1
0
mirror of https://github.com/paboyle/Grid.git synced 2024-11-10 07:55:35 +00:00
Commit Graph

6344 Commits

Author SHA1 Message Date
Peter Boyle
3aab983760 Flop count set as in DiRAC-ITT-2020 (mistaken 20% low, but must maintain consistency) 2020-11-16 17:13:58 +01:00
Peter Boyle
9c4dcc5ea3 Merge branch 'master' into develop 2020-11-16 16:34:57 +01:00
Peter Boyle
a1063ddbb9 Update options and simplify 2020-11-13 04:11:03 +01:00
Peter Boyle
18ef8056ec Hide Shared Memory 2020-11-13 04:10:40 +01:00
Peter Boyle
1c673977fa Must ask for COMMMS_THREADS 2020-11-13 03:59:36 +01:00
Peter Boyle
e9bc748828 Useful GPU machine benchmark for GDR used to shakeout Booster at Juelich - see slack earlyaccess channel 2020-11-13 03:58:34 +01:00
Peter Boyle
f48156529b Work on 2,2,2,8 ranks 2020-11-13 03:57:58 +01:00
Peter Boyle
d05ce01809 TOFU behaviour now optional THREAD_MULTIPLE or THREAD_SERIALIZED 2020-11-13 03:52:19 +01:00
Peter Boyle
cf23eff60e Device to Device, Memset, cannot assume UVM == Communicable 2020-11-13 03:51:08 +01:00
Peter Boyle
6e313575be Use of default GPU is behaviour, not a system property. Move Summit specific to configure.ac 2020-11-13 03:50:16 +01:00
Peter Boyle
b13d1f7238 TOFU compat flag to help Isaaku 2020-11-13 03:49:44 +01:00
Peter Boyle
b5e7945dd9 Option for host or device Cshift implementation 2020-11-13 01:38:54 +01:00
Peter Boyle
7535566f54 Option for bounce through the SHM buffer 2020-11-12 22:54:27 +01:00
Peter Boyle
50b808ab33 Configure option between host and device 2020-11-12 22:28:12 +01:00
Peter Boyle
f16c2665f5 Host memory explict 2020-11-12 20:29:58 +01:00
Peter Boyle
41e28015ae Volume divisible guarantee 2020-11-07 13:32:16 +01:00
Peter Boyle
a0ccbb3bd6 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2020-11-01 01:16:35 +00:00
Peter Boyle
5eeabaa2bb HIP fix 2020-11-01 01:16:01 +00:00
Peter Boyle
00d0d6d008 Hip Free managed 2020-10-31 18:14:31 -04:00
Peter Boyle
537a9f7030 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2020-10-31 18:13:30 -04:00
Peter Boyle
cc9c993f74 Project on group fix on GPU tracked to reciprocal sqrt collision between CUDA and Grid rsqrt 2020-10-31 18:12:47 -04:00
Peter Boyle
d10422ded8 Test project on group 2020-10-31 18:12:30 -04:00
Peter Boyle
f313565a3c HiP compile 2020-10-31 12:12:40 +00:00
61d5860b46
Merge pull request #318 from rrhodgson/feature/BaryonSpinMat
Added untraced baryon contraction code
2020-10-28 18:39:59 +00:00
52d17987dc BaryonUtils.h updated debug output 2020-10-23 11:41:08 +01:00
19d8bba97d BaryonUtils function naming change 2020-10-21 11:58:53 +01:00
463d72d322 Added untraced baryon contraction code 2020-10-19 16:13:28 +01:00
Peter Boyle
3362f8dfa0 happy compile 2020-10-14 22:59:41 -04:00
Peter Boyle
bf3c9857e0 Closure changes 2020-10-14 21:37:14 -04:00
Peter Boyle
a88b3ceca5 Closure cases 2020-10-14 21:33:51 -04:00
Peter Boyle
aa135412f5 toComplex, toReal 2020-10-13 22:25:01 -04:00
Peter Boyle
9945399e60 Reaality issues fix by drop from ET 2020-10-13 22:24:32 -04:00
Peter Boyle
5eeffa49e8 Reality forced included 2020-10-13 22:23:57 -04:00
Peter Boyle
3f06209720 Pretty print 2020-10-13 22:18:51 -04:00
Peter Boyle
12e239dd9f Merge branch 'release/dirac-ITT-2020' 2020-10-13 13:38:29 -04:00
Peter Boyle
af2301afbb
Merge pull request #312 from i-kanamori/debug_512
add reordring of random number generators in IO
2020-10-13 11:42:12 -04:00
Peter Boyle
f98856a26f
Merge pull request #314 from smangham/issue_readme_precision
Fix for deprecated configure options in documentation (issue #313)
2020-10-13 11:41:38 -04:00
Sam Mangham
d55cc5b380 Fixed typo on --enable-comm, removed all references to --enable-precision except for config options, where it is listed as deprecated. Removed travis test for single precision. 2020-10-12 12:33:13 +01:00
c2b688abc9 Benchmark_IO: reducing max local volume to 32^4 2020-10-10 16:52:56 +01:00
b0d61b9687 Benchmark_IO cleaner output 2020-10-09 21:46:45 +01:00
5f893bf9af Benchmark_IO procurement sizes 2020-10-09 21:31:59 +01:00
0e17bd6597 I/O benchmark cleanup 2020-10-09 20:29:57 +01:00
22caa158cc multi-pass I/O benchmark, with statistic and robustness summary 2020-10-09 20:29:40 +01:00
b24a504d7c hook to access last parallel I/O performance measurement 2020-10-09 20:28:54 +01:00
Peter Boyle
992ef6e9fc more runtime 2020-10-08 22:19:20 -04:00
Peter Boyle
f32a320bc3 Single prec benchmark in double prec compile 2020-10-08 19:52:08 -04:00
Peter Boyle
5f0fe029d2 Improve meemory benchmarks for GPU (avoid host mem ping pong) 2020-10-08 19:51:28 -04:00
6b1486e89b fixing number of colours defaulting to 4 in most cases 2020-10-08 16:31:24 +01:00
Peter Boyle
3f9c427a3a Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2020-10-07 13:12:57 -04:00
Peter Boyle
d201277652 Expose Nc as a compile time configure option.
Remove precision option
2020-10-07 13:07:00 -04:00