portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-16 23:07:05 +01:00
Code Releases Activity

Hadrons: meson field threaded cache copy

Browse Source
This commit is contained in:
Antonin Portelli
2018-08-14 14:02:37 +01:00
parent 89d2fac92e
commit f4878d3a13
5 changed files with 21 additions and 18 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
lib/cshift/Cshift_common.h
View File

@ -94,7 +94,7 @@ Gather_plane_extract(const Lattice<vobj> &rhs,std::vector<typename vobj::scalar_
int n1=rhs._grid->_slice_stride[dimension];
if ( cbmask ==0x3){
parallel_for_nest2(int n=0;n<e1;n++){
parallel_for_nest(2)(int n=0;n<e1;n++){
for(int b=0;b<e2;b++){
int o = n*n1;
@ -110,7 +110,7 @@ Gather_plane_extract(const Lattice<vobj> &rhs,std::vector<typename vobj::scalar_
// Case of SIMD split AND checker dim cannot currently be hit, except in
// Test_cshift_red_black code.
std::cout << " Dense packed buffer WARNING " <<std::endl;
parallel_for_nest2(int n=0;n<e1;n++){
parallel_for_nest(2)(int n=0;n<e1;n++){
for(int b=0;b<e2;b++){
int o=n*n1;
@ -191,7 +191,7 @@ template<class vobj> void Scatter_plane_merge(Lattice<vobj> &rhs,std::vector<typ
int e2=rhs._grid->_slice_block[dimension];
if(cbmask ==0x3 ) {
parallel_for_nest2(int n=0;n<e1;n++){
parallel_for_nest(2)(int n=0;n<e1;n++){
for(int b=0;b<e2;b++){
int o = n*rhs._grid->_slice_stride[dimension];
int offset = b+n*rhs._grid->_slice_block[dimension];