portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2025-06-12 20:27:06 +01:00
Code Releases Activity

Sizable improvement in multigrid for unsquared.

Browse Source 
6000 matmuls CG unprec
2000 matmuls CG prec (4000 eo muls)
1050 matmuls PGCR on 16^3 x 32 x 8 m=.01

Substantial effort on timing and logging infrastructure
This commit is contained in:
Peter Boyle
2015-07-24 01:31:13 +09:00
parent 11c99d5e66
commit d1afebf71e
67 changed files with 945 additions and 753 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
lib/tensors/Tensor_extract_merge.h
View File

@ -57,11 +57,11 @@ inline void extract(typename std::enable_if<!isGridTensor<vsimd>::value, const v
extracted[i]=buf[i*s];
for(int ii=1;ii<s;ii++){
if ( buf[i*s]!=buf[i*s+ii] ){
std::cout << " SIMD extract failure splat = "<<s<<" ii "<<ii<<" " <<Nextr<<" "<< Nsimd<<" "<<std::endl;
std::cout<<GridLogMessage << " SIMD extract failure splat = "<<s<<" ii "<<ii<<" " <<Nextr<<" "<< Nsimd<<" "<<std::endl;
for(int vv=0;vv<Nsimd;vv++) {
std::cout<< buf[vv]<<" ";
std::cout<<GridLogMessage<< buf[vv]<<" ";
}
std::cout<<std::endl;
std::cout<<GridLogMessage<<std::endl;
assert(0);
}
assert(buf[i*s]==buf[i*s+ii]);