 Samuel Thibault
|
2a5b38d244
heat: Ignore pivoting errors when running in valgrind
|
7 years ago |
|
Samuel Thibault
|
441644b632
Benefit from cublas v2
|
8 years ago |
|
Samuel Thibault
|
a41080d2a6
Fix setting the proper CUDA stream when using several streams from the same driver thread
|
8 years ago |
|
Samuel Thibault
|
65b1ae67ce
fix classifying complex numbers
|
9 years ago |
|
Samuel Thibault
|
4a7a6b7d7d
Enable CPU code on MIC/SCC
|
9 years ago |
|
Samuel Thibault
|
9ebf0fcd00
Remove spurious where initialization
|
9 years ago |
|
Samuel Thibault
|
a95613dc3b
Enable codelets using BLAS on MIC
|
9 years ago |
|
Samuel Thibault
|
00b1ef6f74
Add more verbose comments in examples source code
|
9 years ago |
 Nathalie Furmento
|
bb61f3808d
update cnrs copyright
|
10 years ago |
|
Nathalie Furmento
|
4f6b40a1da
src/ examples/ tests/ tools/: fix coding style, curly braces go on a newline
|
10 years ago |
|
Samuel Thibault
|
6f02123c98
assume double support is available in simgrid mode
|
10 years ago |
|
Samuel Thibault
|
41d4c8a7fe
Also enable ASYNC in simgrid mode
|
10 years ago |
|
Samuel Thibault
|
6757bca1b5
Remove implementation array terminator in tests, examples, and documentation, since it is not actually used
|
10 years ago |
|
Samuel Thibault
|
4b1a16edde
Fix University name after the fusion..
|
10 years ago |
|
Samuel Thibault
|
3287a44e0e
Allow cuda execution on simgrid too
|
10 years ago |
|
Samuel Thibault
|
7d2bf0c630
Enable asynchronous CUDA kernels wherever possible
|
11 years ago |
|
Samuel Thibault
|
b227dbcfdc
Use STARPU_ATTRIBUTE_* macros
|
12 years ago |
 Cyril Roelandt
|
281f5e37b5
Make sure CUDA cards with capabilities < 1.3 do not try to perform opertions on doubles.
|
12 years ago |
|
Cyril Roelandt
|
750f030c04
examples/lu/xlu_kernels.c: do not just abort when a cublas call fails; print an error message first to help us debug the code.
|
12 years ago |
|
Samuel Thibault
|
29b7f32f77
Provide good examples by always using streams for cublas
|
12 years ago |
|
Samuel Thibault
|
57e59bf2d9
provide good examples by always using cudaMemcpyAsync, not cudaMemcpy
|
12 years ago |
|
Cyril Roelandt
|
3e6185ecab
xlu_kernels.c: use fpclassify(3) rather than comparing floats to 0.0.
|
13 years ago |
|
Nathalie Furmento
|
3a30c58c65
examples: Update towards new codelet/task interface which defines access modes for data handles into starpu_codelet and no longer in starpu_task
|
13 years ago |
|
Nathalie Furmento
|
b9476e6d4a
update code w.r.t coding style
|
13 years ago |
|
Nathalie Furmento
|
61940dab45
examples: inside struct starpu_codelet, use xxx_funcs fields instead of the xxx_func ones
|
13 years ago |
|
Nathalie Furmento
|
b3a6c1f52b
do not use typedef for starpu_codelet, use it directly as a struct
|
13 years ago |
|
Nathalie Furmento
|
c80a8acea1
Datatypes renaming
|
13 years ago |
 Olivier Aumage
|
66e35b6222
- add support for complex number cases in the LU example
|
13 years ago |
|
Nathalie Furmento
|
a9f4293583
minor fixes for warnings detected by Intel Compiler
|
14 years ago |
|
Samuel Thibault
|
68c2135caf
fix yet more c++-like comments
|
14 years ago |
