1
0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-13 04:37:05 +01:00

Commit Graph

  • 14591c72d6 Expression template engin Peter Boyle 2015-05-10 15:34:20 +01:00
  • 015fbee772 Expression template engin Peter Boyle 2015-05-10 15:34:20 +01:00
  • 5a7751d9df Updated TODO list Peter Boyle 2015-05-10 15:32:56 +01:00
  • 933ccfdc4f Updated TODO list Peter Boyle 2015-05-10 15:32:56 +01:00
  • 8215893152 Updated TODO list Peter Boyle 2015-05-10 15:32:56 +01:00
  • 79c51ac51f Hack; must bring norm2 into the unary operator list. ET's are still incomplete. Peter Boyle 2015-05-10 15:30:29 +01:00
  • 4e596da589 Hack; must bring norm2 into the unary operator list. ET's are still incomplete. Peter Boyle 2015-05-10 15:30:29 +01:00
  • 5fcf42cb30 Hack; must bring norm2 into the unary operator list. ET's are still incomplete. Peter Boyle 2015-05-10 15:30:29 +01:00
  • 7119bce9f3 Default to single node. Move to command line args. Peter Boyle 2015-05-10 15:27:38 +01:00
  • 41c9785f3b Default to single node. Move to command line args. Peter Boyle 2015-05-10 15:27:38 +01:00
  • e647cf0459 Default to single node. Move to command line args. Peter Boyle 2015-05-10 15:27:38 +01:00
  • cd90f55536 Single node default. Should expose this as command line args, but haven't sorted out Grid_initialize to handle this. Should put this on the TODO list. Peter Boyle 2015-05-10 15:26:06 +01:00
  • 443efd875e Single node default. Should expose this as command line args, but haven't sorted out Grid_initialize to handle this. Should put this on the TODO list. Peter Boyle 2015-05-10 15:26:06 +01:00
  • 8919bf9e0a Single node default. Should expose this as command line args, but haven't sorted out Grid_initialize to handle this. Should put this on the TODO list. Peter Boyle 2015-05-10 15:26:06 +01:00
  • dc7132af71 Small tweak to enable benchmarking to suppress gauge field bandwidth as a test. This is a short term hack while I benchmark. Peter Boyle 2015-05-10 15:25:23 +01:00
  • 02ae26d091 Small tweak to enable benchmarking to suppress gauge field bandwidth as a test. This is a short term hack while I benchmark. Peter Boyle 2015-05-10 15:25:23 +01:00
  • 133493dc79 Small tweak to enable benchmarking to suppress gauge field bandwidth as a test. This is a short term hack while I benchmark. Peter Boyle 2015-05-10 15:25:23 +01:00
  • 961fbb2718 Assertion should never hit, but did due to a bug Peter Boyle 2015-05-10 15:24:37 +01:00
  • 2ffd941d67 Assertion should never hit, but did due to a bug Peter Boyle 2015-05-10 15:24:37 +01:00
  • 58d32a4d0e Assertion should never hit, but did due to a bug Peter Boyle 2015-05-10 15:24:37 +01:00
  • 4a8fd55f52 Moving operator stuff into separate file so that we can switch on/off replacement with expression templates Peter Boyle 2015-05-10 15:23:49 +01:00
  • ca554f661b Moving operator stuff into separate file so that we can switch on/off replacement with expression templates Peter Boyle 2015-05-10 15:23:49 +01:00
  • 6bb17502f9 Moving operator stuff into separate file so that we can switch on/off replacement with expression templates Peter Boyle 2015-05-10 15:23:49 +01:00
  • e02cbaa016 Fixing breakage in the Comms non compile Peter Boyle 2015-05-10 15:23:09 +01:00
  • 29be76f958 Fixing breakage in the Comms non compile Peter Boyle 2015-05-10 15:23:09 +01:00
  • 8299bc39ea Fixing breakage in the Comms non compile Peter Boyle 2015-05-10 15:23:09 +01:00
  • 463c31ae09 Bringing expression templates for faster vector loops Peter Boyle 2015-05-10 15:22:31 +01:00
  • e3acb36de6 Bringing expression templates for faster vector loops Peter Boyle 2015-05-10 15:22:31 +01:00
  • 7f04b85368 Bringing expression templates for faster vector loops Peter Boyle 2015-05-10 15:22:31 +01:00
  • 3657f2303d ET ready benchmark with bytes counted assuming loop interchange Peter Boyle 2015-05-10 15:18:04 +01:00
  • b2e0f72a7e ET ready benchmark with bytes counted assuming loop interchange Peter Boyle 2015-05-10 15:18:04 +01:00
  • a115f3b086 ET ready benchmark with bytes counted assuming loop interchange Peter Boyle 2015-05-10 15:18:04 +01:00
