porting work to Maxeler DFE

Nathalie Furmento e07abf83c9 merge branches/sched_ctx in trunk 12 år sedan
build-aux 039b226a94 build: Add a rule to generate `cyclomatic-complexity.html'. 12 år sedan
doc 08f7b9607c new merge 12 år sedan
examples 92f989a021 examples/sched_ctx/sched_ctx.c: add warning 12 år sedan
gcc-plugin f91d480518 Remove heft. 12 år sedan
include e22791b9f3 merge fixes 12 år sedan
m4 07f9303473 Another attempt at fixing the plugin build. 12 år sedan
mpi c6e8c55fe1 insert_task: do not call _starpu_codelet_pack_args() if not needed 12 år sedan
sched_ctx_hypervisor 34aa059de1 clean printfs 12 år sedan
socl d679f897c2 SOCL: do not make StarPU crash when a kernel cannot be executed 12 år sedan
src 0b207f269e src/core/jobs.h: merge fix 12 år sedan
starpu-top a4e05df11e starpu-top: fix display of package version information (thanks to Ludovic Courtès for the patch) 12 år sedan
starpufft b4d43505aa starpufft/Makefile.am: use our specific loader to run tests 12 år sedan
tests b9f1897f10 new merge after the merge of heft+dmda 12 år sedan
tools cab880bfd5 new merge with trunk, heft & dmda important changes 12 år sedan
.dir-locals.el 288af06552 remove un-needed copyright 13 år sedan
.gitignore f206bc52fe update list of files to ignore in git 12 år sedan
AUTHORS baed2167c5 AUTHORS: sort names alphabetically 12 år sedan
COPYING.LGPL fc22dad676 create a trunk/, branches/ and a tags/ directory 16 år sedan
ChangeLog aee5120751 dmda/heft merge: update the Changelog and the Copyrights. 12 år sedan
INSTALL 6a40e86cac Fix build windows instructions 13 år sedan
Makefile.am b74fc182bb allow submitting to several ctxs from 1 th + small example 12 år sedan
README 03ae1a5728 README: update requirements for StarPU-Top 12 år sedan
README.dev 9db5ddc0c2 README.dev: add a section about error handling. 13 år sedan
STARPU-VERSION a5788cb8d7 STARPU-VERSION: add rules which explain how to change numbers 13 år sedan
TODO ca5b979c06 tests: tests old functionalities are still working 13 år sedan
autogen.sh d49977bfb2 On Mac OS X, we should look for glibtool, not for libtool 14 år sedan
configure.ac e22791b9f3 merge fixes 12 år sedan
libstarpu.pc.in 927b47d57d new merge 12 år sedan
starpu-1.0.pc.in 74be3d8c8c starpu-1.0.pc.in: fix syntax 12 år sedan

README

# StarPU --- Runtime system for heterogeneous multicore architectures.
#
# Copyright (C) 2009-2012 Université de Bordeaux 1
# Copyright (C) 2010, 2011 Centre National de la Recherche Scientifique
#
# 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.

++=================++
|| I. Introduction ||
++=================++

+---------------------
| I.a. What is StarPU?

StarPU is a runtime system that offers support for heterogeneous multicore
machines. While many efforts are devoted to design efficient computation kernels
for those architectures (e.g. to implement BLAS kernels on GPUs or on Cell's
SPUs), StarPU not only takes care of offloading such kernels (and implementing
data coherency across the machine), but it also makes sure the kernels are
executed as efficiently as possible.

+------------------------
| I.b. What StarPU is not

StarPU is not a new language, and it does not extend existing languages either.
StarPU does not help to write computation kernels.

+---------------------------------
| I.c. (How) Could StarPU help me?

While StarPU will not make it easier to write computation kernels, it does
simplify their actual offloading as StarPU handle most low level aspects
transparently.

Obviously, it is crucial to have efficient kernels, but it must be noted that
the way those kernels are mapped and scheduled onto the computational resources
also affect the overall performance to a great extent.

StarPU is especially helpful when considering multiple heterogeneous processing
resources: statically mapping and synchronizing tasks in such a heterogeneous
environment is already very difficult, making it in a portable way is virtually
impossible. On the other hand, the scheduling capabilities of StarPU makes it
possible to easily exploit all processors at the same time while taking
advantage of their specificities in a portable fashion.

++==================++
|| II. Requirements ||
++==================++

* make
* gcc (version >= 4.1)
* if CUDA support is enabled
* CUDA (version >= 2.2)
* CUBLAS (version >= 2.2)
* if OpenCL support is enabled
* AMD SDK >= 2.3 if AMD driver is used
* CUDA >= 3.2 if NVIDIA driver is used
* extra requirements for the svn version (we usually use the Debian testing
versions)
* autoconf (version >= 2.60)
* automake
* makeinfo
* libtool (version >= 2)

* Remark: It is strongly recommanded that you also install the hwloc library
before installing StarPU. This permits StarPU to actually map the processing
units according to the machine topology. For more details on hwloc, see
http://www.open-mpi.org/projects/hwloc/ .

* To build the StarPU-Top tool the following packages (along with
their development files) are also required:
* libqt4-dev >= 4.7
* libqt4-network
* libqt4-opengl
* libqt4-sql
* qt4-qmake

++=====================++
|| III. Getting StarPU ||
++=====================++

StarPU is available on https://gforge.inria.fr/projects/starpu/.

The project's SVN repository can be checked out through anonymous
access with the following command(s).

$ svn checkout svn://scm.gforge.inria.fr/svn/starpu/trunk
$ svn checkout --username anonsvn https://scm.gforge.inria.fr/svn/starpu/trunk

The password is 'anonsvn'

++=============================++
|| IV. Building and Installing ||
++=============================++

+---------------------------
| IV.a. For svn version only

$ ./autogen.sh

+-----------------------
| IV.b. For all versions

$ ./configure
$ make
$ make install

+---------------------
| IV.c. Windows build:

StarPU can be built using MinGW or Cygwin. To avoid the cygwin dependency,
we provide MinGW-built binaries. The build process produces libstarpu.dll,
libstarpu.def, and libstarpu.lib, which should be enough to use it from e.g.
Microsoft Visual Studio.

Update the video drivers to the latest stable release available for your
hardware. Old ATI drivers (< 2.3) contain bugs that cause OpenCL support in
StarPU to hang or exhibit incorrect behaviour.

For details on the Windows build process, see the INSTALL file.

++==================++
|| V. Documentation ||
++==================++

Texinfo documentation is available in doc/ . If LaTeX is available on the
machine, a pdf can be generated by running

$ make -C doc pdf

If makeinfo is available on the machine, html pages can be generated by running

$ make -C doc html

++============++
|| VI. Trying ||
++============++

Some examples ready to run are installed into $prefix/lib/starpu/{examples,mpi}

++==============++
|| VII. Upgrade ||
++==============++

To upgrade your source code from older version (there were quite a few
renamings), use the tools/dev/rename.sh script

++===============++
|| VIII. Contact ||
++===============++

For any questions regarding StarPU, please contact the starpu-devel
mailing-list at starpu-devel@lists.gforge.inria.fr .