LLM inference in C/C++
C++ ggml
opencl: limit local workgroup size for GLU operation (#26383)
mgroeber91100ab9d6fed73d
diff --git a/ggml/src/ggml-opencl/ggml-opencl.cpp b/ggml/src/ggml-opencl/ggml-opencl.cppindex 388c585fa1..915c8e7b90 100644--- a/ggml/src/ggml-opencl/ggml-opencl.cpp+++ b/ggml/src/ggml-opencl/ggml-opencl.cpp@@ -24260,7 +24260,7 @@ static void ggml_cl_glu(ggml_backend_t backend, const ggml_tensor * src0, const}const size_t nrows = ggml_nrows(src0);- size_t nth = 512;+ size_t nth = backend_ctx->max_workgroup_size < 512 ? backend_ctx->max_workgroup_size : 512;size_t global_work_size[] = {nrows*nth, 1, 1};size_t local_work_size[] = {nth, 1, 1};