CFD/hypre
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
12,191 Commits 1 Branch 84 Tags 207 MiB
2e29e665bb
Commit Graph

2212 Commits

Author SHA1 Message Date
Ruipeng Li
2fa29169c6 bug fix 2022-06-29 10:06:09 -07:00
Ruipeng Li
172787d7d9 delete files 2022-06-28 22:18:39 -07:00
Ruipeng Li
b03f350bf1 fix after merge 2022-06-28 22:15:45 -07:00
Ruipeng Li
4ed68414e5 Merge branch 'master' of github.com:hypre-space/hypre into interp_trunc 2022-06-28 22:10:57 -07:00
Ruipeng Li
b3573fb7a5 .saved 2022-06-28 21:09:23 -07:00
Wayne Mitchell
4411530e76
hypre_Item (#645) 
Introduce hypre_DeviceItem to further unify cuda/hip/sycl implementation.
Unify some wrappers for thread/warp-level kernel routines.
 2022-06-28 08:27:36 -07:00
Ruipeng Li
e270c561b0
Spgemm (#639) 
This PR includes optimizations for hypre's SpGEMM and ParSpGEMM kernels

Co-authored-by: Wayne Mitchell <mitchell82@llnl.gov>
Co-authored-by: Paul T. Bauman <ptbauman@gmail.com>
Co-authored-by: Sarah Osborn <30503782+osborn9@users.noreply.github.com>
 2022-06-24 10:42:16 -07:00
Victor A. Paludetto Magri
8268b9f1e1
hypre_ParCSRMatrixPrintIJ on device (#655) 
hypre_ParCSRMatrixPrintIJ works for matrices living on the device w/o the need of UVM support. A explicit copy is to host memory is performed in this function prior to printing the files.
 2022-06-22 20:49:57 -04:00
Victor A. Paludetto Magri
850fd47d07
Fix chebyshev smoother for singular problems (#657) 
See PR's description for additional info
 2022-06-22 20:47:09 -04:00
Ruipeng Li
b58585e0f0
add a func (#646) 
This PR adds a function to perform local transposition of ParCSR.
 2022-06-21 08:52:21 -07:00
Ruipeng Li
322e6a5e6e astyle 2022-06-17 10:14:00 -07:00
Ruipeng Li
18f85886ff remove debug code 2022-06-17 09:15:08 -07:00
Ruipeng Li
b9b93c45ef save debug code of Pass0 2022-06-17 09:13:58 -07:00
Ruipeng Li
8d54b78730 fix some nvtx region names 2022-06-15 23:18:00 -07:00
Ruipeng Li
3509640354 optimized interp_trunc 2022-06-15 22:55:35 -07:00
Rob Falgout
03e0150ee4 Change release number to 2.25.0 2022-06-13 17:17:54 -07:00
Ruipeng Li
6b34e985eb minor change 2022-06-13 16:20:03 -07:00
Ruipeng Li
dfa39f680b filtering out srun warnings 2022-06-13 11:11:08 -07:00
Ruipeng Li
5dbe0a2a10 saved.tioga 2022-06-12 23:46:01 -07:00
Ruipeng Li
a5b10499cb tune for AMD GPUs 2022-06-12 11:34:22 -07:00
Ruipeng Li
ee88d0b1a4 header 2022-06-11 11:49:56 -07:00
Ruipeng Li
c76cd1fd04 bug fix 2022-06-10 15:08:56 -07:00
Ruipeng Li
84baffc1da add license 2022-06-10 15:08:41 -07:00
Ruipeng Li
cb5814c4f4 minor change 2022-06-10 15:07:57 -07:00
Ruipeng Li
6ce70c135e bug fix 2022-06-10 14:47:46 -07:00
Ruipeng Li
7a7a63f802 Merge branch 'spgemm' of github.com:hypre-space/hypre into spgemm 2022-06-10 12:50:32 -07:00
Ruipeng Li
