Proposed patch 4.3 for simplifying compilation errors for tbb::atomic

Dear TBB Developers,

Please consider the attached patch for tbb43_20141009oss_src which simplifies the compilation errors for misusing tbb::atomic.

Basically, tbb users sometimes misuse tbb::atomic with unsupported types, for example.
And the compiler-error message about the misuse is not easy to understand since it exposes internal details about tbb.

error: class "__m128" has no suitable assignment operator

This code 

#include <xmmintrin.h>


volatile __m128 a, b;


void test(void)


     a = b;



produces this error

$ /opt/intel/composerxe/bin/icpc -c test.cc

test.cc(7): error: class "__m128" has no suitable assignment operator

       a = b;



compilation aborted for test.cc (code 2)


when compiled with icpc.  There is no error if the variables are not volatile.  There is no error with icc or gcc or g++.

Is profiling information running on SDE accurate and trustable?


         I am trying to look at AVX 512 performance, currently, I wrote a simple function for evaluation as below, I configured the optimization and enabled AVX-512 etc in the project properties setting(vs2013 integrated with Intel-parallel-studio ), and I see the AVX-512 instructions are used from the asm files generated by compiler.

void complexVectorConjMpy(float *inputPtr1, float *inputPtr2, float *outputPtr, int numData)
    int idxData;
    float data1Re, data1Im, data2Re, data2Im;

'Segmentation violation signal raised' when xiar runs

I am trying to compile chromium 41.0.2272.64 (64 bit) using icc on Linux. icpc --version shows

icpc (ICC) 15.0.2 20150121
Copyright (C) 1985-2015 Intel Corporation.  All rights reserved.

I compile the whole thing with -ipo option on. It compiles obj/content/browser/gpu/content_browser.gpu_process_host.o as usual with the following command (some unrelated stuff is skipped, and I know that some options shown below are not supported by icc or duplicated):

Knights Corner Architecture


I have a question about the architecture of knights corner. I wonder how many Vector Processing Unit (VPU) within one physical core? Because the hardware can support 4 threads within one core, does it that mean there are four VPUs within one core? or there is only one VPU within one core, and four threads share one VPU? I am not familiar with this, Can someone answer me ?


Thank you!


Problem compiling with armadillo

Hello all,

I'm using aramadillo (http://arma.sourceforge.net) to elegantly manipulate arrays. All was working well till the last version (4.650.2).
Now it fails compiling with icpc (15.0.1, Linux) pretending there is a resolution problem.  The snippet code still compile fine with g++(4.9.2) and also clang (3.5) showing no error or warning.
I've filed a bug to armadillo team, but they told me it is an Intel issue.

configure infiniband connection(rdma) for 2 mics

Hi, I'm trying to set up infiniband connection between host and mic, mic and mic. Host is showing this on ifconfig:

mic0:ib: flags=67<UP,BROADCAST,RUNNING>  mtu 64512
        inet  netmask  broadcast
        ether 4c:79:ba:20:06:63  txqueuelen 1000  (Ethernet)

and two: mic0 and mic1 interfaces.

I have two coprocessors installed.  I can run ib_read_bw between host and mic0, but not host and mic1 or mic0 and mic1.  Getting error:  

Received 10 times ADDR_ERROR
Unable to perform rdma_client function.

Suscribirse a Optimización