1
0
mirror of https://github.com/paboyle/Grid.git synced 2026-04-20 02:31:01 +01:00
Commit Graph

6992 Commits

Author SHA1 Message Date
Peter Boyle 806b02bddf Simplify dead code 2022-08-04 15:23:13 -04:00
Peter Boyle de40395773 More timing. Think I should start to use nvtx and rocmtx ?? 2022-08-04 13:37:16 -04:00
Peter Boyle 7ba4788715 Fix 2022-08-04 13:36:44 -04:00
Peter Boyle 06d9ce1a02 Synch ranks on node here for GPU - GPU memcopy 2022-08-04 13:35:56 -04:00
Peter Boyle 75bb6b2b40 Move barrier into the StencilSend begin routine 2022-08-04 13:35:26 -04:00
Peter Boyle 74f10c2dc0 Move barrier into Stencil Send 2022-08-04 13:34:11 -04:00
Peter Boyle 188d2c7a4d PVC default, ignore ATS 2022-08-02 08:38:53 -07:00
Peter Boyle 17d7177105 Files for SYCL 2022-08-02 08:33:39 -07:00
Peter Boyle bb0a0da47a inon blocking caution due to SYCL 2022-08-02 08:09:43 -07:00
Peter Boyle 84110166e4 Fix the fence 2022-08-02 08:00:43 -07:00
Peter Boyle d32b923b6c Fencing on a stream in SYCL is needed. Didn't know that ... gulp 2022-08-02 07:58:04 -07:00
Peter Boyle a93d5459d4 Better mpi request completion 2022-07-28 12:18:35 -04:00
Peter Boyle 9c21add0c6 High res timer replaces getttimeofday 2022-07-28 12:14:03 -04:00
Peter Boyle 639aab6563 High res timer instead of gettimeofday 2022-07-28 12:13:35 -04:00
Peter Boyle 8137cc7049 Allways concurrent comms 2022-07-28 12:01:51 -04:00
Peter Boyle 60e63dca1d Add memory logging channel 2022-07-28 11:39:15 -04:00
Peter Boyle 486409574e Expanded cach to avoid any allocs in HMC 2022-07-28 11:38:34 -04:00
Peter Boyle a913b8be12 Dslash self timing. Might want to not have this 2022-07-28 11:37:55 -04:00
Peter Boyle 2239751850 Better logging 2022-07-28 11:37:36 -04:00
Peter Boyle 9b20f1449c Better timing 2022-07-28 11:37:12 -04:00
Peter Boyle b99453083d Updated timing 2022-07-28 11:37:02 -04:00
Peter Boyle 2ab1af5754 Ensure no synchronize and not optoin dependent 2022-07-19 09:51:06 -07:00
Peter Boyle 5f8892bf03 Mistake pointed out by Camilo 2022-07-19 09:31:51 -07:00
Peter Boyle f14e7e51e7 Grid accelerator 2022-07-12 10:56:22 -07:00
Peter Boyle 943fbb914d Merge branch 'feature/dirichlet' of https://github.com/paboyle/Grid into feature/dirichlet 2022-07-11 13:48:42 -04:00
Peter Boyle ca4603580d Verbose 2022-07-11 13:48:35 -04:00
Peter Boyle f73db8f1f3 Synch clocks 2022-07-11 13:47:39 -04:00
Peter Boyle f7217d12d2 World barrier for clock synch 2022-07-11 13:45:31 -04:00
Peter Boyle fab50c57d9 More loggin 2022-07-11 18:42:27 +01:00
Peter Boyle 3440534fbf MixedPrec support 2022-07-10 21:35:18 +01:00
Peter Boyle 177b1a7ec6 Mixed prec 2022-07-10 21:34:10 +01:00
Peter Boyle 58182fe345 Different approach to default dirichlet params 2022-07-10 21:32:58 +01:00
Peter Boyle 1f907d330d Different default params for dirichlet 2022-07-10 21:31:48 +01:00
Peter Boyle b0fe664e9d Better force log info 2022-07-10 21:31:25 +01:00
Peter Boyle c0f8482402 Remove SSC marks 2022-07-07 17:49:36 +01:00
Peter Boyle 3544965f54 Stream doesn't work 2022-07-07 17:49:20 +01:00
Christopher Kelly 33e4a0caee Imported changes from feature/gparity_HMC branch:
Rework of WilsonFlow class
		Fixed logic error in smear method where the step index was initialized to 1 rather than 0, resulting in the logged output value of tau being too large by epsilon
		Previously smear_adaptive would maintain the current value of tau as a class member variable whereas smear would compute it separately; now both methods maintain the current value internally and it is updated by the evolve_step routines. Both evolve methods are now const.
		smear_adaptive now also maintains the current value of epsilon internally, allowing it to be a const method and also allowing the same class instance to be reused without needing to be reset
		Replaced the fixed evaluation of the plaquette energy density and plaquette topological charge during the smearing with a highly flexible general strategy where the user can add arbitrary measurements as functional objects that are evaluated at an arbitrary frequency
	        By default the same plaquette-based measurements are performed, but additional example functions are provided where the smearing is performed with different choices of measurement that are returned as an array for further processing
		Added a method to compute the energy density using the Cloverleaf approach which has smaller discretization errors
	Added a new tensor utility operation, copyLane, which allows for the copying of a single SIMD lane between two instances of the same tensor type but potentially different precisions
	To LocalCoherenceLanczos, added the option to compute the high/low eval of the fine operator on every restart to aid in tuning the Chebyshev
	Added Test_field_array_io which demonstrates and tests a single-file write of an arbitrary array of fields
	Added Test_evec_compression which generates evecs using Lanczos and attempts to compress them using the local coherence technique
	Added Test_compressed_lanczos_gparity which demonstrates the local coherence Lanczos for G-parity BCs
	Added HMC main programs for the 40ID and 48ID G-parity lattices
2022-07-01 14:12:12 -04:00
Peter Boyle 1f903d9296 Merge branch 'feature/dirichlet' into feature/dirichlet-gparity 2022-07-01 12:12:50 -04:00
Peter Boyle 4df1e0987f Merge branch 'feature/dirichlet-gparity' of https://github.com/paboyle/Grid into feature/dirichlet-gparity 2022-07-01 09:55:43 -04:00
Peter Boyle 588c2f3cb1 Faster axpy_norm and innerProduct 2022-07-01 09:44:58 -04:00
Peter Boyle bd99fd608c Introduce a non-default stream for compute operatoins 2022-07-01 09:42:53 -04:00
Peter Boyle 57b442d0de Log memory operations 2022-07-01 09:42:17 -04:00
Peter Boyle 751a4562d7 Timing improvement 2022-07-01 09:41:43 -04:00
Peter Boyle ca66301dee Remove debug 2022-06-30 14:53:12 -04:00
Peter Boyle 808bb59206 Mixed prec DD-RHMC 2022-06-30 13:50:09 -04:00
Peter Boyle 4b7f51d19d Create a new RNG file 2022-06-30 13:49:50 -04:00
Peter Boyle d03152fac4 New file under debug 2022-06-30 13:49:35 -04:00
Peter Boyle 137f190258 Dirichlet implementation 2022-06-30 13:45:07 -04:00
Peter Boyle 53d01312b3 Rough flop counting, need to add M5D, M5Ddag, MooeeInv flops 2022-06-30 13:44:09 -04:00
Peter Boyle 220050822a Speed up M5D and M5Ddag 2022-06-30 13:43:27 -04:00