Commit Graph
100 Commits
Author SHA1 Message Date
Peter Boyle 63dbaeefaa Extra barrier prior to finalize just in case it fixes an issue on Tursa 2022-02-16 14:01:43 +00:00
Peter Boyle e8c187b323 SyCL happier? 2022-02-15 11:24:38 -05:00
Peter Boyle 0c1618197f Faster intranode MPI works now 2022-02-15 08:52:07 -05:00
Peter Boyle f49d5c2d22 Updated scripts for crusher 2022-02-14 17:55:16 -05:00
Peter Boyle a3b022d469 Crusher compile 2022-02-14 15:09:08 -05:00
Peter Boyle c322420580 Dont instantiate an Nc=3 and non-GP hardwired code for other implementations 2022-02-14 16:04:08 +00:00
Peter Boyle 135808dcfa Less verbose 2021-12-07 16:24:24 -05:00
Peter Boyle 7f7d06d963 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-12-07 09:06:42 -08:00
Peter Boyle 2bf3b4d576 Update to reduce memory footpring in benchmark test 2021-12-07 09:02:02 -08:00
Peter Boyle f34d34bd17 2 nodes 2021-11-22 22:27:16 -05:00
Peter Boyle e32d5141b4 Updated to make MPI reliable still gives good perf, but MPI will be slow
intranode
2021-11-22 21:46:31 -05:00
Peter Boyle 6d5277f2d7 Update to Spock 2021-11-22 20:58:02 -05:00
Peter Boyle 14d82777e0 Best modules for spock 2021-11-22 20:47:16 -05:00
Peter Boyle 2a4e739513 Enable XGMI copy (need to rename nvlink to cover NVLINK/XGMI/XeLink) 2021-11-22 20:46:09 -05:00
Peter Boyle 8079dc2a14 Cray MPI not working right yet 2021-11-22 20:45:44 -05:00
Peter Boyle 6ceb556684 Intranode asynch hipMemCopy 2021-11-22 20:45:12 -05:00
Peter Boyle 76cde73705 HIP improvements on messaging and intranode hipMemCopyAsynch 2021-11-22 20:44:39 -05:00
Peter Boyle fe9edf8526 Merge branch 'develop' of https://www.github.com/paboyle/Grid into develop 2021-10-29 02:03:27 +01:00
Peter Boyle 44204c7e06 Extra code 2021-10-29 02:02:56 +01:00
Peter Boyle 85f750d753 Merge branch 'develop' of https://www.github.com/paboyle/Grid into develop 2021-10-27 00:28:05 +01:00
Peter Boyle a4ce6e42c7 Warning free compile on make all and make tests under nvcc 2021-10-27 00:27:03 +01:00
Peter Boyle 5398b7e7e3 Max 128 size 2021-10-26 09:16:29 -07:00
Peter Boyle ba7e371b90 Warning free compile on Tursa.
Hopefully got all reqd virtual dtors
2021-10-21 19:56:52 +01:00
Peter Boyle 749b8022a4 Linear operator and SparseMatrix virtual destructors 2021-10-15 20:47:18 +01:00
Peter Boyle 7e0057d2c4 Merge branch 'develop' of https://www.github.com/paboyle/Grid into develop 2021-10-15 20:46:51 +01:00
Peter Boyle cfe9e870d3 Stream 2021-10-15 20:46:44 +01:00
Peter Boyle 16c2a99965 Overlap cudamemcpy - didn't set up stream right 2021-10-11 13:31:26 -07:00
Peter Boyle cda915a345 Better options 2021-10-07 20:29:09 +01:00
Peter Boyle a8eda8f6da Summit scripts 2021-10-05 21:22:10 -04:00
Peter Boyle 9b1a0653cf Summit results 2021-10-05 21:22:01 -04:00
Peter Boyle 7cb1ff7395 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-10-05 20:13:42 -04:00
Peter Boyle ab6ea29913 Print removal 2021-10-05 20:13:25 -04:00
Peter Boyle 4016e705fc Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-10-05 14:56:57 -04:00
Peter Boyle 2f4e85e5d6 Summit set up 2021-10-05 14:56:17 -04:00
Peter Boyle 8ed0b57b09 Memory verbose and tracking, shrink default cache
Print PCI device IDs on node 0
2021-10-05 11:41:03 -04:00
Peter Boyle 67e08aa952 New file not run yet 2021-09-23 23:39:55 +02:00
Peter Boyle cffc736bb3 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-09-22 06:03:06 -07:00
Peter Boyle c0d56a1c04 Perlmutter tune up 2021-09-22 06:02:34 -07:00
Peter Boyle 3206f69478 SYCL happy 2021-09-21 18:01:35 -07:00
Peter Boyle b2ccaad761 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-09-21 12:18:05 -07:00
Peter Boyle 8eb1232683 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-09-21 09:25:07 -07:00
Peter Boyle c6ce3ad03b Some properties 2021-09-21 09:20:21 -07:00
Peter Boyle b3b033d343 Clean 2021-09-21 09:18:54 -07:00
Peter Boyle ca9816bfbb Typo 2021-09-21 04:12:04 +02:00
Peter Boyle 814d5abc7e Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-09-21 04:05:51 +02:00
Peter Boyle a29122e2bf Rebench 2021-09-21 04:05:04 +02:00
Peter Boyle e188c0512e Udpdate 2021-09-21 01:04:30 +02:00
Peter Boyle 1fb6aaf150 Device 2 Device with cudaMemcpy 2021-09-21 01:03:07 +02:00
Peter Boyle 894654f7ef Simplificatoin, always gather faces 2021-09-21 01:02:34 +02:00
Peter Boyle 109507888b Option to force use of MPI over Nvlink 2021-09-21 00:53:25 +02:00
Peter Boyle 68650b61fe Options controlling behaviour 2021-09-21 00:51:01 +02:00
Peter Boyle 8bd70ad8b5 Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-09-16 10:22:38 -07:00
Peter Boyle 145acf2919 Perf results 2021-09-16 01:06:28 +01:00
Peter Boyle cc4a27b9e6 Scripts and performance 2021-09-16 00:15:35 +01:00
Peter Boyle b4690e6091 Adding build basics for different systems 2021-09-16 00:00:38 +01:00
Peter Boyle 9d2238148c Merge branch 'develop' of https://www.github.com/paboyle/Grid into develop 2021-09-15 19:25:57 +01:00
Peter Boyle c15493218d Two extra routines to break out SchurRedBlack on many RHS into stages to allow efficient deflation & split grid
Split grid solver still to do.
