portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2024-11-09 23:45:36 +00:00
Code Releases Activity

Merge branch 'feature/dirichlet' of https://github.com/paboyle/Grid into feature/dirichlet

Browse Source
This commit is contained in:
Peter Boyle 2023-03-28 08:35:38 -07:00
commit 5068413cdb
2 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
Grid/algorithms/iterative/ConjugateGradient.h
View File

@ -191,7 +191,7 @@ public:
std::cout << GridLogMessage << "\tAxpyNorm " << AxpyNormTimer.Elapsed() <<std::endl; std::cout << GridLogMessage << "\tAxpyNorm " << AxpyNormTimer.Elapsed() <<std::endl;
std::cout << GridLogMessage << "\tLinearComb " << LinearCombTimer.Elapsed() <<std::endl; std::cout << GridLogMessage << "\tLinearComb " << LinearCombTimer.Elapsed() <<std::endl;
std::cout << GridLogMessage << "\tMobius flop rate " << DwfFlops/ usecs<< " Gflops " <<std::endl; std::cout << GridLogDebug << "\tMobius flop rate " << DwfFlops/ usecs<< " Gflops " <<std::endl;
if (ErrorOnNoConverge) assert(true_residual / Tolerance < 10000.0); if (ErrorOnNoConverge) assert(true_residual / Tolerance < 10000.0);

4
Grid/lattice/Lattice_reduction_gpu.h
View File

@ -219,14 +219,14 @@ inline typename vobj::scalar_objectD sumD_gpu_small(const vobj *lat, Integer osi
commVector<sobj> buffer(numBlocks); commVector<sobj> buffer(numBlocks);
sobj *buffer_v = &buffer[0]; sobj *buffer_v = &buffer[0];
sobj result; sobj result;
reduceKernel<<< numBlocks, numThreads, smemSize >>>(lat, buffer_v, size); reduceKernel<<< numBlocks, numThreads, smemSize, computeStream >>>(lat, buffer_v, size);
accelerator_barrier(); accelerator_barrier();
acceleratorCopyFromDevice(buffer_v,&result,sizeof(result)); acceleratorCopyFromDevice(buffer_v,&result,sizeof(result));
#else #else
Vector<sobj> buffer(numBlocks); Vector<sobj> buffer(numBlocks);
sobj *buffer_v = &buffer[0]; sobj *buffer_v = &buffer[0];
sobj result; sobj result;
reduceKernel<<< numBlocks, numThreads, smemSize >>>(lat, buffer_v, size); reduceKernel<<< numBlocks, numThreads, smemSize, computeStream >>>(lat, buffer_v, size);
accelerator_barrier(); accelerator_barrier();
result = *buffer_v; result = *buffer_v;
#endif #endif