Azusa Yamaguchi
8f87950dc1
FIx miistake
2015-06-01 12:26:20 +01:00
Peter Boyle
66d997e031
Large scale change to support 5d fermion formulations.
...
Have 5d replicated wilson with 4d gauge working and matrix regressing
to Ls copies of wilson.
2015-05-31 15:09:02 +01:00
Peter Boyle
20100d0a40
Hand unrolled version of dslash in a separate class.
...
Useful to compare; raises Intel compiler from 9GFlop/s to 17.5 Gflops.
on ivybridge core. Raises Clang form 14.5 to 17.5
2015-05-26 19:54:03 +01:00
Peter Boyle
f6cade41b4
Better checkerboard tracking.
2015-05-25 13:45:08 +01:00
Peter Boyle
3954792f37
Better pragma use
2015-05-23 09:32:37 +01:00
Peter Boyle
e0cc5ba920
Streaming store option ifdef
2015-05-21 06:47:05 +01:00
Peter Boyle
1b9ecbac3b
Compile time select if we do the streaming store copy. Relies on Clang++ eliminating object copies,
...
and other compliers do not necessarily cope.
2015-05-21 06:39:00 +01:00
Peter Boyle
a6e1ea216d
Got unpreconditioned conjugate gradient to run and converge on a random (uniform random,
...
not even SU(3) for now) gauge field. Convergence history is correctly indepdendent of decomposition
on 1,2,4,8,16 mpi tasks.
Found a couple of simd bugs which required fixed and enhanced the Grid_simd.cc test suite.
Implemented the Mdag, M, MdagM, Meooe Mooee schur type stuff in the wilson dop.
2015-05-19 13:57:35 +01:00
Peter Boyle
53260e7a39
Typoo xifed
2015-05-16 05:49:32 +01:00
Peter Boyle
a900790b44
Update Grid_lattice_trace.h
2015-05-16 04:40:28 +01:00
Peter Boyle
5f8b82b90c
Optimisation and syntax pretty
2015-05-16 04:36:22 +01:00
Peter Boyle
afda459886
strong inline
2015-05-16 04:33:10 +01:00
Peter Boyle
e8efa6320e
Parallel for replace
2015-05-15 11:48:04 +01:00
Peter Boyle
074430af0d
Formatting change
2015-05-15 11:38:54 +01:00
Peter Boyle
873110d482
Filed bug report Bug 66153 on GCC-5.
...
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=66153
2015-05-15 11:38:04 +01:00
Peter Boyle
4cee0e8653
Silly formatting change
2015-05-15 11:37:07 +01:00
Peter Boyle
d388b831b4
cout IO for all types
2015-05-13 09:24:10 +01:00
Peter Boyle
52174da232
Enhanced SIMD interfacing
2015-05-12 20:41:44 +01:00
Peter Boyle
65c91eae64
Threading support rework.
...
Placed parallel pragmas as macros; implemented deterministic thread reduction in style of
BFM.
2015-05-12 07:51:41 +01:00
Peter Boyle
8b765be2b1
Moving some things around for pretty
2015-05-11 19:09:49 +01:00
Peter Boyle
242e447bc5
Lots of changes required to compile for MIC under ICPC
2015-05-10 23:29:21 +01:00
Peter Boyle
c946e77143
Expression template hack
2015-05-10 15:35:30 +01:00
Peter Boyle
015fbee772
Expression template engin
2015-05-10 15:34:20 +01:00
Peter Boyle
58d32a4d0e
Assertion should never hit, but did due to a bug
2015-05-10 15:24:37 +01:00
Peter Boyle
6bb17502f9
Moving operator stuff into separate file so that we can switch on/off replacement with
...
expression templates
2015-05-10 15:23:49 +01:00
Peter Boyle
cb4b82b09f
streaming store cases
2015-05-05 18:14:09 +01:00
Peter Boyle
9d93d1e6d4
Comms and memory benchmarks added
2015-05-03 09:44:47 +01:00
Peter Boyle
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.
2015-04-28 08:11:59 +01:00
Peter Boyle
35cfef2129
Big updates with progress towards wilson matrix
2015-04-26 15:51:09 +01:00
Peter Boyle
fc32450360
Improved the gamma quite a bit.
...
Serial rng's which are set on node zero and broadcaste
2015-04-24 20:21:40 +01:00
Peter Boyle
74432432b6
Moved code from summation into transfer and reduction
2015-04-24 18:40:44 +01:00
Peter Boyle
62e8d2d127
Slice summation working. May move this into lattice/Grid_lattice_reduction however
2015-04-23 15:13:00 +01:00
Peter Boyle
1851327d19
Got the NERSC IO working and fixed a bug in cshift.
2015-04-22 22:46:48 +01:00
Peter Boyle
a5b0c492d7
Rework of RNG to use C++11 random. Should work correctly maintaining parallel RNG across
...
a machine. If a "fixedSeed" is used, randoms should be reproducible across different machine
decomposition since the generators are physically indexed and assigned in lexico ordering.
2015-04-19 14:55:58 +01:00
Peter Boyle
f64d39ab57
Split all OMP directives into lattice subdir for easy maintainance of
...
parallelism and future OMP 4.0 offload.
2015-04-18 22:17:01 +01:00
Peter Boyle
aee6669d0b
Build reorg with which I am a bit happier
2015-04-18 21:22:50 +01:00
