portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2024-11-10 15:55:37 +00:00
Code Releases Activity
96f1d1b828
Grid/lib/qcd/action/fermion/WilsonKernels.h

103 lines
4.4 KiB
C
Raw Normal View History

Global edit adding copyright and license info to every source file.
2016-01-02 14:51:32 +00:00
/*************************************************************************************
Grid physics library, www.github.com/paboyle/Grid
Source file: ./lib/qcd/action/fermion/WilsonKernels.h
Copyright (C) 2015
Author: Peter Boyle <pabobyle@ph.ed.ac.uk>
Author: Peter Boyle <paboyle@ph.ed.ac.uk>
Author: paboyle <paboyle@ph.ed.ac.uk>
This program is free software; you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation; either version 2 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
You should have received a copy of the GNU General Public License along
with this program; if not, write to the Free Software Foundation, Inc.,
51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA.
See the full license in the file "LICENSE" in the top level distribution directory
*************************************************************************************/
/* END LEGAL */
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
#ifndef GRID_QCD_DHOP_H
#define GRID_QCD_DHOP_H
namespace Grid {
namespace QCD {
////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Rework/global edit to enforce type templating of fermion operators. Allows multi-precision work and paves the way for alternate BC's and such like allowing for example G-parity which is important for K pipi programme. In particular, can drive an extra flavour index into the fermion fields using template types.
2015-08-10 20:47:44 +01:00
// Helper routines that implement Wilson stencil for a single site.
* Finished the template/policy style introduction of gparity, except the gparity force terms. So valence sector looks ok. FermionOperatorImpl.h provides the policy classes. Expect HMC will introduce a smearing policy and a fermion representation change policy template param. Will also probably need multi-precision work. * HMC is running even-odd and non-checkerboarded (checked 4^4 wilson fermion/wilson gauge). There appears to be a bug in the multi-level integrator -- <e-dH> passes with single level but not with multi-level. In any case there looks to be quite a bit to clean up. This is the "const det" style implementation that is not appropriate yet for clover since it assumes that Mee is indept of the gauge fields. Easily fixed in future.
2015-08-15 23:25:49 +01:00
// Common to both the WilsonFermion and WilsonFermion5D
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
////////////////////////////////////////////////////////////////////////////////////////////////////////////////
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
class WilsonKernelsStatic {
public:
// S-direction is INNERMOST and takes no part in the parity.
static int AsmOpt; // these are a temporary hack
static int HandOpt; // these are a temporary hack
};
Rework/global edit to enforce type templating of fermion operators. Allows multi-precision work and paves the way for alternate BC's and such like allowing for example G-parity which is important for K pipi programme. In particular, can drive an extra flavour index into the fermion fields using template types.
2015-08-10 20:47:44 +01:00
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
template<class Impl> class WilsonKernels : public FermionOperator<Impl> , public WilsonKernelsStatic {
Rework/global edit to enforce type templating of fermion operators. Allows multi-precision work and paves the way for alternate BC's and such like allowing for example G-parity which is important for K pipi programme. In particular, can drive an extra flavour index into the fermion fields using template types.
2015-08-10 20:47:44 +01:00
public:
* Finished the template/policy style introduction of gparity, except the gparity force terms. So valence sector looks ok. FermionOperatorImpl.h provides the policy classes. Expect HMC will introduce a smearing policy and a fermion representation change policy template param. Will also probably need multi-precision work. * HMC is running even-odd and non-checkerboarded (checked 4^4 wilson fermion/wilson gauge). There appears to be a bug in the multi-level integrator -- <e-dH> passes with single level but not with multi-level. In any case there looks to be quite a bit to clean up. This is the "const det" style implementation that is not appropriate yet for clover since it assumes that Mee is indept of the gauge fields. Easily fixed in future.
2015-08-15 23:25:49 +01:00
INHERIT_IMPL_TYPES(Impl);
typedef FermionOperator<Impl> Base;
Rework/global edit to enforce type templating of fermion operators. Allows multi-precision work and paves the way for alternate BC's and such like allowing for example G-parity which is important for K pipi programme. In particular, can drive an extra flavour index into the fermion fields using template types.