834bac5d8f update bench perf. on lassen 2022-06-10 12:50:24 -07:00
Ruipeng Li
637af55397 Merge branch 'spgemm' of github.com:hypre-space/hypre into spgemm 2022-06-10 12:14:35 -07:00
Ruipeng Li
c9b9a36354 update bench perf. on ray 2022-06-10 12:14:20 -07:00
Wayne Mitchell
3a735f9ccd Merge branch 'spgemm' of github.com:hypre-space/hypre into spgemm 2022-06-10 19:07:39 +00:00
Wayne Mitchell
951fa56baa Fix copy paste error 2022-06-10 19:06:34 +00:00
Wayne Mitchell
7dc8321f4a Additional fixes for sycl build. Note that CSR matrix memory location must be set correctly before calling hypre_CSRMatrixSetRownnz. 2022-06-10 19:02:35 +00:00
Ruipeng Li
eb77ea2da2 Merge branch 'spgemm' of github.com:hypre-space/hypre into spgemm 2022-06-10 11:05:48 -07:00
Ruipeng Li
2782ac5079 lassen saved files 2022-06-10 11:04:56 -07:00
Ruipeng Li
d34a1fac60 update saved ray 2022-06-10 09:35:15 -07:00
Ruipeng Li
68934f9604 bug fix? 2022-06-09 15:35:23 -07:00
Ruipeng Li
4224e7227a change the default (non-binned) back 2022-06-09 15:34:24 -07:00
Ruipeng Li
78eb2ed9b3 minor changes 2022-06-09 15:16:03 -07:00
Ruipeng Li
f7e652ca63 add more printf 2022-06-09 15:14:53 -07:00
Wayne Mitchell
3cc0138a9b Merge changes from Ruipeng and finish fixing up sycl. Seems to work now. 2022-06-09 20:34:00 +00:00
Wayne Mitchell
8730346c03 More debugging code and astyle. About to sync up with Ruipeng. 2022-06-09 19:39:03 +00:00
Ruipeng Li
67a2588888 Cmake and make files 2022-06-09 08:51:49 -07:00
Ruipeng Li
5b35ee3a41 bug fix 2022-06-09 00:31:47 -07:00
Ruipeng Li
26d53aacce bug fix 2022-06-09 00:11:31 -07:00
Ruipeng Li
f6edbf7631 Merge branch 'spgemm' of github.com:hypre-space/hypre into spgemm 2022-06-08 17:21:42 -07:00
Ruipeng Li
2eb2cc7143 saved ray 2022-06-08 17:21:24 -07:00
Ruipeng Li
d461c7184a update lassen saved 2022-06-08 17:19:42 -07:00
Ruipeng Li
db4fd264db bug fix 2022-06-08 15:33:47 -07:00
Ruipeng Li
9036263552 minor changes 2022-06-08 13:30:38 -07:00
Ruipeng Li
ae31f37135 change binned to be the default 2022-06-08 12:40:43 -07:00
Ruipeng Li
1112186d12 add some comments 2022-06-08 11:42:46 -07:00
Wayne Mitchell
d80c0f3550 Merge branch 'spgemm' of github.com:hypre-space/hypre into spgemm 2022-06-08 18:32:56 +00:00
Wayne Mitchell
d8c6556e7e Fixes for sycl. Still debugging incorrect results. 2022-06-08 18:32:49 +00:00
Ruipeng Li
430241d7d0 fix matrix num cols 2022-06-08 11:29:01 -07:00
Ruipeng Li
0eab14592b code style 2022-06-08 10:53:39 -07:00
Ruipeng Li
a436b670db update saved.ray 2022-06-08 10:30:45 -07:00
Ruipeng Li
f867d600b2 remove commented code 2022-06-07 15:28:33 -07:00
Ruipeng Li
67c545fce0 bug fix matvecT 2022-06-07 15:28:11 -07:00
Ruipeng Li
3f39f5d4fa update printMM 2022-06-07 15:26:40 -07:00
Ruipeng Li
8b02ab88d3 GPU support for rownnz using IntArray 2022-06-07 12:08:43 -07:00
