This code works with gcc and clang but doesn't compile with icc 2016 with std=c++14
I would like to see the Intel Intrinsics Guide include all of the intrinsics, not just the SSE-based intrinsics.
It would be helpful if the 'interlocked' and other intrinsics were also included, especially with timing information.
I am using AVX/SSE instructions to replace memcmp and our workload includes comparing 64 bytes and occasionally 64 and 128 bytes. I am using following function cmp32 for 32byte comparisons and extend it 2 times for 64 or 4 times for 128 bytes and I am hardly getting 1% performance improvement. Testing was done on Intel(R) Core(TM) i7-4790 CPU @ 3.60GHz, Ubuntu 14.04 x86_64.
I tried replacing following lines
vcmp = _mm256_cmpeq_epi64(xmm0, xmm1);
vmask = _mm256_movemask_epi8(vcmp);
I note that premier.intel.com is open for submissions for the first time in months, so I submitted a ticket. It still looks like setting up premier for beta 16 has not been completed, as the form requires choosing compiler version 15.0 or earlier from the pull-down.
I'm curious whether Cilk(tm) Plus is under deprecation, in view of the comments at IDF last year that Intel would not sponsor publications on it, and the lack of follow-through to make gcc -fcilkplus viable.
As to changes in behavior in the beta test:
[root@amax ~]# micctrl -s
mic0: booting (mode: linux image: /lib/firmware/mic/uos.img)
mic1: booting (mode: linux image: /lib/firmware/mic/uos.img)
I have two mic cards, but they just get stuck at booting state, and never be online. No matter I reboot the host or restart the mpss service, it just stayed at this state. Any help please? Thank you very much!
I am trying to do General Exploration analysis, and the following error appears almost immediately after trying to attach to the process,
"Communication with sampling driver failed".
I'm running on Windows 7, on a Intel Xeon CPU E5-2680 (2 processors). Doing the Basic Hotspot Analysis went fine. Any ideas - or things I could try?
I have Xeon Phi 3120A and I wonder if I have to buy "Intel Parallel Studio XE 2015" or any other software package to make it work.
I'm in process of converting from Microsoft Visual Studio 2008 to MSV2012 for my mixed code, C and FORTRAN application.
The code compiles and executes perfectly for MSV2008 but I receive many, many error messages for the C portion of my code
when I try to compile using MSV2012 wrapper.
The compiler does not like the line:
int _stdcall OPEN_MATLAB(name,ll)
The first couple of error messages are :
1>Source\matlab.c(38): error C2065: 'name' : undeclared identifier
1>Source\matlab.c(38): error C2065: 'll' : undeclared identifier
I'm looking at converting from a different library to Intel ones so I'm looking at function analogues.
There are vector set functsions( ippsSet_???), but I was wondering if there was an analogue for matrices, preferable for complex data.
I'm trying to get a very large Visual Studio 2013 solution to compile (100+ big projects). All of the projects are being compiled with the /fast flag. After waiting around 20-30 minutes for the solution to compile, the linker chokes with:
xilink: : error #10014: problem during multi-file optimization compilation (code 1)
I can see the RAM usage in Task Manager, increasing very quickly when linker starts running. It runs up to +2gb, and then drops back to where it was before. This is on a system running Windows 7 Pro 64 bit with 16gb of RAM and an Intel 3770.