Olivier Aumage a110078dab fix case where the number of subcontexts does not divide the number of cpu workers, in which case some cpu workers were not assigned to the subcontexts, and would end-up executing parallel tasks without the prologue being executed first %!s(int64=8) %!d(string=hai) anos
..
audio 441644b632 Benefit from cublas v2 %!s(int64=8) %!d(string=hai) anos
axpy 441644b632 Benefit from cublas v2 %!s(int64=8) %!d(string=hai) anos
basic_examples 91a65f66ef this example doesn't support Master-Slave transfers %!s(int64=8) %!d(string=hai) anos
binary 087fb6729c Fix number of MPI Slaves workers when testing OpenCL %!s(int64=8) %!d(string=hai) anos
callback 4b0182bdb0 examples: fixes for cppcheck warnings %!s(int64=9) %!d(string=hai) anos
cg 5e9e6b367f Use cublasdot and cudaMemsetAsync instead of self-baked implementations %!s(int64=8) %!d(string=hai) anos
cholesky f26c50ae1d fix warning %!s(int64=8) %!d(string=hai) anos
common 8155434aa6 fix warning in simgrid mode %!s(int64=10) %!d(string=hai) anos
cpp 80cf162dc9 MIC/SCC/MPI-Master-Slave devices cannot read properly function's names in C++ env %!s(int64=8) %!d(string=hai) anos
filters 4b0182bdb0 examples: fixes for cppcheck warnings %!s(int64=9) %!d(string=hai) anos
fortran 00b1ef6f74 Add more verbose comments in examples source code %!s(int64=10) %!d(string=hai) anos
fortran90 eaebdeebf1 update dm* policies with new synchro scheme %!s(int64=8) %!d(string=hai) anos
gl_interop 4f6b40a1da src/ examples/ tests/ tools/: fix coding style, curly braces go on a newline %!s(int64=10) %!d(string=hai) anos
heat e9a865605f disable shape 2, it seems not well conditioned %!s(int64=8) %!d(string=hai) anos
incrementer a8784b8805 fix build %!s(int64=9) %!d(string=hai) anos
interface 4b0182bdb0 examples: fixes for cppcheck warnings %!s(int64=9) %!d(string=hai) anos
lu fc0049658d Add missing unregistration %!s(int64=8) %!d(string=hai) anos
mandelbrot fa947f5c97 update peager with new synchro scheme %!s(int64=8) %!d(string=hai) anos
matvecmult 4b0182bdb0 examples: fixes for cppcheck warnings %!s(int64=9) %!d(string=hai) anos
mlr 02352696b7 mlr: changing example to also use data handles %!s(int64=8) %!d(string=hai) anos
mult 441644b632 Benefit from cublas v2 %!s(int64=8) %!d(string=hai) anos
native_fortran eac4759964 fix memleak %!s(int64=8) %!d(string=hai) anos
openmp 3953060972 update pheft with new synchro scheme %!s(int64=8) %!d(string=hai) anos
pi 904414d6f0 Fix used type %!s(int64=8) %!d(string=hai) anos
pipeline 9335a5c0b5 fix coverity warning %!s(int64=8) %!d(string=hai) anos
ppm_downscaler debcbc561e Fix terminating string %!s(int64=8) %!d(string=hai) anos
profiling f942ab9adc examples/profiling: display the number of executed tasks per worker %!s(int64=9) %!d(string=hai) anos
reductions 441644b632 Benefit from cublas v2 %!s(int64=8) %!d(string=hai) anos
sched_ctx a110078dab fix case where the number of subcontexts does not divide the number of cpu workers, in which case some cpu workers were not assigned to the subcontexts, and would end-up executing parallel tasks without the prologue being executed first %!s(int64=8) %!d(string=hai) anos
sched_ctx_utils 4b0182bdb0 examples: fixes for cppcheck warnings %!s(int64=9) %!d(string=hai) anos
scheduler 1f52be43ed reenable test %!s(int64=8) %!d(string=hai) anos
spmd 3953060972 update pheft with new synchro scheme %!s(int64=8) %!d(string=hai) anos
spmv 9c38741e02 fix code to follow coding conventions for braces %!s(int64=8) %!d(string=hai) anos
stencil 038343c2df examples/stencil/stencil-blocks.c: add missing variable %!s(int64=8) %!d(string=hai) anos
tag_example 4b0182bdb0 examples: fixes for cppcheck warnings %!s(int64=9) %!d(string=hai) anos
top d476c09ccf fix bugs detected by coverity %!s(int64=9) %!d(string=hai) anos
worker_collections c108a816e7 examples: fix cppcheck warnings %!s(int64=8) %!d(string=hai) anos
.gitignore 27eaf1ca36 Add `.gitignore'. %!s(int64=14) %!d(string=hai) anos
Makefile.am c79052d75a multiple calls of MPI app is not allowed in shell script %!s(int64=8) %!d(string=hai) anos
README.txt 5cbef2b984 fix copyright %!s(int64=9) %!d(string=hai) anos
loader-cross.sh.in 999afbb767 Fix cross-loader for the examples/stencil directory %!s(int64=10) %!d(string=hai) anos

README.txt

# StarPU --- Runtime system for heterogeneous multicore architectures.
#
# Copyright (C) 2016 Université de Bordeaux
# Copyright (C) 2016 CNRS
#
# StarPU is free software; you can redistribute it and/or modify
# it under the terms of the GNU Lesser General Public License as published by
# the Free Software Foundation; either version 2.1 of the License, or (at
# your option) any later version.
#
# StarPU is distributed in the hope that it will be useful, but
# WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
#
# See the GNU Lesser General Public License in COPYING.LGPL for more details.

audio
This applies a simple band filter over audio files

axpy
This computes the AXPY BLAS over a big vector

basic_examples
This contains very trivial examples: hello world, scaling a vector, etc.

binary
This shows how to store and load compiled OpenCL kernels on and from the
file system

callback
This shows how to use task callbacks

cg
This computes a Conjugate Gradient

cholesky
This computes a Cholesky factorization

common
This holds common code for BLAS kernels

cpp
This shows how to use StarPU from C++

filters
This contains several partitioning examples

fortran90
This shows how to use StarPU from Fortran90

gl_interop
This shows how interoperation can be done between StarPU CUDA
computations and OpenGL rendering

heat
This uses a finite element method to compute heat propagation thanks to
an LU factorization or a conjugate gradient

incrementer
This just increments a variable

interface
This shows how to implement a user-defined data type, here simply
complex floats

lu
This computes an LU factorization

mandelbrot
This computes and outputs the mandelbrot set

matvecmult
This computes a matrix-vector multiplication

mult
This computes a matrix-matrix multiplication

openmp
This shows how to use an OpenMP code inside a StarPU parallel task

pi
This computes Pi thanks to random numbers

pipeline
This shows how to submit a pipeline to StarPU with limited buffer
use, and avoiding submitted all the tasks at once

ppm_downscaler
This downscales PPM pictures

profiling
This examplifies how to get profiling information on executed tasks

reductions
This examplifies how to use value reductions

sched_ctx
This examplifies how to use scheduling contexts

sched_ctx_utils
This is just common code for scheduling contexts

scheduler
This examplifies how to implement a user-defined scheduler

spmd
This shows how to define a parallel task

spmv
This computes a sparse matrix-vector multiplication

stencil
This computes a dumb 3D stencil with 1D subdomain decomposition

tag_example
This examplifies how to use tags for dependencies

top
This examplifies how to enrich StarPU-top with information

worker_collections
This examplifies how to use worker collections