Ottimizzazione

braced-init-list in ranged for loop is not supported (C++11)?

Hi, according C++11 specification (in my case I refer to draft n3376) the following 'for' construction is valid:

#include <iostream>
int main(void) {
    for (auto x : { 0, 1, 2 }){
        std::cout << x;
    }
}

But trying to compile it with intel c++ compiler gives me error:

main.cpp(3): error : assertion failed at: "shared/cfe/edgcpfe/il.c", line 15066

I have installed the following compiler: Intel® C++ Composer XE 2013 SP1 Update 2 Integration for Microsoft* Visual Studio* 2013, Version 14.0.1290.12

Quicksync doesn't work in 64-bit on some computers (Windows N + HD 3000)

Hi. We find that QuickSync does not work in 64-bit on computer with Windows N and Intel HD 3000 graphics card. We investigated this issue and find following.

1. 32-bit and 64-bit hardware dlls present and located in same folder.

2. Quicksync works in 32-bit application but in 64-bit application it does not work.

3. MFXInit function call fails in 64-bit with MFX_ERR_UNSUPPORTED error code, I tried also automatic detection.

Analyzer tool logs for 32-bit and 64-bit attached.

Is it efficient: Copy decoded video frame into OpenGL by OpenCL kernel?

Hi

I integrated Intel QuickSync with Intel MediaSDK into a product, which uses OpenGL for video displaying.



My integration receives frame data into system memory.

While benchmarking I encountered, that decoding  into Direct3D memory is much faster.

So, I imagine, that QuickSync decodes frame into Direct3D, and then copies it (probably via DMA) into system memory.

So, in my system, each decoded frame happens double copying of frame data from GPU memory to System memory and back.

Encoding frames with large amounts of small text/lines

We're interested in configuring Intel hardware h.264 encoding in a way which is more optimized for video containing lots of tiny repetitive lines/characters, such as entire pages of small text or finely-spaced grids. I recognize this is not an ideal case for compressed video, but since x264 seems to be able to produce entirely acceptable video of this sort of content, hopefully there is a way to get Media SDK to do the same.

Advisor XE GUI support for 32 bit

I’m using Intel Parallel Studio XE 2013 on a 32bit VM machine on my laptop. But the advisor XE does not have GUI support for 32 bit.

# ls /opt/intel/advisor_xe_2013/bin32/

advixe-cl  advixe-feedback  advixe-runss  advixe-runtc  dicer_transform_helper2  libadvixe_pythonhelpers_embed_1.5.a  pin  pinbin  sitelib.zip

# ls /opt/intel/advisor_xe_2013/bin64/

advixe-cl        advixe-gui    advixe-runtc  dicer_transform_helper2             libadvixe_pythonhelpers_embed_1.5.a  pinbin

Resolving Symbols for Intel® Manycore Platform System Stack (Intel® MPSS) in Intel® VTune™ Amplifier XE Analysis

Background

Whenever Intel VTune Amplifier XE is unable to resolve symbols for libraries or the operating system, it lumps all the counts for that module together. Often, these lumped counts end up at the top of the hotspot list, skewing the analysis. By setting the correct search library path in an analysis, you can help Intel VTune Amplifier XE further resolve the counts into its constituent functions and thus rank them appropriately. 

  • Sviluppatori
  • Professori
  • Studenti
  • Server
  • Intel® VTune™ Amplifier XE
  • Architettura Intel® Many Integrated Core
  • Ottimizzazione
  • Recipe: Building and Optimizing the Hogbom Clean Benchmark for Intel® Xeon Phi™ Coprocessors

    Overview

    This article provides a recipe for compiling and running the Hogbom Clean benchmark for the Intel® Xeon Phi™ coprocessor and discusses the various optimizations applied to the code. 

  • Sviluppatori
  • Professori
  • Studenti
  • Linux*
  • Server
  • C/C++
  • Avanzato
  • Intermedio
  • Compilatore C++ Intel®
  • OpenMP*
  • Architettura Intel® Many Integrated Core
  • Ottimizzazione
  • Iscriversi a Ottimizzazione