CUDA backend: layernorm (#2271)

2025-12-16 01:49:05 +08:00 · 2025-06-12 07:48:32 +09:00
parent c371baf53a
commit d7e680ffe4
3 changed files with 391 additions and 2 deletions
--- a/mlx/backend/cuda/CMakeLists.txt
+++ b/mlx/backend/cuda/CMakeLists.txt
@@ -20,6 +20,7 @@ target_sources(
          ${CMAKE_CURRENT_SOURCE_DIR}/fence.cpp
          ${CMAKE_CURRENT_SOURCE_DIR}/kernel_utils.cu
          ${CMAKE_CURRENT_SOURCE_DIR}/matmul.cpp
+          ${CMAKE_CURRENT_SOURCE_DIR}/layer_norm.cu
          ${CMAKE_CURRENT_SOURCE_DIR}/logsumexp.cu
          ${CMAKE_CURRENT_SOURCE_DIR}/primitives.cu
          ${CMAKE_CURRENT_SOURCE_DIR}/random.cu