2021-09-15 19:24:39 +01:00
Peter Boyle c7baeb5bae Merge branch 'develop' of https://github.com/paboyle/Grid into develop 2021-09-14 08:31:11 -07:00
Peter Boyle 402d80e197 Merge branch 'develop' of https://www.github.com/paboyle/Grid into develop 2021-09-14 16:16:06 +01:00
Peter Boyle 86e33c8ab2 Significant GPU perf speed up finished 2021-09-14 16:14:23 +01:00
Peter Boyle 5dae6a6dac Deprecate half prec comms 2021-09-14 15:06:59 +01:00
Peter Boyle 361bb8a101 Remove half prec comms 2021-09-14 15:06:29 +01:00
Peter Boyle 7efdb3cd2b Remove half prec comms 2021-09-14 15:06:06 +01:00
Peter Boyle 65ef4ec29f Move tables to device memory 2021-09-14 15:05:01 +01:00
Peter Boyle d5835c0222 Switch to coalesced stencil face gather 2021-09-14 15:04:14 +01:00
Peter Boyle a7b943b33e Remove half prec comms 2021-09-14 05:05:33 +01:00
Peter Boyle 7440cde92f No half prec comms; coalesced access on GPU 2021-09-14 05:04:56 +01:00
Peter Boyle 0fc662bb24 Dirac cuda 11.4 happy ; force host for functions accessing mult table
ET runs these on host BEFORE lodging result in AST for kernel
2021-09-14 05:00:44 +01:00
Peter Boyle 8195890640 Force MPI over NVLINK 2021-09-14 05:00:17 +01:00
Peter Boyle 4c88104a73 Fix compile warns 2021-09-11 23:08:05 +01:00
Peter Boyle 73b944c152 Drop half prec comms for now. 2021-09-11 23:07:18 +01:00
Peter Boyle d1b0b7f5c6 Half prec comms dropping 2021-09-11 23:05:40 +01:00
Peter Boyle 381d8797d0 Drop half prec comms for now 2021-09-11 23:05:02 +01:00
Peter Boyle b06526bc1e Comment update 2021-08-30 21:15:39 -04:00
Peter Boyle 3044419111 Some sample code 2021-08-30 20:32:11 -04:00
Peter Boyle bcfa9cf068 Improvement of output 2021-08-28 08:08:15 -07:00
Peter Boyle 114920b8de Some example clean up 2021-08-25 12:24:17 +01:00
Peter Boyle 0d588b95f4 Bug fix to Example_Laplacian test 2021-08-23 23:14:26 +01:00
Peter Boyle 5b3c530aa7 Return value 2021-08-23 15:30:45 +01:00
Peter Boyle c6a5499c8b Fail on non-apple 2021-08-22 18:40:55 +01:00
Peter Boyle ec9c3fe77a Remove the file 2021-08-22 18:28:39 +01:00
Peter Boyle 6135ad530e Extra examples / solutions 2021-08-22 18:25:07 +01:00
Peter Boyle 40098424c7 Examples 2021-08-22 14:17:12 +01:00
Peter Boyle 7163b31a26 Examples 2021-08-20 01:15:23 +01:00
Peter Boyle ffbdd91e0e Apple happiness 2021-08-20 01:15:00 +01:00
Peter Boyle 5d29e175d8 Typo fix 2021-08-10 18:25:43 +01:00
Peter Boyle 417dbfa257 Fix 2021-08-10 08:55:35 -07:00
Peter Boyle 75030637cc Improved comms benchmark, same as benchmark_comms_host_device 2021-08-10 05:16:30 -07:00
Peter Boyle fe5aaf7677 Make comms benchmark same as Benchmark_comms_host_device 2021-08-09 04:06:30 -07:00
Peter Boyle 80ac2a73ca Check is wrong (HtoD / DtoH) 2021-08-05 18:33:20 -04:00
Peter BoyleandChristoph Lehner b5aeae526f Make Cshift fields static to avoid repeated reallocaate overhead 2021-05-28 16:33:08 +02:00
Peter Boyle 8cd4263974 Tests compile 2021-04-25 22:20:37 -04:00
Peter Boyle d45c868656 Change interface 2021-04-25 10:53:34 -04:00
Peter Boyle 955a8113de Expose label only to reduce number of parameters 2021-04-25 10:36:38 -04:00
Peter Boyle dbe210dd53 Open the ens_id 2021-04-25 10:25:59 -04:00
Peter Boyle 86e11743ca set twists 2021-04-20 10:19:11 -04:00
Peter Boyle a7fb25adf6 Make Cshift fields static to avoid repeated reallocaate overhead 2021-03-29 21:44:14 +02:00
Peter Boyle e947992957 Improved force terms 2021-03-29 20:04:06 +02:00
Peter Boyle bb89a82a07 Staggered coalseced read 2021-03-29 20:01:15 +02:00
Peter Boyle 8bdadbadac Cold start 2021-03-18 15:41:14 -04:00