Inital code drop

Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com> Signed-off-by: Przemek Tredak <ptredak@nvidia.com>

Inital code drop
Co-authored-by: Kirthi Shankar Sivamani <ksivamani@nvidia.com> Signed-off-by: Przemek Tredak <ptredak@nvidia.com>
996ea169 · Przemek Tredak · 996ea169 · 996ea169 · 996ea169 · 996ea169
Commit 996ea169 authored Sep 27, 2022 by Przemek Tredak
20 changed files
--- a/transformer_engine/common/gemm/cublaslt_gemm.cu
+++ b/transformer_engine/common/gemm/cublaslt_gemm.cu
--- a/transformer_engine/common/include/transformer_engine/activation.h
+++ b/transformer_engine/common/include/transformer_engine/activation.h
--- a/transformer_engine/common/include/transformer_engine/cast.h
+++ b/transformer_engine/common/include/transformer_engine/cast.h
--- a/transformer_engine/common/include/transformer_engine/gemm.h
+++ b/transformer_engine/common/include/transformer_engine/gemm.h
--- a/transformer_engine/common/include/transformer_engine/layer_norm.h
+++ b/transformer_engine/common/include/transformer_engine/layer_norm.h
--- a/transformer_engine/common/include/transformer_engine/logging.h
+++ b/transformer_engine/common/include/transformer_engine/logging.h
--- a/transformer_engine/common/include/transformer_engine/transformer_engine.h
+++ b/transformer_engine/common/include/transformer_engine/transformer_engine.h
--- a/transformer_engine/common/include/transformer_engine/transpose.h
+++ b/transformer_engine/common/include/transformer_engine/transpose.h
--- a/transformer_engine/common/layer_norm/ln.h
+++ b/transformer_engine/common/layer_norm/ln.h
--- a/transformer_engine/common/layer_norm/ln_api.cpp
+++ b/transformer_engine/common/layer_norm/ln_api.cpp
--- a/transformer_engine/common/layer_norm/ln_bwd_kernels.cuh
+++ b/transformer_engine/common/layer_norm/ln_bwd_kernels.cuh
--- a/transformer_engine/common/layer_norm/ln_bwd_semi_cuda_kernel.cu
+++ b/transformer_engine/common/layer_norm/ln_bwd_semi_cuda_kernel.cu
--- a/transformer_engine/common/layer_norm/ln_fwd_cuda_kernel.cu
+++ b/transformer_engine/common/layer_norm/ln_fwd_cuda_kernel.cu
--- a/transformer_engine/common/layer_norm/ln_fwd_kernels.cuh
+++ b/transformer_engine/common/layer_norm/ln_fwd_kernels.cuh
--- a/transformer_engine/common/layer_norm/ln_kernel_traits.h
+++ b/transformer_engine/common/layer_norm/ln_kernel_traits.h
--- a/transformer_engine/common/recipe.py
+++ b/transformer_engine/common/recipe.py
--- a/transformer_engine/common/transformer_engine.cpp
+++ b/transformer_engine/common/transformer_engine.cpp
--- a/transformer_engine/common/transpose/cast_transpose.cu
+++ b/transformer_engine/common/transpose/cast_transpose.cu
--- a/transformer_engine/common/transpose/cast_transpose_fusion.cu
+++ b/transformer_engine/common/transpose/cast_transpose_fusion.cu
--- a/transformer_engine/common/transpose/transpose.cu
+++ b/transformer_engine/common/transpose/transpose.cu