portelli/Grid
1
0
Fork 0
mirror of https://github.com/paboyle/Grid.git synced 2026-05-24 11:04:17 +01:00
Code Releases Activity

Lattice_reduction_gpu_cub: define GRID_REDUCTION_TIMING in header

Browse Source 
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
This commit is contained in:
Peter Boyle
2026-05-18 14:54:08 -04:00
parent e12bc7f07c
commit fca2c5dba0
1 changed files with 2 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
Grid/lattice/Lattice_reduction_gpu_cub.h
+2
View File
@@ -52,6 +52,8 @@ NAMESPACE_BEGIN(Grid);
#if defined(GRID_CUDA) || defined(GRID_HIP) #if defined(GRID_CUDA) || defined(GRID_HIP)
#define GRID_REDUCTION_TIMING
// Direct CUB reduction on the full scalar_objectD. // Direct CUB reduction on the full scalar_objectD.
// Only safe when sizeof(sobjD)*64 <= device sharedMemPerBlock. // Only safe when sizeof(sobjD)*64 <= device sharedMemPerBlock.
// Do not call directly for large composite types (e.g. LatticePropagator). // Do not call directly for large composite types (e.g. LatticePropagator).