Ruipeng Li
c09ab567e7 remove a function 2022-06-07 10:13:11 -07:00
Ruipeng Li
31d4f891bf turn off timing 2022-06-06 21:46:46 -07:00
Ruipeng Li
f4917850e7 Merge branch 'master' of github.com:hypre-space/hypre into spgemm 2022-06-06 21:45:49 -07:00
Ruipeng Li
57f0ddd75c removed old impl 2022-06-06 21:40:36 -07:00
Ruipeng Li
c12449c44c sycl impl. 2022-06-06 16:42:26 -07:00
Ruipeng Li
0fee4f3c80 astyle 2022-06-06 13:23:32 -07:00
Ruipeng Li
4efa15836d support P100 2022-06-06 13:17:19 -07:00
Victor A. Paludetto Magri
07a8def6f8
Fix compilation warnings (#643) 
This PR fixes compilation warnings obtained with gcc-11, clang-12, and clang-14. A list of the warnings is given below:

* -Wundef
* -Wunused
* -Wdouble-promotion
* -Wsometimes-uninitialized
* -Wunused-variable
* -Wunused-but-set-variable
 2022-05-31 20:32:49 -04:00
Ruipeng Li
0aa6d52f92 style change 2022-05-27 10:56:32 -07:00
Ruipeng Li
8070877390 code style 2022-05-27 10:21:07 -07:00
Ruipeng Li
8c58483b56 Merge branch 'master' of github.com:hypre-space/hypre into spgemm 2022-05-27 09:44:49 -07:00
Ruipeng Li
270bbe6e01 astyle 2022-05-27 09:42:35 -07:00
Ruipeng Li
96991d3d30 name changes 2022-05-26 20:03:07 -07:00
Ruipeng Li
875180594d change some var names 2022-05-26 19:54:32 -07:00
Victor A. Paludetto Magri
edb91b4a50
Add -auxfromfile option to IJ driver (#633) 
Add -auxfromfile option for reading an auxiliary matrix from file, which is then used to build the preconditioner. This is useful, for example, for the case when a filtered version of A is used to build the preconditioner.
 2022-05-26 21:23:31 -04:00
Ruipeng Li
17bc61ed48 Merge branch 'spgemm' of github.com:hypre-space/hypre into spgemm 2022-05-26 16:56:43 -07:00
Ruipeng Li
9cd6eb1ff8 bug fix 2022-05-26 16:56:32 -07:00
Ruipeng Li
4ac9d00577
Parspgemm (#641) 
Merge two spgemm branches into one
 2022-05-26 14:38:15 -07:00
Ruipeng Li
09eb54520c fix after merge 2022-05-26 10:07:11 -07:00
Ruipeng Li
4e78801d2a Merge branch 'master' of github.com:hypre-space/hypre into spgemm 2022-05-25 16:36:44 -07:00
Ruipeng Li
e766e36e76
Add header to remove header-transitivity issue (#636) 
Add header for `thrust::remove_if.`
Co-authored-by: Paul T. Bauman <ptbauman@gmail.com>
 2022-05-24 14:18:57 -07:00
Ruipeng Li
1c1bf95b10
fix with umpire 2022 (#625) 
This PR fixes hypre with umpire 2022
 2022-05-24 13:55:08 -07:00
Ruipeng Li
a8d423013b
missing -f in softlink commands (#594) 
This PR adds in a few of the Mafefiles the missing -f in the softlink commands.

Co-authored-by: Paul Mullowney <Paul.Mullowney@nrel.gov>
 2022-05-24 13:39:33 -07:00
Paul T. Bauman
565bbe0511
Need to read start/end indices as HYPRE_BigInt (#605) 
This fixes an issue with the ParCSRMatrixRead when compiled with mixed-int enabled.
