 Samuel Thibault
|
7cff24cd66
move -lgfortran to STARPU_CUDA_FORTRAN_LDFLAGS
|
14 anni fa |
 Nathalie Furmento
|
e626672b57
doc: fix typo
|
14 anni fa |
|
Nathalie Furmento
|
2f2862b9c2
doc: 1st dratf of the MPI documentation
|
14 anni fa |
 Cédric Augonnet
|
f369f68805
It seems that $SIZEOF_VOID_P needs to be set manually
|
14 anni fa |
|
Nathalie Furmento
|
50b9d20c0e
configure.ac: fix FxT case
|
14 anni fa |
|
Nathalie Furmento
|
0c03f02ff7
doc: revert #3536. starpu_insert_task() returns an int
|
14 anni fa |
|
Samuel Thibault
|
6f51eb3371
move -lcufft to its own STARPU_CUFFT_LDFLAGS variable
|
14 anni fa |
|
Samuel Thibault
|
135f7897af
more items in the configuration summary
|
14 anni fa |
|
Samuel Thibault
|
be937f5151
Add configuration option summary
|
14 anni fa |
|
Nathalie Furmento
|
703469c5fb
doc: fix deftypefun declarations.
|
14 anni fa |
|
Samuel Thibault
|
ca2898b8c5
Thanks $DEITY for :g/^@deftypefun/s/[ *]\([^ *()@{}]\+\)\([,)]\)/@var{\1}\2/gc to add $arg{} where appropriate
|
14 anni fa |
|
Samuel Thibault
|
72674a38f7
use deftypefun where it's trivial to do so
|
14 anni fa |
|
Samuel Thibault
|
65391824bb
typo
|
14 anni fa |
|
Samuel Thibault
|
8a138f3417
Reference to the documentation and explain how to generate pdf/html
|
14 anni fa |
|
Samuel Thibault
|
bfcb59f8cc
document STARPU_REGRESSION_BASED vs STARPU_NL_REGRESSION_BASED
|
14 anni fa |
|
Samuel Thibault
|
091808305a
document more about calibration
|
14 anni fa |
|
Samuel Thibault
|
a118be6bfa
Document more on prefetch
|
14 anni fa |
|
Samuel Thibault
|
30dfe46a78
Document STARPU_PER_ARCH model type
|
14 anni fa |
|
Samuel Thibault
|
bbe0cdbbda
Add comments
|
14 anni fa |
|
Nathalie Furmento
|
7bbc5e5b0c
configure.ac: also look for libcuda in directory specified by --with-cuda-lib-dir
|
14 anni fa |
|
Nathalie Furmento
|
59640039cb
tests/datawizard/data_lookup.c: detect when no PU can execute the codelet and skip the test
|
14 anni fa |
|
Cédric Augonnet
|
d8e4ed1724
Only compile CUDA code if STARPU_USE_CUDA is defined
|
14 anni fa |
|
Nathalie Furmento
|
0d914cc730
tests/datawizard/lazy_allocation: properly implement the CUDA codelet to check buffer content
|
14 anni fa |
|
Nathalie Furmento
|
ce21c179fc
MPI: properly set id when FxT is enabled
|
14 anni fa |
 Ludovic Courtès
|
1e153b7a83
gcc: In task submission, pass raw pointers to `starpu_insert_task'.
|
14 anni fa |
|
Ludovic Courtès
|
007a5cf66c
gcc: Initialize the `nbuffers' field of codelets.
|
14 anni fa |
|
Ludovic Courtès
|
8177b305f4
gcc: Factorize the test stubs into <lib.h>.
|
14 anni fa |
|
Ludovic Courtès
|
91e09797ef
gcc: Pass field initializers to the codelet constructor in the right order.
|
14 anni fa |
|
Nathalie Furmento
|
fa4f1b9a81
cuda: fix type for cudaError
|
14 anni fa |
|
Samuel Thibault
|
37772c92ec
Use cl_mem for OpenCL buffer pointers, as that's what they actually are, and to avoid confusing beginners with CPU buffers.
|
14 anni fa |
