Archive for the ‘Programming’ Category
Know more about Flops
If you want to know more about flops (on CPU or on GPU) a good first (but good step) is to use this link:
http://folding.stanford.edu/English/FAQ-flops
They give lots of details and are very clear, in bref, a good reference.
[CUDA] Compiling with cmake
Here is an simple cmake file to compile a cuda project.
Click to continue…
[C++][Qt] Parallel Quick Sort with QtConcurrent (Shared memory generic quick sort)
PS : I developed several quick sort (available on this blog), a sequential version, an openmp tasks version, a openmp not inplace version, an mpi version and a Qt concurent version.
Here I developed a quick sort based on the great qt feature QtConcurrent.
It is mostly similar to the openmp tasks versions.
Click to continue…
[Qt][GPU] C++ Application for dynamic transition images (QtConcurrent, QtOpenCL)
In this post, I will briefly introduce my first test of the qtopencl tool.
[GPU] Install Cuda Opencl on Ubuntu
The steps to install cuda, opencl for nvidia gpu on ubuntu.
Click to continue…
[C++][Omp] OpenMP version _OPENMP directive
Sometime you may want to know what version of openmp you are using at compile time.
This is possible using the _OPENMP directive.
[Qt][OpenMP][MPI] mpi/openmp qt creator .pro flag
Add the right options to your .pro to compile with openmp or mpi under Qt creator.
[C++][OpenMP] Custom barrier, a barrier for a group of threads
Openmp give a barrier for all threads.
Here is a class to perform a barrier with only a group of threads.
[C++][OpenMP] A shared memory quick sort openmp tasks (example, source code)
After a first shoot of quick sort on shared memory (to be able to create an mpi version, but that is not inplace)
I created a real shared memory version.
Be aware that these versions need the OpenMP task!
PS : I developed several quick sort (available on this blog), a sequential version, an openmp tasks version, a openmp not inplace version, an mpi version and a Qt concurent version.