 2022-05-24 13:32:56 -07:00
Ruipeng Li
aa0446d720
ij help (#634) 
Minor fix to `ij -help`
 2022-05-24 13:30:23 -07:00
Ruipeng Li
ef3f890d4b
Nvcollab (#591) 
This PR contains various GPU optimizations in the collaboration with the NVIDIA team. 

Co-authored-by: Peng Wang <penwang@nvidia.com>
 2022-05-24 13:27:32 -07:00
Ruipeng Li
bec8645cf9
script option for runtest.sh (#632) 
This PR changes runtest.sh to run an executable with a "script"  and allows valgrind and mpibind on all platforms.
 2022-05-19 10:10:03 -07:00
Daniel Osei-Kuffuor
63208e3e34
Hotfix for issues with dsuperlu in regression test. (#631) 
Commented out unnecessary memory deallocation check.
 2022-05-15 21:42:28 -07:00
Wayne Mitchell
dfdd1cd12f
Sycl par matmat (#611) 
Further unification of GPU implementation across cuda/hip/sycl.
Implements the parallel matrix matrix product in sycl.
HYPRE_CUDA_LAUNCH and HYPRE_SYCL_LAUNCH macros have 
been unified under HYPRE_GPU_LAUNCH for kernel launches.
Replace HYPRE_SetSpGemmUseCusparse with HYPRE_SetSpGemmUseVendor.
 2022-05-09 15:24:44 -07:00
Daniel Osei-Kuffuor
00d1dfd3f7
Mgr block jacobi (#607) 
* Added new capabilities to allow multilevel assignment of solver options
* New (local) block Jacobi option for smoothers and intergrid operators
* Added capabilities to do CPR in MGR
* Updated non-Galerkin strategy for constructing the coarse grid.

Co-authored-by: Quan Bui <mquan.bui@gmail.com>
 2022-05-09 08:30:05 -07:00
Ruipeng Li
117963fb19 fixed a debugging check 2022-04-13 17:01:07 -07:00
Ruipeng Li
3819f474f0 A: row nnz = 1, numer 2022-04-13 16:01:42 -07:00
Ruipeng Li
6450927874 A: row nnz = 1, numer (1st version) 2022-04-12 11:18:42 -07:00
Ruipeng Li
df10f0b5ed A: row nnz = 1, symbl 2022-04-12 11:17:56 -07:00
Ruipeng Li
4025b32ad8 update driver 2022-04-07 13:11:32 -07:00
Victor A. Paludetto Magri
8017ce459b
Fix segfault on HYPRE_SStructGraphDestroy (#617) 
This PR fixes a segmentation fault on HYPRE_SStructGraphDestroy. The error occurred when the number of graph entries added to the SStructGraph via HYPRE_SStructGraphAddEntries was larger than 1000.
 2022-04-06 21:12:41 -07:00
Ruipeng Li
4421ebd83a Parprint; function name change of sync stream 2022-04-06 18:13:49 -07:00
Ruipeng Li
096ce6c1dd driver changes 2022-04-06 17:39:36 -07:00
Ruipeng Li
e21778682b minor changes 2022-04-06 17:39:23 -07:00
Victor A. Paludetto Magri
70d055a994
Fix complex build (#616) 
This PR fixes compilation of the "complex" build variant of hypre. It also adds hypre_csqrt for computing the square root of an HYPRE_Complex number. This function/macro works when enable-complex is turned on/off.
 2022-04-06 15:02:04 -07:00
Ruipeng Li
224bb78d4f Merge branch 'spgemm' of github.com:hypre-space/hypre into parspgemm 2022-04-05 23:35:19 -07:00
Ruipeng Li
046b278c66 bug fix 2022-04-05 22:26:29 -07:00
Ruipeng Li
db8fbbf119 minor change in timing 2022-04-05 22:26:11 -07:00
Rob Falgout
4c5529810a Updating one missed copyright date in user manual 2022-04-05 16:40:02 -07:00
Victor A. Paludetto Magri
e16167fe46
Fix copyright (#615) 
This PR updates Copyright headers from "Copyright 1998-2019 ..." to "Copyright (c) 1998 ..."
