Struggling to create dynamic library (interop C# - C++ -> MIC)
Hello everyone,I'm working on a project C# and I want to aim a MIC within a computation kernel. Here's the scheme of my project:C# main project--> sending data to C++/CLIC++/CLI project (bridge...
View ArticleStruggling to create dynamic library (interop C# - C++ -> MIC)
Hello everyone,I'm working on a project C# and I want to aim a MIC within a computation kernel. Here's the scheme of my project:C# main project--> sending data to C++/CLIC++/CLI project (bridge...
View ArticlePerformance degradation if loop count is not known at compile time
I am creating a simple matrix multiplication procedure, operating on the Intel Xeon Phi architecture.After many attempts with autovectorization, trying to get better performances, I had to use Intel...
View ArticleOpenCL rpm issues
Hi all,while investigating https://software.intel.com/en-us/forums/topic/534808 I decided to install the OpenCL drivers for the Xeon Phi; I ran into two issues while installing the package...
View ArticleAgner's library for KNC
Hello,I've seen before that some of you have been mentioning Agner Fog'sĀ VCL library (http://www.agner.org/optimize/#vectorclass) as a very good alternative for dealing with ICC intrinsics directly. I...
View Articleoffload-attribute-target
When I compile my program with just 'ifort', everything compiles fine. But when IĀ do 'ifort -qoffload-attribute-target=mic' it gives me errors, why would that be? For example, one of the errors is...
View ArticleClik here to view.
Take care on short runs of Intel(r) VTune-TM Amplifier XE to collect Intel(r)...
It was an easy mistake to fall into.Ā The customer had collected data on an application taking less than a second to run on an IntelĀ® Architecture host processor, and then tried the same workload on...
View Articlemkl_?csrmv() for multiplying transpose of a matrix
Ā p, li { white-space: pre-wrap; }When mkl_?csrmv() is used with the option transa='N', the performance of mkl_?csrmv on Intel Xeon Phi coprocessor is good. However, when it is used with the option...
View ArticleClik here to view.
Poor FFT mkl performance
I am optimizing a new application (written with Xeon Phi in mind) which performs a lot of FFT transforms.The transforms are done on 512x512 arrays separately in each thread. This works quite well on...
View ArticleXeon Phi and Gentoo
Hifor my master thesis I am supposed to write a kernel module which uses functions of another module on the Xeon Phi.The testmachine I have is running RHEL 6.4.As for the Kernel development I need and...
View ArticleClik here to view.
Regarding GROMACS installation on using MIC
Ā Hi guys,i am installing GROMACS using MIC and got the following error while using cmake to compile. I have tried with icc and ifort (even with mpicc and mpiifort), but still the same problem occurred....
View Articlemp_linpack HPL in auto offload
Hello.I'm trying to run precompiled intel AO benchmark. For some reason, when running it the MIC never goes beyond 50% utilization, as if HT was disabled on it.By default all cores should be used. Am I...
View ArticleRHEL 6.6 support
Will there be an update to mpss to support RHEL 6.6 (and derivatives) ?FWIW: CentOS 6.6 / 2.6.32.504 does work with MPSS 3.3.1 (I did not test MPSS 3.4.X). The 'weak-updates' directory links in the...
View ArticleGCC 4.7 bug with C++11 atomics
We use C++11 atomics for the implementation of communication queues. With the gcc 4.7 from MPSS 3.2.1 the compiler generates following error:{standard input}: Assembler messages: {standard input}:5116:...
View ArticleAny chance of ganglia monitoring being ported to more recent versions?
The one shipping with MPSS is for 3.4.1. The current Ganglia release, as of 3rd August, is 3.6.2 and we're on 3.5.7. We need to add the functionality to our existing Ganglia installation, rather than...
View ArticleClik here to view.
Poor FFT mkl performance (2)
(This is a new issue after the issue with mkl_dft_grasp_user_thread() was solved in previous post by using separate fft plans in each thread.)I am optimizing a new application (written with Xeon Phi in...
View ArticleHDF5 error
Hi,I have some compiler assisted offload code that when I use a wrapper for the HDF library (v5 1.8.8) breaks with the following error:On the remote process, dlopen() failed. The error message sent...
View ArticleClik here to view.
Intel Xeon Phi 5110P - device is not online
Dear friends, can you help us with the following problem?We have Intel Xeon Phi 5110 P installed on Asus p8z77ws motherboard.OS - CentOS 6 with necessary kernel version.We have installed mpss 3.3.2 and...
View ArticleMPSS-3.4.1 installs smoothly on Fedora 20 -with release kernel-3.11.10
We have installed two C0 stepping 3120A cards on a Dual-Xeon Serverboard.While MPSS-3.4.1 installation failed for the F20 update kernel (we tried kernel-headers-3.16.7-200.fc20.x86_64), everything...
View ArticleXeon Phi on Vanilla Kernel (3.13 and above)
HiI am currently programming for my master thesis in computer science.My next step is to make the MIC runable on the vanilla kernel with MIC Modules in version 3.13 and above.The Kernel documentations...
View Article