 'Single Threaded Static Library'. I also made a small program to test it, which also needs 'Using Intel IPP' set to the same value, and that works fine.

How to use IIR functions for variable filters with multiple states?


I'm currently investigating IPP 8 and have a few problems with the biquad IIR support:

- ippsIIRInitAlloc is now deprecated, fine by me, we can allocate manually. However I need to be able to changes the IIR taps WITHOUT changing the filter state. ippsIIRInit seems to destroy the delay line, which makes it unusable. So the only way to change taps without using deprecated functions is to copy the delay line and set it back with ippsIIRInit, which is unacceptable. So what am I missing?

1D Box Blur


I want to impelment fast 1D Box Blur on an image.
I saw IPP's function 'SumWindowColumns' and 'SumWindowRow':

I was wondering, do they have efficient implementation which is independent of the Kernel Size?
This could be done using Integral Images or the trick to add a pixel and remove one.

ippsIIRGetStateSize returns HUGE numbers


I'm currently checking IPP 8 and it is saying that many functions related to memory allocation are deprecated. I understand the point, so I'm checking the ways to get the memory allocations done. 


int sz;
ippsIIRGetStateSize_BiQuad_32f(1, &sz); => 4576!!!
ippsIIRGetStateSize_BiQuad_64f(1, &sz); => 8560!!!
ippsIIRGetStateSize64f_BiQuad_32f(1, &sz); => 16752!!!

C++ Operator Multi-Versioning

This is my first posting here, so I apologize if this isn't quite the right place for this question.

So I just got very interested in multiversioning in order to do some performance comparisons of instruction sets between Haswell and SandyBridge, and with functions it's a breeze, but when I tried to do the same with operators I received the following error:

internal error: assertion failed: mangled_operator_name: bad kind (shared/cfe/edgcpfe/lower_name.c, line 10192)

intrinsic for mulx



I read the white paper "New Instructions Support Large Integer Arithmetic". I read that the umul128  intrinsic compiler will provide the instruction mulx to perform the full product of two integers.

But I can't find any reference in the compiler C++ (2015.1.108) documentation.

Is the intrinsic available for mulx using the intel c++ compiler ?

Or how I can replace it with an asm instruction ? An example is welcome for the multiplication of two 64-bits integers.



Intel - Distribution - Program Fails

I'm trying to distribute some mkl files around for client use. Here's a quick synopsis of the problem.

1 - I have a Intel License Server running on a remote machine.

2 - I have some math libraries which get picked up in a ribbon for use in Excel.

3 - When I distribute the mkl libraries around and a client tries to run function which calls MKL in Excel, it crashes. Using the event logger and other logging tools have proven fruitless.

Things I've tried to resolved the issue.

Viable configuration for a home lab - 8 31S1P on 4 slot pcie 3.0 motherboard?

 Building my basement laboratory for math, machine learning, parallel programming, kagglng...

Is it electrically possible to mount 2 cards per PCIE 3.0 x16 slot with daughter boards and extenders on a motherboard.

At 300 watts per card tdp  it would likely take - 3 1000 watt power supplies BUT can we get enough power to the individual MIC's in this kind of configuration?

Being able to spend on compute cards rather than infiniband switches/cards and platforms seems like a better way to spend the allowance my wife lets me keep if possible.



Intel MPSS on Ubuntu and Mellanox OFED 2.3


We recently installed a PHI on a Dell R720 server running Ubuntu 14.04 with Mellanox OFED 2.3

From at section 2.3, there is instructions to install Mellanox OFED 2.1 to support host IB adapter. I am running Mellanox OFED 2.3.1 and can't take it down to 2.1. The question is: Is it possible to work with Mellanox OFED 2.3.1 on the PHI?

S’abonner à Informatique parallèle