 2022-04-05 16:19:51 -07:00
Victor A. Paludetto Magri
9415d6aa08
FSAI implementation on CPUs (#610) 
Thir PR adds a factorized sparse approximate inverse (FSAI) implementation on hypre, which can be used as a standalone solver, preconditioner to Krylov methods, or complex smoother to BoomerAMG. Particularly, we consider the adaptive algorithm version, where the sparsity pattern of the lower triangular factor G is built dynamically, i.e., during an iterative procedure that tries to find the best nonzero positions for a given row of G. This implementation was performed on top of the IJ interface. It uses the diagonal portion of A for constructing G, i.e., it's a block-Jacobi method in the MPI sense. List of additional changes:

* Add caliper instrumentation to FSAI.
* Add ZeroGuess option to FSAI.
* Performance optimizations.
* Add OpenMP support to FSAI.
* Make internal BLAS/LAPACK functions thread-safe. 
* Update CMake build.
* Add new test cases: beam_tet_dof459_np1, beam_hex_dof459_np2, and beam_tet_dof2475_np4.
* Add documentation for FSAI.

Co-authored-by: Heather Switzer <switzer4@lassen36.coral.llnl.gov>
Co-authored-by: heatherms27 <hmswitzer@email.wm.edu>
Co-authored-by: Sarah Osborn <30503782+osborn9@users.noreply.github.com>
 2022-04-05 11:18:39 -07:00
Ruipeng Li
7b8377d3b1 new impl. of partial add 2022-04-05 10:19:44 -07:00
Ruipeng Li
e9c019e40f parcsr matvec T 2022-04-02 14:19:45 -07:00
Ruipeng Li
87b0b6669a update hypre's spmv 2022-04-02 14:16:08 -07:00
Ruipeng Li
d1774a1ded Merge branch 'sycl_par_matmat' of github.com:hypre-space/hypre into parspgemm 2022-04-01 12:45:42 -07:00
Ruipeng Li
d4b5c26ef9 new impl. of partial add 2022-04-01 12:21:19 -07:00
Ruipeng Li
f5950dee92 whitespace changes 2022-04-01 12:20:33 -07:00
Ruipeng Li
dcead56dd2 header 2022-03-31 19:03:42 -07:00
Ruipeng Li
aea2e7303f configure 2022-03-31 19:01:36 -07:00
Ruipeng Li
8ea39950b1 Merge branch 'nvcollab' of github.com:hypre-space/hypre into spgemm 2022-03-31 18:58:36 -07:00
Ruipeng Li
66b624de9e update ij_mm 2022-03-31 18:23:03 -07:00
Wayne Mitchell
bc30046d08 Removing guards for zero size onedpl calls where not needed 2022-04-01 00:50:22 +00:00
Ruipeng Li
1c35a59c1d astyle 2022-03-31 17:48:19 -07:00
Ruipeng Li
6086b07a96 minor change 2022-03-31 17:38:35 -07:00
Ruipeng Li
6e8607fd47 reorg parcsrmm 2022-03-31 17:38:19 -07:00
Wayne Mitchell
996dfaecd5 Revert changes in solvers.jobs 2022-03-31 22:01:35 +00:00
Ruipeng Li
8e34f6400f
fix domp build (#613) 2022-03-31 14:25:57 -07:00
Ruipeng Li
23c7777045 Merge branch 'sycl_par_matmat_domp' of github.com:hypre-space/hypre into parspgemm 2022-03-30 09:19:52 -07:00
ulrikeyang
303457abae
fixed MM-multipass interpolation for case of no C-points (#606) 
* fixed MM-multipass interpolation for case of no C-points

* fixed the issue of isolated groups of fine points and added a regression test.

