psocratis
ff1c2deb4f
fixing device runs for slu, mumps and umfpack
2020-11-17 19:16:02 -08:00
Veselin Dobrev
78a4889fd0
Move some Host{Read,Write,ReadWrite} calls just before the
...
calls that need the data on host. This should make it clearer
why they are needed and should help with future updates that
move operations to device.
In HypreSolver and HyprePCG, moved the output vector initialization
in non-iterative mode before the setup calls. This makes sure we
do not give uninitialized data to the setup calls.
Added some more host transfers in the sparse matrix-sparse matrix
multiplication function.
2020-11-17 12:40:51 -08:00
camierjs
f82d078562
Simplify
2020-11-13 16:17:22 -08:00
camierjs
dab294dabf
Initial run
2020-11-13 15:25:36 -08:00
Tzanio
d472fe16d9
Changes in the linalg/ directory before the mfem-4.2 release
2020-10-24 19:06:53 -07:00
Tzanio Kolev
00550ef4f7
Merge pull request #1489 from mfem/feature/artv3/cusparse-Spmv
...
SpMV with CuSPARSE
2020-08-16 15:16:10 -07:00
Tzanio
a16b216c91
Final editorial changes
2020-08-16 15:14:03 -07:00
stefanhenneking
0db7b843f5
Using std::abs
2020-07-29 15:23:11 -05:00
Arturo Vargas
4e8a531bb1
add guards for cpu with cuda codes
2020-07-21 16:23:43 -07:00
Arturo Vargas
9937009eab
clean up pass, guard for resizing
2020-07-21 11:59:42 -07:00
Arturo Vargas
9bd06e360e
temp_buffer->new_buffer
2020-07-21 11:13:33 -07:00
Arturo Vargas
264886c511
PR comments
2020-07-21 11:11:38 -07:00
Arturo Vargas
2a85ec5f97
Merge branch 'master' into feature/artv3/cusparse-Spmv
2020-07-18 22:05:07 -07:00
Stefan Henneking
79aa92e217
Merge branch 'master' into abs-mult-dev
2020-07-15 10:51:00 -05:00
Tzanio Kolev
ae880b4ee8
Merge branch 'master' into yohann/fa
2020-07-14 16:42:03 -07:00
stefanhenneking
6fc40b1ee5
Adding AbsMult and AbsMultTranspose to SparseMatrix.
2020-07-13 14:55:57 -05:00
Dylan Copeland
c1320238ae
Minor changes to MFEM_ABORT_KERNEL.
2020-07-02 10:51:12 -07:00
Dylan Copeland
237905c956
Addressing Veselin's comments about MFEM_ABORT_KERNEL.
2020-06-30 11:15:14 -07:00
Arturo Vargas
1d35d74e85
fix logic for using cusparse
2020-06-25 11:40:28 -07:00
Arturo Vargas
f761e4d033
add runtime option to use cusparse - on by default
2020-06-25 10:41:19 -07:00
Arturo Vargas
684785eb64
clean up pass
2020-06-23 15:55:46 -07:00
Arturo Vargas
bddf1110b4
PR review updates
2020-06-23 15:48:25 -07:00
Arturo Vargas
6c1ee0c854
skip computation if matrix is zero
2020-06-11 20:33:43 -07:00
Dylan Copeland
68fd30d2f9
Adding MFEM_ABORT_KERNEL, currently used only in SparseMatrix::DiagScale.
2020-06-08 22:42:48 -07:00
Dylan Copeland
e4a6e1c724
Added MFEM_FORALL to SparseMatrix::DiagScale. Made some minor style improvements.
2020-06-08 21:40:11 -07:00
Yohann Dudouit
c2de4ee220
Move synchronization inside Print().
2020-06-08 18:37:11 -07:00
Yohann Dudouit
e831e8007b
Use OperatorJacobiSmoother in ex9.
...
- implement GetDiag() for SparseMatrix on device.
2020-06-08 18:10:13 -07:00
Vargas
9b83346ed3
make style
2020-06-01 10:42:45 -07:00
Arturo Vargas
26d3646c1b
add guards for non cuda
2020-06-01 10:42:04 -07:00
Vargas
f7724b30d9
make style
2020-06-01 10:15:09 -07:00
Arturo Vargas
c28cfb92ac
clean up pass, add diffusion benchmark
2020-06-01 10:13:55 -07:00
Arturo Vargas
c61af0cce9
comment out debugging code
2020-05-29 10:28:24 -07:00
Dylan Copeland
f5aa751bba
Fixed SparseMatrix::DiagScale to work on GPU.
2020-05-28 16:46:41 -07:00
Yohann Dudouit
8ecb802662
More code cleaning.
2020-05-22 19:01:00 -07:00
Yohann Dudouit
15f6269ad4
Add some debugging prints.
2020-05-20 17:30:42 -07:00
Arturo Vargas
8b9b0f7a0d
uncomment inportant code
2020-05-18 17:58:43 -07:00
Arturo Vargas
fd0ac87506
added driver for testing performance
2020-05-18 17:46:39 -07:00
Arturo Vargas
83d753c036
proof of concept
2020-05-18 15:53:26 -07:00
Arturo Vargas
fb249c5775
fixed configuration for sparse matvec
2020-05-18 14:32:32 -07:00
Arturo Vargas
a3e73ee1a3
init commit of cuSparse Spmv
2020-05-18 13:32:15 -07:00
Tzanio
05125e19d5
Removed a space
2020-03-09 15:36:41 -07:00
Tzanio
58872c9e7b
Merge branch 'master' into okina-mmu
...
Conflicts:
fem/datacollection.cpp
tests/unit/CMakeLists.txt
2020-03-04 15:19:33 -08:00
Tzanio
ca55ef0217
The release number for mfem-4.1 is LLNL-CODE-806117
2020-03-02 17:18:46 -08:00
Tzanio
88a78bb6aa
Merge branch 'master' into bsd-relicense
...
Conflicts:
CONTRIBUTING.md
2020-03-02 09:28:37 -08:00
camierjs
e4d0496931
Merge branch 'master' into okina-mmu
2020-02-29 10:20:37 -08:00
Tzanio
2caa222edf
A number of changes to switch the code from LGPL-2.1 to BSD-3:
...
- Updated LICENSE, README, CHANGELOG and CONTRIBUTING.md
- Removed COPYRIGHT, added NOTICE
- Updated the headers in all source files, makefiles, etc.
2020-02-23 11:16:46 -08:00
camierjs
519a0cbb10
Revert to keep the verify and device disable
2020-02-18 14:32:06 -08:00
camierjs
39d6614642
Revert sample-runs and examples
...
HIP managed addons
Sparsmat transpose on host
2020-02-18 11:52:00 -08:00
camierjs
4eecaf7625
Revert transpose kernel to stay on host for the atomics
2020-02-15 08:37:13 -08:00
Veselin Dobrev
9499d49e9f
Variour tweaks and simplifications, coordinated with @camierjs.
2020-02-14 16:25:40 -08:00