gabrielequatrana
|
efccdc88cd
|
Fixed another typo
|
9 months ago |
|
gabrielequatrana
|
fabf53a225
|
Fixed some typos
|
9 months ago |
|
gabrielequatrana
|
4ff0f112a9
|
SpMM HDIAG working
|
9 months ago |
|
gabrielequatrana
|
bdd04a6911
|
Adding support to HDIAG SpMM
|
9 months ago |
|
gabrielequatrana
|
0490dd77db
|
ELG SpMM now working
|
9 months ago |
|
gabrielequatrana
|
dc6e5bb942
|
ELG SpMM (compiling but not working)
|
10 months ago |
|
gabrielequatrana
|
6b8199f84b
|
ELG SpMM (not compiling)
|
10 months ago |
|
gabrielequatrana
|
9daa04c3dc
|
Updated HLG SpMM (s,d,c,z)
|
10 months ago |
Salvatore Filippone
|
ccb4f73dca
|
Choose version of HLG-multivect kernel
|
10 months ago |
sfilippone
|
148a5e5e14
|
Check for shmemsize
|
10 months ago |
|
Salvatore Filippone
|
378c126055
|
Use MMBSZ=8, prepare to check shmem size
|
10 months ago |
|
Salvatore Filippone
|
897cfb4028
|
Multicolumn HLG product
|
10 months ago |
|
Salvatore Filippone
|
5b95f1920c
|
Regenerate configure, fix typo in hlg_vect_mv
|
10 months ago |
|
gabrielequatrana
|
08984619dc
|
Fixed SpMM for HLG
|
10 months ago |
|
gabrielequatrana
|
beb418e00b
|
Fixed SpMM for ELG (AXPBY GPU not working)
|
10 months ago |
|
gabrielequatrana
|
c807d88c57
|
SpMM using Cusparse dedicated routine (CSRG)
|
10 months ago |
|
Salvatore Filippone
|
399818a482
|
Never do arithmetic on a (void *)
|
10 months ago |
|
Salvatore Filippone
|
1ecf36c9c3
|
Merge branch 'psblas-bgmres' of github.com:sfilippone/psblas3 into psblas-bgmres
|
10 months ago |
|
Salvatore Filippone
|
7d74ebf5c4
|
Make multivectors work
|
10 months ago |
|
gabrielequatrana
|
82715fec9b
|
Fixed SpMM for CSRG
|
10 months ago |
|
gabrielequatrana
|
409b51e609
|
Try to fix SpMM
|
10 months ago |
|
gabrielequatrana
|
ee140bc8dd
|
Read/Write multivect fixed (SpMM bug)
|
10 months ago |
|
gabrielequatrana
|
a624b7098b
|
Cuda multivect methods implementation
|
10 months ago |
|
sfilippone
|
0760e4d553
|
Fix C function declarations for compilation with LLVM/clang in CUDA
|
10 months ago |
|
sfilippone
|
3a25d7b04a
|
Fixes for LLVM compilation
|
11 months ago |
|
sfilippone
|
e0a4d362fa
|
Define flag TRACK_CUDA_MALLOC
|
11 months ago |
|
Salvatore Filippone
|
b5f1442ac8
|
Merge branch 'nond-rep' into repackage
|
11 months ago |
|
sfilippone
|
48455190ec
|
Add GPU version of XYZW
|
11 months ago |
|
sfilippone
|
a11f328e62
|
Added CUDA version of XYZW
|
11 months ago |
|
sfilippone
|
b5d5f97661
|
Improve cuda%zero()
|
12 months ago |
|
sfilippone
|
0e269ed641
|
typo in Cabgdxyz
|
12 months ago |
|
Salvatore Filippone
|
d95077ffd6
|
Fix typo in vectordev_mod
|
12 months ago |
|
Salvatore Filippone
|
2d3773df98
|
CUDA kernels for ABGDXYZ
|
12 months ago |
|
Salvatore Filippone
|
2a75d677d0
|
ABGDXYZ in vectordev_mod
|
12 months ago |
|
sfilippone
|
2391f64df6
|
X_cuda_vect%abgdxyz
|
12 months ago |
|
sfilippone
|
93c71c4316
|
Fix %ZERO() on cuda
|
12 months ago |
|
sfilippone
|
0568a83734
|
Fix ifdef and old code
|
12 months ago |
|
Salvatore Filippone
|
35d68aa4e3
|
Reuse calls to getDeviceProperties done at init time
|
12 months ago |
|
Salvatore Filippone
|
1ba8dfc7b7
|
Switch FOR and IF in AXPBY
|
12 months ago |
|
Salvatore Filippone
|
f9677bc892
|
Enabled new CUDA version of ABGDXYZ
|
12 months ago |
|
Salvatore Filippone
|
4681767ef8
|
New implementation for ABGDXYZ in CUDA
|
12 months ago |
|
Salvatore Filippone
|
105aa3c570
|
Intermediate impl of ABGDXYZ
|
12 months ago |
|
Salvatore Filippone
|
864872ecac
|
Intermediate implementation of abgdxyz on cuda
|
12 months ago |
|
Salvatore Filippone
|
a41b209144
|
Better AXPBY implementation in CUDA.
|
12 months ago |
|
Salvatore Filippone
|
ebc7c6b3b4
|
Fix call to base%abgdxyz
|
1 year ago |
|
Salvatore Filippone
|
14c4ff0f32
|
Added new methd for two combined axpbys
|
1 year ago |
|
sfilippone
|
6433dc797e
|
Fix CUDA implementation of %set_scal and %zero
|
1 year ago |
|
sfilippone
|
097d63147a
|
Fix cuda dir makefile
|
1 year ago |
|
Salvatore Filippone
|
20a01d4d71
|
Attempt at fixing CSRG in CUDA 10.2. Not complete yet.
|
1 year ago |
|
sfilippone
|
1bc2a884e2
|
Adjust conditional compilation on CUDA version
|
1 year ago |