tloloum
|
10ec5eafab
|
ELL oacc impl
|
2 years ago |
|
tloloum
|
162b2fc78f
|
beginning of oacc_ell
|
2 years ago |
|
tloloum
|
08a9744413
|
moving test
|
2 years ago |
|
tloloum
|
b5a8c549dd
|
psb_d_oacc_pde3d draft
|
2 years ago |
sfilippone
|
9e18545151
|
Fix typos
|
2 years ago |
|
sfilippone
|
686bac4224
|
Account for S/D/C/Z variants
|
2 years ago |
|
sfilippone
|
b6fe0f3344
|
New version for modules and methods
|
2 years ago |
|
sfilippone
|
7006665d82
|
Introduce submodules, adjust Makefile
|
2 years ago |
|
sfilippone
|
0707cc0a72
|
Take out reference to elldev_mod
|
2 years ago |
|
sfilippone
|
9e8294066d
|
Fix Makefile
|
2 years ago |
|
tloloum
|
7c256df451
|
Merge branch 'oacc_loloum' of https://github.com/sfilippone/psblas3 into oacc_loloum
|
2 years ago |
|
tloloum
|
2b5f09ddf9
|
all methods implementations for psb_d_oacc_csr_sparse_mat
|
2 years ago |
|
sfilippone
|
99a334d93d
|
Merge branch 'repackage' into oacc_loloum
|
2 years ago |
Salvatore Filippone
|
1911fec97b
|
Update docs
|
2 years ago |
|
sfilippone
|
e9147c089e
|
Update docs
|
2 years ago |
|
sfilippone
|
681ea2fff7
|
Updated docs
|
2 years ago |
|
tloloum
|
c7bbfb8b68
|
new Makefile, test compile well + data vect
|
2 years ago |
|
sfilippone
|
a81d1d9b68
|
Now oacc_vect compiles cleanly
|
2 years ago |
|
sfilippone
|
3bb05fab54
|
Merge branch 'oacc_loloum' of github.com:sfilippone/psblas3 into oacc_loloum
|
2 years ago |
|
sfilippone
|
1f0e591827
|
Reworked oacc_vect_mod. oacc_mlt_v_X generates ICE, to be fixed
|
2 years ago |
|
tloloum
|
6be998ac66
|
oacc_env_mod
|
2 years ago |
|
sfilippone
|
e0d7091ecc
|
Makefile in openacc/impl
|
2 years ago |
|
sfilippone
|
93c9df0277
|
Adjust Makefiles & source
|
2 years ago |
|
sfilippone
|
e3a3e39caf
|
Modify configure --enable-openacc --with-fcopenacc=.....
|
2 years ago |
|
Théophane Loloum
|
2b8671fba6
|
oacc : d_vect + i_vect + d_csr
|
2 years ago |
|
sfilippone
|
9f2b8a2623
|
Cleanup
|
2 years ago |
|
sfilippone
|
e3a55967a5
|
Modify CUDA code to compile with 12.4/12.5
|
2 years ago |
|
sfilippone
|
39cfcd3893
|
Fix allocation in coo_impl
|
2 years ago |
|
sfilippone
|
a38867be25
|
Fix allocation in coo_impl
|
2 years ago |
|
sfilippone
|
b9ad357648
|
Improve temp memory allocation in fix_coo
|
2 years ago |
|
Salvatore Filippone
|
42293c62b6
|
Fix usage of sync()
|
2 years ago |
|
Salvatore Filippone
|
a177e94ba5
|
Fix comments,
|
2 years ago |
|
sfilippone
|
d71d355b68
|
Refactor cusparse includes..
|
2 years ago |
|
sfilippone
|
2e3f862e42
|
Start refactoring cusparse.h
|
2 years ago |
|
sfilippone
|
ee66db5efd
|
Refactor interface to cusparse in preparation for CSR Adaptive
|
2 years ago |
|
sfilippone
|
12a4c21fed
|
Fixes for OpenMP compilation in map_mod
|
2 years ago |
|
sfilippone
|
e19284eb6c
|
Small omp addition
|
2 years ago |
|
sfilippone
|
10f81577f4
|
Merge branch 'repackage' of github.com:sfilippone/psblas3 into repackage
|
2 years ago |
|
sfilippone
|
35096a2ef9
|
Cosmetic changes to coo_impl
|
2 years ago |
|
sfilippone
|
add3389a81
|
Merge branch 'repackage' of github.com:sfilippone/psblas3 into repackage
|
2 years ago |
|
sfilippone
|
c8cc2275d0
|
Fix cuda/makefile for make -j
|
2 years ago |
|
sfilippone
|
70f51b9da8
|
Improve handling of fix_coo buffers with OpenMP
|
2 years ago |
|
sfilippone
|
ecccb13914
|
Fix COO fix_coo_inner_rowmajor not to overflow on integers.
|
2 years ago |
|
sfilippone
|
a613e963db
|
First step in fix for coo_impl on OpenMP
|
2 years ago |
|
sfilippone
|
d01b8145c6
|
Fix cuda makefile dependencies
|
2 years ago |
|
sfilippone
|
d8ed01218d
|
Cleanup hash_map using new indx_map%set_lc
|
2 years ago |
|
sfilippone
|
7ec394ce1c
|
Rename indx_map_mod and put SET_LR/C under ifdef
|
2 years ago |
|
sfilippone
|
7dc64692cc
|
Fix for OpenMP runs in hash_map_mod
|
2 years ago |
|
Salvatore Filippone
|
e711c53fab
|
Make sure we compile when LPK /= IPK
|
2 years ago |
|
Salvatore Filippone
|
b5a32a59f9
|
Merge branch 'repackage' of github.com:sfilippone/psblas3 into repackage
|
2 years ago |