Intel® Many Integrated Core Architecture (Intel MIC Architecture)

What does __kmp_wait_yield_4 imply in hot spot analysis of vTune


I am rewriting part of LibSVM code for vectorization in order to accelerate it on Xeon Phi. Unfortunately, the performance of my rewriting code is the same as, if not slightly worse than, the original performance on Xeon Phi. To figure out the reason, I profiled two programs using vTune hot spot analysis. According to the results, the target code segment takes significantly less time in the rewriting version, however, most of its time goes to __kmp_wait_yield_4, which only occupies a tiny amount of time in the original version.

ICC non-commercial development

I am doing research on neural networks, and plan to port this to the Xeon Phi. This work is not commercial. I noticed that Intel has a web page for getting tools for non-commercial development:

What is the state of this program, and what opportunities will I have to get ICC to compile my research to run on the Xeon Phi?

SCIF data transfer


Hi, I want to send data using RMA function of SCIF.

Register is success. But do not write to remote memory.

strerror (errno) output "No such device or address" message is displayed.

Below is a simple code of both side(host, mic)

host side---------------

Xeon Phi stops working after a while

I am running SUSE 3.0.76-0.11-default. I installed the RPM's with zypper, which seemed to load fine. The flash command did not work:

/usr/bin/micflash -update -device all -smcbootloader

there was some issue finding the smc bootloader.

So, then I set up a key, and was able to ssh mic0.

After a while say 15 minutes, I lost my connection to the Phi, the ssh session froze up, and I was not able to ssh to the card. I have a fan directly on the card, and it was warm, but not hot to the touch. When I reboot my system, I can again ssh to mic0.

Fortran Calling C Offloaded Function

Hi Guys,

  I would like to know if it is possible to do a code like this:

!dir$ offload begin target (mic:0)

call C_SUBROUTINE(x, y)

!dir$ end offload

Where the subroutine code is:

__attribute__ ((target(mic:0)))
void c_subroutine_(float * __restrict x, float * __restrict y)



Where x and y are arrays of floats. I've tried but I'm getting invalid pointer error when the subroutine is called, even including "!dir$ attributes offload:mic :: C_SUBROUTINE" inside the calling Fortran subroutine.

Overlay is not integrated into initrd (MPSS 3.4.2)


I just upgrade to the MPSS 3.4.2 and since filelists are not supported anymore for the base initrd, I tried the overlay feature with a filelist. However, I quickly realized that the files I wanted to add are not ending up in the final initrd image. So I investigated the problem and took a look at the final CPIO. What I found was the following structure:

What options to use building in Visual Studio / command line in Windows?

Sorry the Subject looks a lot like already asked question but the actual question is different.

I am trying a simple code to run on my Phi and I can get past the compilation. Is there a detailed instruction how to setup building process (through Visual Studio or command line) for different builds (eg native, MPI, OMP etc) including setting global vars etc?

I have MPSS 3.4.2. Visual Studio 2013. Windows 7. Intel Parallel Studio 2015.

I could find a lot of scripts for Linux but not for Windows.

Subscribe to Intel® Many Integrated Core Architecture (Intel MIC Architecture)