Commit Graph
37 Commits
Author SHA1 Message Date
sfilippone 5903c0b272 Fix DOT in OpenACC 2024-10-08 17:41:42 +02:00
sfilippone 4822861b73 Merge branch 'oacc_loloum' of github.com:sfilippone/psblas3 into oacc_loloum 2024-10-08 17:37:28 +02:00
sfilippone 49469ce021 Various changes into openacc 2024-10-08 17:36:44 +02:00
sfilippone 740609a4d8 Fix present() clauses 2024-10-07 12:45:18 +02:00
sfilippone 68f20c0e7a Modify init 2024-10-07 12:44:45 +02:00
sfilippone e5504ddddc Fix memory traffic in GTH/SCT 2024-09-12 15:08:21 +02:00
sfilippone 096bce08c1 Merged changes from V4 OpenACC 2024-09-11 12:43:46 +02:00
sfilippone bcbe0c89c7 Backporting fixes from version 4 2024-09-10 17:37:35 +02:00
sfilippone 6236f3489c Remove obsolete files 2024-08-30 16:05:08 +02:00
sfilippone f783478df3 Merge updates from V4 2024-08-30 16:03:04 +02:00
sfilippone fbb974fb8b Change name sync|free space, unify allocate impl 2024-08-30 10:20:58 +02:00
sfilippone 479135c62d Merge some changes from V4 2024-08-29 16:36:24 +02:00
sfilippone 95c546aadd Fix OpenACC version of ELL vect_mv 2024-08-26 08:22:35 +02:00
sfilippone fa5e7ff945 Fixes for vector methods and sync() 2024-08-20 19:38:34 +02:00
sfilippone 9314b2cf53 Fix missing method in oacc_ell 2024-08-08 15:08:45 +02:00
sfilippone cf2cc6cab9 Precedence of oacc_vect modules 2024-08-08 14:34:28 +02:00
sfilippone 2709aa9f16 Fix upd_xyz name 2024-08-08 11:10:04 +02:00
sfilippone 2982aaee27 Implementation in OpenACC for ELL and HLL into templates. Merge from development 2024-08-08 11:07:28 +02:00
sfilippone 464baceb13 HLL loop nest cannot run with collapse 2024-08-07 10:54:19 +02:00
tloloum 55d1067ec2 collapse loop 2024-08-06 11:42:52 +02:00
tloloum e6fa1d17a2 oacc hll 2024-07-30 14:25:01 +02:00
tloloum 10ec5eafab ELL oacc impl 2024-07-26 09:56:32 +02:00
tloloum 162b2fc78f beginning of oacc_ell 2024-07-22 15:27:33 +02:00
sfilippone 9e18545151 Fix typos 2024-07-17 13:04:07 +02:00
sfilippone 686bac4224 Account for S/D/C/Z variants 2024-07-17 13:03:46 +02:00
sfilippone b6fe0f3344 New version for modules and methods 2024-07-17 08:52:43 +02:00
sfilippone 7006665d82 Introduce submodules, adjust Makefile 2024-07-12 16:34:48 +02:00
sfilippone 0707cc0a72 Take out reference to elldev_mod 2024-07-12 13:01:32 +02:00
sfilippone 9e8294066d Fix Makefile 2024-07-12 13:01:22 +02:00
tloloum 2b5f09ddf9 all methods implementations for psb_d_oacc_csr_sparse_mat 2024-07-12 09:26:06 +02:00
sfilippone a81d1d9b68 Now oacc_vect compiles cleanly 2024-07-09 13:45:16 +02:00
sfilippone 3bb05fab54 Merge branch 'oacc_loloum' of github.com:sfilippone/psblas3 into oacc_loloum 2024-07-09 13:18:34 +02:00
sfilippone 1f0e591827 Reworked oacc_vect_mod. oacc_mlt_v_X generates ICE, to be fixed 2024-07-09 13:16:36 +02:00
tloloum 6be998ac66 oacc_env_mod 2024-07-09 11:22:28 +02:00
sfilippone e0d7091ecc Makefile in openacc/impl 2024-07-09 10:04:32 +02:00
sfilippone 93c9df0277 Adjust Makefiles & source 2024-07-09 10:03:57 +02:00
Théophane Loloum 2b8671fba6 oacc : d_vect + i_vect + d_csr 2024-07-08 10:14:42 +02:00