1
0
mirror of https://github.com/paboyle/Grid.git synced 2025-04-23 20:25:56 +01:00

404 Commits

Author SHA1 Message Date
Peter Boyle
442336bd96 Hand unrolled to use optimised code paths on GPU for coalesced reads in Wilson case.
Other cases to do. This now includes comms code path.
2021-03-02 14:50:51 +01:00
Christoph Lehner
9c9566b9c9
Merge pull request #23 from paboyle/develop
Sync
2021-03-01 12:33:51 +01:00
Daniel Richtmann
e3d019bc2f
Enable performance counting in WilsonFermion like in others 2021-02-22 15:25:40 +01:00
Peter Boyle
eda9ab487b MADWF 5d source option for hadrons - look at Grid of source
Abort on GPU error
2021-02-08 10:47:22 -05:00
Peter Boyle
69f1f04f74 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-01-21 21:39:59 -05:00
Peter Boyle
ff1fa98808 Fix for GPU conserveed current 2021-01-21 21:38:23 -05:00
Peter Boyle
99445673f6 Gparity fix, and plaquette IO 2021-01-14 21:00:36 -05:00
Christoph Lehner
299d0de066
Merge pull request #21 from paboyle/develop
Sync
2020-12-22 20:59:15 +01:00
Nils Meyer
45d49d8648 clean up 2020-12-19 03:35:18 +01:00
Nils Meyer
3f9ae6e7e7 Merge branch 'develop' into feature/a64fx-3 2020-12-19 02:37:11 +01:00
Nils Meyer
4dd9e39e0d up to +36% performance gain for dslash/dwf on QPACE 4 using GCC 10.1.1 2020-12-19 00:54:31 +01:00
Michael Marshall
873519e960 Enable existing conserved current code for CUDA (compiles OK for CUDA 10.1). Add option to Test_cayley_mres to load a configuration 2020-12-14 16:06:10 +00:00
Daniel Richtmann
c438118fd7
Change access specifier of clover fields in order to allow deriving classes to access these 2020-12-08 14:42:11 +01:00
Christopher Kelly
2ef1fa66a8 Improved performance of G-parity kernel for GPUs by simplifying multLink implementation 2020-12-07 11:53:35 -05:00
Peter Boyle
321f0f51b5 Project to SU(N) 2020-11-24 21:46:10 -05:00
Daniel Richtmann
b3881d2636
Thread inversion of clover term 2020-10-30 16:18:58 +01:00
Peter Boyle
bf3c9857e0 Closure changes 2020-10-14 21:37:14 -04:00
Peter Boyle
ace9cd64bb dpcpp happy 2020-09-29 08:03:46 -07:00
Peter Boyle
ecd3f890f5 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2020-09-16 02:30:14 +01:00
Peter Boyle
2859955a03 HIP requires "inline" 2020-09-16 00:36:13 +01:00
Peter Boyle
cc220abd1d inline for HIP 2020-09-16 00:35:38 +01:00
Peter Boyle
d1c0c0197e HipCC requires inline on definition 2020-09-16 00:35:06 +01:00
Peter Boyle
fd9424ef27 innlines required to make HIP happy 2020-09-16 00:34:32 +01:00
Peter Boyle
a5c35c4024 Make HIP / Vega happy 2020-09-16 00:33:53 +01:00
Peter Boyle
6dbd117aa5 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2020-09-03 20:30:49 -04:00
Peter Boyle
b4255140d6 Stale data member eliminated 2020-09-03 15:47:46 -04:00
Peter Boyle
8807d998bc closure improved 2020-09-01 15:29:11 -04:00
Christoph Lehner
0e88bf4bff remove Nils's default pragma 2020-07-29 10:24:35 -04:00
nmeyer-ur
bbd145382b enable --enable-simd=A64FX in configure 2020-07-08 12:43:51 +02:00
nmeyer-ur
1b08cb7300 Merge branch 'develop' into feature/a64fx-2 2020-07-08 08:18:18 +02:00
nmeyer-ur
8726e94ea7 merge upstream develop 2020-07-07 20:26:47 +02:00
f1f655d92b
Merge pull request #304 from Heinrich-BR/develop
ScalarImpl.h updates
2020-07-06 10:16:03 +01:00
43334e88c3 Tiny change in a comment for clarity 2020-07-04 16:11:16 +01:00
4f1e66b044 Fixed HMC SU(N) integrator which was causing fields to leave Lie Algebra manifold for N>2 2020-07-04 03:53:06 +01:00
Peter Boyle
b949cf6b12 PeekLocal needs a view to keep thread safe.
ALLOCATION_CACHEE reenable
2020-06-19 17:13:27 -04:00
Peter Boyle
1aa988b2af Comms overlap fix UVM case 2020-06-19 01:21:14 -04:00
Peter Boyle
237ce92540 Offload loops 2020-06-10 19:59:11 -04:00
Peter Boyle
fd97f64612 Merge branch 'sycl' of https://github.com/paboyle/Grid into sycl 2020-06-10 12:58:13 -04:00
Peter Boyle
8720aecb80 Offload more loops 2020-06-10 12:57:55 -04:00
Peter Boyle
cdf0a04fc5 Merge branch 'develop' into sycl 2020-06-09 04:00:12 -04:00
Peter Boyle
e97f3688db Fix the HMC issue - kernel was launchnig asynchronously 2020-06-08 17:01:15 -04:00
nmeyer-ur
433766ac62 revert Add/SubTimesI and prefetching in stencil
This reverts commit 9b2699226c7a3ca8d45f843f4f8e4658fa082163.
2020-06-08 12:02:53 +02:00
Peter Boyle
1a4c8c3387 Global edit with change to View usage. autoView() creates a wrapper object that closes the view when scope closes. 2020-06-05 18:52:35 -04:00
nmeyer-ur
5ee3ea2144 round-up after testing of prefetches in stencil close 2020-06-03 11:58:20 +02:00
nmeyer-ur
91c81cab30 some corrections; compiles on my laptop; untested 2020-05-29 18:19:22 +02:00
nmeyer-ur
38164f8480 include counters in WilsonFermionImplementation.h 2020-05-29 17:59:26 +02:00
nmeyer-ur
f013979791 add counter support in WilsonFermion.h 2020-05-29 17:13:59 +02:00
Peter Boyle
1d252d0922 Accelerator inline 2020-05-28 11:45:25 -04:00
Peter Boyle
006cc8a8f1 Staggereed move to accelerator 2020-05-28 08:33:06 -04:00
Peter Boyle
7860a50f70 Make view specify where and drive data motion - first cut.
This is a compile tiime option --enable-unified=yes/no
2020-05-21 16:13:16 -04:00