Add IGLP intrinsic

3ab3cf39 · Po-Yen, Chen · 0345963e · 3ab3cf39
Commit 3ab3cf39 authored Mar 22, 2023 by Po-Yen, Chen
Hide whitespace changes
Inline Side-by-side

Showing with 2 additions and 0 deletions

include/ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_v2.hpp ...k/tensor_operation/gpu/grid/gridwise_gemm_pipeline_v2.hpp +2 -0

No files found.
--- a/include/ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_v2.hpp
+++ b/include/ck/tensor_operation/gpu/grid/gridwise_gemm_pipeline_v2.hpp
@@ -79,6 +79,8 @@ struct GridwiseGemmPipeline_v2

            do
            {
+                __builtin_amdgcn_iglp_opt(0);
+
                block_sync_lds();

                // GEMM i