  • 9ed1fb45e1 Updated todo list Peter Boyle 2015-05-10 15:13:50 +01:00
  • ebed239e49 Updated todo list Peter Boyle 2015-05-10 15:13:50 +01:00
  • 27c2d13968 Updated todo list Peter Boyle 2015-05-10 15:13:50 +01:00
  • 52403d587c Wilson perf improvements with Gauge prefetching Peter Boyle 2015-05-06 06:37:21 +01:00
  • 55ccb8ccf4 Wilson perf improvements with Gauge prefetching Peter Boyle 2015-05-06 06:37:21 +01:00
  • 5415180676 Wilson perf improvements with Gauge prefetching Peter Boyle 2015-05-06 06:37:21 +01:00
  • cdd5cdeda2 Cleaned up for Linux Peter Boyle 2015-05-05 22:09:22 +01:00
  • 35d949cc17 Cleaned up for Linux Peter Boyle 2015-05-05 22:09:22 +01:00
  • 7b0dd6c5d6 Cleaned up for Linux Peter Boyle 2015-05-05 22:09:22 +01:00
  • b9d16a7191 streaming store cases Peter Boyle 2015-05-05 18:14:09 +01:00
  • cb4b82b09f streaming store cases Peter Boyle 2015-05-05 18:14:09 +01:00
  • 07d57b6d55 Streaming store option Peter Boyle 2015-05-05 18:13:06 +01:00
  • cd990ba13d Streaming store option Peter Boyle 2015-05-05 18:13:06 +01:00
  • 5ebc7a1756 Added streaming stores Peter Boyle 2015-05-05 18:09:28 +01:00
  • 249165d1b2 Added streaming stores Peter Boyle 2015-05-05 18:09:28 +01:00
  • bf60764e4b Updated bandwidth test Peter Boyle 2015-05-05 18:08:53 +01:00
  • b720222d98 Updated bandwidth test Peter Boyle 2015-05-05 18:08:53 +01:00
  • 890b13dd5b Added a makefile Peter Boyle 2015-05-05 17:56:42 +01:00
  • 0e8415de1b Added a makefile Peter Boyle 2015-05-05 17:56:42 +01:00
  • aeda7b923d Back to vector for now; cost of init loop is clear in the a*x + y loop in memory benchmark and must move to better container class. Peter Boyle 2015-05-03 09:48:13 +01:00
  • 2b46ad38e2 Back to vector for now; cost of init loop is clear in the a*x + y loop in memory benchmark and must move to better container class. Peter Boyle 2015-05-03 09:48:13 +01:00
  • 193860dbc8 Comms and memory benchmarks added Peter Boyle 2015-05-03 09:44:47 +01:00
  • 9d93d1e6d4 Comms and memory benchmarks added Peter Boyle 2015-05-03 09:44:47 +01:00
  • 99a1ff423d Added a comms benchmark Peter Boyle 2015-05-02 23:51:43 +01:00
  • 253362f978 Added a comms benchmark Peter Boyle 2015-05-02 23:51:43 +01:00
  • f663be2a6c Added a comms benchmark Peter Boyle 2015-05-02 23:42:30 +01:00
  • ea52562527 Added a comms benchmark Peter Boyle 2015-05-02 23:42:30 +01:00
  • 4a1d4f1b3c Starting a benchmarking sub dir Peter Boyle 2015-05-02 17:52:36 +01:00
  • 6a39089a43 Starting a benchmarking sub dir Peter Boyle 2015-05-02 17:52:36 +01:00
  • 31fd146cc0 Improving the byte swap support for portability Peter Boyle 2015-05-01 10:57:33 +01:00
  • bdf18941a2 Improving the byte swap support for portability Peter Boyle 2015-05-01 10:57:33 +01:00
  • c770f96be7 Merge branch 'master' of https://github.com/paboyle/Grid Peter Boyle 2015-04-30 16:40:13 +01:00
  • d904e2b9ac Merge branch 'master' of https://github.com/paboyle/Grid Peter Boyle 2015-04-30 16:40:13 +01:00
  • a98c01c86a Integrated Lebesgue code and been playing with alternate implementations of the wilson dop without any particular success in increasing the performance. Peter Boyle 2015-04-30 16:39:06 +01:00
  • c0ead94791 Integrated Lebesgue code and been playing with alternate implementations of the wilson dop without any particular success in increasing the performance. Peter Boyle 2015-04-30 16:39:06 +01:00
  • d5b1bfb4bb Merge pull request #1 from mspraggs/patch-1 Peter Boyle 2015-04-30 09:46:48 +01:00
  • 7ac997bd58 Merge pull request #1 from mspraggs/patch-1 Peter Boyle 2015-04-30 09:46:48 +01:00
  • 6f05404cb8 Added <map> include to GridNerscIO.h mspraggs 2015-04-29 23:44:03 +01:00
  • 24fc71b2e9 Added <map> include to GridNerscIO.h mspraggs 2015-04-29 23:44:03 +01:00
  • b7090ebba4 Benchmark wilson dhop now; 14.6GF on one core, not as fast as SU(3)xSU(3) [23GF] but still not too shabby. Disassembling output shows ugly sequences in the permute sector. Could comparatively benchmark with and without the if-else structure to see how much I'm losing. Peter Boyle 2015-04-29 06:50:18 +01:00