* corresponding changes to the device code

Co-authored-by: Ruipeng Li <li50@llnl.gov>
 2022-03-29 15:14:29 -07:00
Ruipeng Li
e5393bc0eb fix domp build 2022-03-28 22:55:18 -07:00
Ruipeng Li
caec03c7ad Merge branch 'sycl_par_matmat' of github.com:hypre-space/hypre into parspgemm 2022-03-25 18:59:19 -07:00
Ruipeng Li
23b8527c1d fix after merge 2022-03-25 18:53:02 -07:00
Ruipeng Li
68be11b70f fix after merge; change CUDAKernel to GPUKernel 2022-03-25 18:52:40 -07:00
Wayne Mitchell
ce30e2faa4 Backward compatibility for UseCusparse 2022-03-26 01:09:17 +00:00
Ruipeng Li
ac65ca58fe changes to mm 2022-03-25 15:53:16 -07:00
Wayne Mitchell
5f7608b875 Change UseCusparse to UseVendor 2022-03-25 22:07:05 +00:00
Wayne Mitchell
112691e2d1 Remove debugging code 2022-03-25 21:07:58 +00:00
Wayne Mitchell
b860e28c72 Clean up 2022-03-25 20:57:41 +00:00
Wayne Mitchell
bb2cb43232 Merge branch 'master' into sycl_par_matmat 2022-03-25 20:27:17 +00:00
Wayne Mitchell
b5de7f7dfd MGR bug fix and no longer throw error for zero len kernel launches for sycl 2022-03-25 19:41:45 +00:00
Wayne Mitchell
6b5a13cedf Bug fix! Still need to clean up some further debugging code. 2022-03-25 00:56:39 +00:00
Wayne Mitchell
511c0ec77e Remove some debugging code. Temporary changes to solvers.jobs 2022-03-23 18:49:46 +00:00
Ruipeng Li
5fe37b2286
hypre_ParPrintf (#604) 
This PR adds hypre_ParPrintf. Prints to standard out, only from the first processor in the communicator. Calls from other processes are ignored.
 2022-03-21 09:10:01 -07:00
Wayne Mitchell
f2fa2e9577 Lots of ugly debugging code in here, but I have also fixed a couple esoteric things. Saving with debugging code in just in case I need to go back and use it. 2022-03-18 23:49:23 +00:00
Ruipeng Li
5513fd5115 Merge branch 'ParPrintf' of github.com:hypre-space/hypre into parspgemm 2022-03-18 15:20:52 -07:00
Ruipeng Li
7300f4b937 Merge branch 'master' of github.com:hypre-space/hypre into parspgemm 2022-03-18 15:17:45 -07:00
Ruipeng Li
04239e769a remove hypre_vprintf 2022-03-18 13:32:22 -07:00
Ruipeng Li
96985f04e2 ParPrintf; hypre_vprintf 2022-03-18 12:10:47 -07:00
Ruipeng Li
adfd07c509
Fix build on FreeBSD/powerpc*. (#603) 
lr collides with lr from machine/frame.h header (link register):
Co-authored-by: Piotr Kubaj <pkubaj@FreeBSD.org>
 2022-03-18 10:16:20 -07:00
Rob Falgout
fa43ea82e3
Bug fix in prefix sum for OpenMP IJ interface (#602) 
This fixes a bug found in issue #522 for the prefix sum openmp code in IJ.
 2022-03-16 10:56:09 -07:00
Ruipeng Li
e5f6655ba0 initial support for pattern only matrices (spgemm only) 2022-03-16 09:32:12 -07:00
Ruipeng Li
d079a90fb8 Merge branch 'master' of github.com:hypre-space/hypre into nvcollab 2022-03-15 08:32:25 -07:00
Ruipeng Li
92faac9748
fix memory location (#600) 
This PR fixes a number of memory location issues in memory copy and memset. It also adds more strict checking in memory.c in the debug mode.
 2022-03-14 11:19:28 -07:00
Victor A. Paludetto Magri
6fd043c9c2
(S)Struct IO on GPUs (#599) 
This PR extends the (semi)-struct matrix/vector IO functions added on #583 with GPU support. Additionally:

* Fix regression tests on Lassen.