2015-08-10 20:47:44 +01:00
public:
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
Enable reordering of the loops in the assembler for cache friendly. This gets in the way of L2 prefetching however. Do next next link in stencil prefetching.
2016-06-19 19:45:58 +01:00
void DiracOptDhopSite(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
* Finished the template/policy style introduction of gparity, except the gparity force terms. So valence sector looks ok. FermionOperatorImpl.h provides the policy classes. Expect HMC will introduce a smearing policy and a fermion representation change policy template param. Will also probably need multi-precision work. * HMC is running even-odd and non-checkerboarded (checked 4^4 wilson fermion/wilson gauge). There appears to be a bug in the multi-level integrator -- <e-dH> passes with single level but not with multi-level. In any case there looks to be quite a bit to clean up. This is the "const det" style implementation that is not appropriate yet for clover since it assumes that Mee is indept of the gauge fields. Easily fixed in future.
2015-08-15 23:25:49 +01:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
Improvements to the assembler interface that let us move chunks of the site and s loop into the kernels. This will save on function call overhead and guarantee L2 prefetching strategy is right since OMP can't distribute the sub-chunks of work.
2016-06-09 09:12:36 +01:00
int sF, int sU,int Ls, int Ns, const FermionField &in, FermionField &out);
big commit fixing nocompiles in defective C++11 compilers (gcc, icpc). stared getting to near the bleeding edge I guess
2015-06-30 15:03:11 +01:00
Enable reordering of the loops in the assembler for cache friendly. This gets in the way of L2 prefetching however. Do next next link in stencil prefetching.
2016-06-19 19:45:58 +01:00
void DiracOptDhopSiteDag(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
* Finished the template/policy style introduction of gparity, except the gparity force terms. So valence sector looks ok. FermionOperatorImpl.h provides the policy classes. Expect HMC will introduce a smearing policy and a fermion representation change policy template param. Will also probably need multi-precision work. * HMC is running even-odd and non-checkerboarded (checked 4^4 wilson fermion/wilson gauge). There appears to be a bug in the multi-level integrator -- <e-dH> passes with single level but not with multi-level. In any case there looks to be quite a bit to clean up. This is the "const det" style implementation that is not appropriate yet for clover since it assumes that Mee is indept of the gauge fields. Easily fixed in future.
2015-08-15 23:25:49 +01:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
Improvements to the assembler interface that let us move chunks of the site and s loop into the kernels. This will save on function call overhead and guarantee L2 prefetching strategy is right since OMP can't distribute the sub-chunks of work.
2016-06-09 09:12:36 +01:00
int sF,int sU,int Ls, int Ns, const FermionField &in,FermionField &out);
* Finished the template/policy style introduction of gparity, except the gparity force terms. So valence sector looks ok. FermionOperatorImpl.h provides the policy classes. Expect HMC will introduce a smearing policy and a fermion representation change policy template param. Will also probably need multi-precision work. * HMC is running even-odd and non-checkerboarded (checked 4^4 wilson fermion/wilson gauge). There appears to be a bug in the multi-level integrator -- <e-dH> passes with single level but not with multi-level. In any case there looks to be quite a bit to clean up. This is the "const det" style implementation that is not appropriate yet for clover since it assumes that Mee is indept of the gauge fields. Easily fixed in future.
2015-08-15 23:25:49 +01:00
Stencil class name global change to StencilImpl typedef
2015-11-06 11:30:17 +00:00
void DiracOptDhopDir(StencilImpl &st,DoubledGaugeField &U,
* Finished the template/policy style introduction of gparity, except the gparity force terms. So valence sector looks ok. FermionOperatorImpl.h provides the policy classes. Expect HMC will introduce a smearing policy and a fermion representation change policy template param. Will also probably need multi-precision work. * HMC is running even-odd and non-checkerboarded (checked 4^4 wilson fermion/wilson gauge). There appears to be a bug in the multi-level integrator -- <e-dH> passes with single level but not with multi-level. In any case there looks to be quite a bit to clean up. This is the "const det" style implementation that is not appropriate yet for clover since it assumes that Mee is indept of the gauge fields. Easily fixed in future.
