Compare commits

...
2 Commits
Author SHA1 Message Date
Peter Boyle 2043072d9c Merge pull request #493 from jdmaia/hip_stencil_padding
Fix alignment on StencilEntry struct for HIP
2026-09-11 17:32:30 -04:00
Julio Maia 487ec8d410 Fix alignment on StencilEntry struct for HIP
CUDA neatly aligns the StencilEntry struct to 16-bytes, which is not used on HIP.
Mapping the StencilEntry struct to 16-bytes on HIP yields around 5% better performance
on MI355X-class GPUs.
2026-09-11 10:56:02 -05:00
+1 -1
View File
@@ -82,7 +82,7 @@ void DslashLogPartial(void);
void DslashLogDirichlet(void);
struct StencilEntry {
#ifdef GRID_CUDA
#if defined(GRID_CUDA) || defined(GRID_HIP)
uint64_t _byte_offset; // 8 bytes
uint32_t _offset; // 4 bytes
#else