* Read data values into host memory
* Update Umatrix read algorithm when the ParCSRMatrix is expected to live on the device
* Reset deallocated pointers at hypre_IJMatrixDestroyParCSR to NULL
* Clone rownnz info if present on a CSRMatrix
* Reduce memory transfer and remove unused variables
* Fix bug with -print option
* Build rownnz info also when the ParCSRMatrix is in assembled state
* Remove a few instances of "return ierr"
* Refactor (s)struct IO - code works with cuda and without UM
* Add executables to gitignore
 2022-03-13 20:14:23 -07:00
Ruipeng Li
c9a6c67229 Merge branch 'master' of github.com:hypre-space/hypre into nvcollab 2022-03-11 18:34:45 -08:00
Ruipeng Li
f7787ab0ae
fixes coarsening.jobs.14 (#598) 
a temporary "fix"
 2022-03-11 18:25:03 -08:00
Ruipeng Li
793b22aaf0 Merge branch 'nvcollab' of github.com:hypre-space/hypre into nvcollab 2022-03-11 08:34:20 -08:00
Ruipeng Li
97f3254d71 Merge branch 'master' of github.com:hypre-space/hypre into nvcollab 2022-03-11 08:33:31 -08:00
Ruipeng Li
8c344aee9a
Invalid assumption on exclusive_scan (#575) 
This PR fixes a number of initialization problems with exclusive_scan on GPUs due to invalid assumptions of this function.
 2022-03-11 08:32:26 -08:00
Rui-peng Li
700b0328bf bug fix 2022-03-10 22:19:22 -06:00
Ruipeng Li
9498625db4 Merge branch 'nvcollab' of github.com:hypre-space/hypre into nvcollab 2022-03-10 18:31:13 -08:00
Ruipeng Li
50c88ab95d minor changes 2022-03-10 18:30:52 -08:00
Ruipeng Li
c143265c61 regression on ray 2022-03-10 17:41:49 -08:00
Ruipeng Li
14a336c28b accidentally deleted io.sh 2022-03-10 13:30:31 -08:00
Ruipeng Li
2d06b53c4e benchmark ij on lassen 2022-03-10 13:26:15 -08:00
Ruipeng Li
815f2d57e0 add cublas/cusparse precision macros 2022-03-10 12:12:32 -08:00
Ruipeng Li
90cbe64fee saved.lassen 2022-03-10 08:45:27 -08:00
Ruipeng Li
7681f7f180 add cublas to makefile 2022-03-10 08:43:44 -08:00
Ruipeng Li
8ecee0b47d update hypre's spmv 2022-03-09 22:36:16 -08:00
Ruipeng Li
31ca2338d2 saved.lassen 2022-03-09 16:22:22 -08:00
Ruipeng Li
009501d51c bug fix 2022-03-09 16:14:56 -08:00
Ruipeng Li
86dae0be5b a minor change 2022-03-09 14:29:22 -08:00
Ruipeng Li
1dc1261fe8 fix cpu regression 2022-03-09 14:26:40 -08:00
Ruipeng Li
f8fd57ab2a updated saved perf on ray 2022-03-09 13:54:00 -08:00
Ruipeng Li
9b8627ce84 update lassen banchmark saved results 2022-03-09 13:09:36 -08:00
Ruipeng Li
7a8cf68b9a add -repeats 2 for struct benchmark jobs 2022-03-09 13:09:07 -08:00
Ruipeng Li
9dda5af3c4 struct.c driver for reps == 2 2022-03-09 12:31:46 -08:00
Ruipeng Li
63c9fa65a2 add using hypre's spmv option 2022-03-08 22:11:31 -08:00
Wayne Mitchell
5e90f43cbc astyle 2022-03-08 23:51:41 +00:00
Wayne Mitchell
34c16787b7 Addition of MatMat and TMatMat routines and clean up 2022-03-08 23:50:48 +00:00
Wayne Mitchell
297ff5d5a7 Par matmat is verified correcct for a small example 2022-03-08 21:08:47 +00:00
Ruipeng Li