2015-08-15 23:25:49 +01:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
int sF,int sU,const FermionField &in, FermionField &out,int dirdisp,int gamma);
Fix a regression failure on Mobius; chroma regression added
2015-12-10 22:55:00 +00:00
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
private:
// Specialised variants
Enable reordering of the loops in the assembler for cache friendly. This gets in the way of L2 prefetching however. Do next next link in stencil prefetching.
2016-06-19 19:45:58 +01:00
void DiracOptGenericDhopSite(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
Improvements to the assembler interface that let us move chunks of the site and s loop into the kernels. This will save on function call overhead and guarantee L2 prefetching strategy is right since OMP can't distribute the sub-chunks of work.
2016-06-09 09:12:36 +01:00
int sF,int sU, const FermionField &in, FermionField &out);
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
Enable reordering of the loops in the assembler for cache friendly. This gets in the way of L2 prefetching however. Do next next link in stencil prefetching.
2016-06-19 19:45:58 +01:00
void DiracOptGenericDhopSiteDag(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
int sF,int sU,const FermionField &in,FermionField &out);
Enable reordering of the loops in the assembler for cache friendly. This gets in the way of L2 prefetching however. Do next next link in stencil prefetching.
2016-06-19 19:45:58 +01:00
void DiracOptAsmDhopSite(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
Changes for AVX512 assembler
2015-11-06 11:45:51 +00:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
Improvements to the assembler interface that let us move chunks of the site and s loop into the kernels. This will save on function call overhead and guarantee L2 prefetching strategy is right since OMP can't distribute the sub-chunks of work.
2016-06-09 09:12:36 +01:00
int sF,int sU,int Ls, int Ns, const FermionField &in, FermionField &out);
Fix a regression failure on Mobius; chroma regression added
2015-12-10 22:55:00 +00:00
Feature for z-Mobius prep
2016-08-15 22:31:29 +01:00
void DiracOptAsmDhopSiteDag(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
int sF,int sU,int Ls, int Ns, const FermionField &in, FermionField &out);
Large change with KNL preparation
2016-06-03 11:24:26 +01:00
Enable reordering of the loops in the assembler for cache friendly. This gets in the way of L2 prefetching however. Do next next link in stencil prefetching.
2016-06-19 19:45:58 +01:00
void DiracOptHandDhopSite(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
Pushed the overlap comms tweaks
2016-01-11 14:34:22 +00:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
Avx512 changes for assembler kernels
2016-03-27 05:24:07 +01:00
int sF,int sU,const FermionField &in, FermionField &out);
Pushed the overlap comms tweaks
2016-01-11 14:34:22 +00:00
Enable reordering of the loops in the assembler for cache friendly. This gets in the way of L2 prefetching however. Do next next link in stencil prefetching.
2016-06-19 19:45:58 +01:00
void DiracOptHandDhopSiteDag(StencilImpl &st,LebesgueOrder &lo,DoubledGaugeField &U,
Pushed the overlap comms tweaks
2016-01-11 14:34:22 +00:00
std::vector<SiteHalfSpinor,alignedAllocator<SiteHalfSpinor> > &buf,
Avx512 changes for assembler kernels
2016-03-27 05:24:07 +01:00
int sF,int sU,const FermionField &in, FermionField &out);
Compiling updates for KNL
2016-06-03 11:47:54 +01:00
public:
Rework/global edit to enforce type templating of fermion operators. Allows multi-precision work and paves the way for alternate BC's and such like allowing for example G-parity which is important for K pipi programme. In particular, can drive an extra flavour index into the fermion fields using template types.
2015-08-10 20:47:44 +01:00
Fix a regression failure on Mobius; chroma regression added
2015-12-10 22:55:00 +00:00
WilsonKernels(const ImplParams &p= ImplParams());
Rework/global edit to enforce type templating of fermion operators. Allows multi-precision work and paves the way for alternate BC's and such like allowing for example G-parity which is important for K pipi programme. In particular, can drive an extra flavour index into the fermion fields using template types.
2015-08-10 20:47:44 +01:00
};
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
}
}
#endif
Copy Permalink