  • d8ffa09e3b Benchmark wilson dhop now; 14.6GF on one core, not as fast as SU(3)xSU(3) [23GF] but still not too shabby. Disassembling output shows ugly sequences in the permute sector. Could comparatively benchmark with and without the if-else structure to see how much I'm losing. Peter Boyle 2015-04-29 06:50:18 +01:00
  • c72db6c6f6 Fixed the stencil sector and Wilson now agrees between stencil based implementation and the cshift based implementation. Managed to reduce the volume of code in this sector a little, but consolidation would be good, perhaps taking common logic out into simple helper functions Peter Boyle 2015-04-29 06:23:56 +01:00
  • dcc23faa4a Fixed the stencil sector and Wilson now agrees between stencil based implementation and the cshift based implementation. Managed to reduce the volume of code in this sector a little, but consolidation would be good, perhaps taking common logic out into simple helper functions Peter Boyle 2015-04-29 06:23:56 +01:00
  • 25d523c0f4 Shaken out stencil to the point where I think wilson dslash is correct. Need to audit code carefully, consolidate between stencil and cshift, and then benchmark and optimise. Peter Boyle 2015-04-28 08:11:59 +01:00
  • b0485894b3 Shaken out stencil to the point where I think wilson dslash is correct. Need to audit code carefully, consolidate between stencil and cshift, and then benchmark and optimise. Peter Boyle 2015-04-28 08:11:59 +01:00
  • f159495a9d Reworking CSHIFT and Stencil. Implementing Wilson and discovered rework is required Peter Boyle 2015-04-27 13:45:07 +01:00
  • 0b7d389258 Reworking CSHIFT and Stencil. Implementing Wilson and discovered rework is required Peter Boyle 2015-04-27 13:45:07 +01:00
  • 94f728bee4 Big updates with progress towards wilson matrix Peter Boyle 2015-04-26 15:51:09 +01:00
  • 35cfef2129 Big updates with progress towards wilson matrix Peter Boyle 2015-04-26 15:51:09 +01:00
  • 51f0da7b93 Starting the implementation of wilson; incomplete and committing non-functional code which is not yet included from elsewhere or linked to the build system. Peter Boyle 2015-04-25 14:33:02 +01:00
  • c678f2d255 Starting the implementation of wilson; incomplete and committing non-functional code which is not yet included from elsewhere or linked to the build system. Peter Boyle 2015-04-25 14:33:02 +01:00
  • 9dacdc947d Update to TODO list Peter Boyle 2015-04-25 13:04:26 +01:00
  • d5fd34b6e8 Update to TODO list Peter Boyle 2015-04-25 13:04:26 +01:00
  • c5fa18eb20 Added two spinor functionality required to support the Wilson hopping term. Peter Boyle 2015-04-25 12:54:06 +01:00
  • 2d8cf9e456 Added two spinor functionality required to support the Wilson hopping term. Peter Boyle 2015-04-25 12:54:06 +01:00
  • 8b4073d84c Dirac done ; remove from TODO Peter Boyle 2015-04-24 22:56:37 +01:00
  • dc970c6442 Dirac done ; remove from TODO Peter Boyle 2015-04-24 22:56:37 +01:00
  • 9ec3529864 Improved the gamma quite a bit. Serial rng's which are set on node zero and broadcaste Peter Boyle 2015-04-24 20:21:40 +01:00
  • fc32450360 Improved the gamma quite a bit. Serial rng's which are set on node zero and broadcaste Peter Boyle 2015-04-24 20:21:40 +01:00
  • 42eac283e2 static names and enum list Peter Boyle 2015-04-24 19:12:14 +01:00
  • 2a67214f9d static names and enum list Peter Boyle 2015-04-24 19:12:14 +01:00
  • 38598190c3 Vectors now too and right multiple of matrix with gamma Peter Boyle 2015-04-24 19:08:29 +01:00
  • 71d5927a66 Vectors now too and right multiple of matrix with gamma Peter Boyle 2015-04-24 19:08:29 +01:00
  • 2e275e1e65 Removed summation Peter Boyle 2015-04-24 18:42:44 +01:00
  • f2ac20e7ab Removed summation Peter Boyle 2015-04-24 18:42:44 +01:00
  • 80463ecaea Cleared the code out from Grid_summation to lattice/Grid_lattice_transfer.h Peter Boyle 2015-04-24 18:41:34 +01:00
  • 750dd5f5fd Cleared the code out from Grid_summation to lattice/Grid_lattice_transfer.h Peter Boyle 2015-04-24 18:41:34 +01:00
  • 128ad0999f Moved code from summation into transfer and reduction Peter Boyle 2015-04-24 18:40:44 +01:00