7443a2ac6c missed some sync in the last commit 2022-03-07 23:54:03 -08:00
Ruipeng Li
e1b9a56405 add gpu sync for mpi 2022-03-07 23:34:33 -08:00
Ruipeng Li
8ee20f4812 cudamallocasync 2022-03-07 16:54:56 -08:00
Ruipeng Li
df0f6dbba7 configure options: cublas; cudamallocasync 2022-03-07 16:40:32 -08:00
Ruipeng Li
d7728d0bce updated ij driver for 2nd solve 2022-03-07 15:16:00 -08:00
Ruipeng Li
b97fbc13ed sync device at ending timing 2022-03-07 15:13:08 -08:00
Ruipeng Li
c2e4836c1e bug fix 2022-03-05 10:36:53 -08:00
Ruipeng Li
a51bb880a8 bug fix 2022-03-05 09:46:16 -08:00
Ruipeng Li
03546b428f Merge branch 'master' of github.com:hypre-space/hypre into nvcollab 2022-03-04 22:18:02 -08:00
Paul T. Bauman
251cd3d269
Need -O1 instead of -O0 for HIP in debug mode (#588) 
This PR changes -O0 in debug mode to -O1 with HIP (at this time).
 2022-03-04 12:40:35 -08:00
Ruipeng Li
95e6433fc7
GPU support with single precision (#572) 
This PR fixes the GPU support with single precision.
 2022-03-04 12:05:32 -08:00
Ruipeng Li
ebd6eb88c3
bug fix; nonsquare rap (#581) 
This PR fixes a corner case of the RAP routine for RAP matrix that is globally square but not locally.
 2022-03-03 21:26:17 -08:00
Wayne Mitchell
d388a2766e Lots of reorganization. This now has all functionality for par matmat and compiles, but needs debugging. 2022-03-01 01:59:27 +00:00
Wayne Mitchell
8112dd736f Further cleanup and reorg of device_utils.c/h and addition of more functionality needed for par matmat 2022-02-19 00:40:27 +00:00
Paul T. Bauman
04af9a4cd9
HYPRE_Int -> HYPRE_BigInt (#585) 2022-02-18 12:16:35 -08:00
Wayne Mitchell
cda5b10a69 Single processor device rap works 2022-02-18 18:58:28 +00:00
Wayne Mitchell
47ae1c8a22 Start major reorganization of device_utils.h 2022-02-18 17:44:22 +00:00
Golam Rabbani
94070dd3a9
Updated CMakeLists.txt for SYCL (#577) 
With CMake, enable CUDA stream by default when using SYCL.
 2022-02-17 18:21:51 -08:00
Victor A. Paludetto Magri
33a5051398
Add SStruct IO functions (#583) 
This PR adds support for native print/read functions of SStructMatrix and SStructVector. Other important changes are:
* Add public functions for reading StructMatrix and StructVector.
* Add a new set of regression tests called "io" to the TEST_sstruct folder.
 2022-02-17 18:06:23 -08:00
Wayne Mitchell
1e289479f1 astyle 2022-02-17 00:56:49 +00:00
Ruipeng Li
9888903445 memory tracker 2022-02-16 15:04:20 -08:00
Ruipeng Li
c336122cc1 move to debug region 2022-02-16 14:19:15 -08:00
Victor A. Paludetto Magri
49dbf7b60a
Fix cross-compilation problem (#580) 
This PR fixes issue #556.

AC_CHECK_FILE was being used to test the existence of the .git folder. However, according to Autoconf manual, it does not work when cross-compiling. This PR implements another strategy for looking for the .git folder which works also when doing cross-compilation.
 2022-02-16 07:55:02 -08:00
Wayne Mitchell
4136c63269 Switch to HYPRE_GPU_LAUNCH 2022-02-15 20:01:08 +00:00