diff --git a/transformer_engine/common/CMakeLists.txt b/transformer_engine/common/CMakeLists.txt
index 647d2c474d..deea5c58ad 100644
--- a/transformer_engine/common/CMakeLists.txt
+++ b/transformer_engine/common/CMakeLists.txt
@@ -135,6 +135,7 @@ set_source_files_properties(fused_softmax/scaled_masked_softmax.cu
                             COMPILE_OPTIONS "--use_fast_math")
 set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} --expt-relaxed-constexpr")
 set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -O3")
+set(CMAKE_CUDA_FLAGS "${CMAKE_CUDA_FLAGS} -diag-suppress 128")
 
 # Number of parallel build jobs
 if(ENV{MAX_JOBS})