Merge branch 'develop' into bnorm_bwd_pr

a781d078 · Qianfeng Zhang · fd76c787 · 4c4c7328 · a781d078 · a781d078
Commit a781d078 authored Nov 16, 2022 by Qianfeng Zhang
20 changed files
--- a/profiler/src/profile_softmax.cpp
+++ b/profiler/src/profile_softmax.cpp
--- a/profiler/src/profiler.cpp
+++ b/profiler/src/profiler.cpp
--- a/script/cmake-ck-dev.sh
+++ b/script/cmake-ck-dev.sh
--- a/script/cmake-ck-release.sh
+++ b/script/cmake-ck-release.sh
--- a/test/CMakeLists.txt
+++ b/test/CMakeLists.txt
--- a/test/batched_gemm_softmax_gemm_permute/CMakeLists.txt
+++ b/test/batched_gemm_softmax_gemm_permute/CMakeLists.txt
--- a/test/batched_gemm_softmax_gemm_permute/test_batched_gemm_softmax_gemm_permute_bf16.cpp
+++ b/test/batched_gemm_softmax_gemm_permute/test_batched_gemm_softmax_gemm_permute_bf16.cpp
--- a/test/batched_gemm_softmax_gemm_permute/test_batched_gemm_softmax_gemm_permute_util.hpp
+++ b/test/batched_gemm_softmax_gemm_permute/test_batched_gemm_softmax_gemm_permute_util.hpp
--- a/test/convnd_bwd_weight/CMakeLists.txt
+++ b/test/convnd_bwd_weight/CMakeLists.txt
--- a/test/elementwise_normalization/CMakeLists.txt
+++ b/test/elementwise_normalization/CMakeLists.txt
--- a/test/elementwise_normalization/test_elementwise_layernorm_fp16.cpp
+++ b/test/elementwise_normalization/test_elementwise_layernorm_fp16.cpp
--- a/test/gemm/gemm_util.hpp
+++ b/test/gemm/gemm_util.hpp
--- a/test/gemm_split_k/gemm_split_k.cpp
+++ b/test/gemm_split_k/gemm_split_k.cpp
--- a/test/grouped_convnd_bwd_weight/CMakeLists.txt
+++ b/test/grouped_convnd_bwd_weight/CMakeLists.txt
--- a/test/convnd_bwd_weight/convnd_bwd_weight.cpp
+++ b/test/convnd_bwd_weight/convnd_bwd_weight.cpp
--- a/test/normalization/CMakeLists.txt
+++ b/test/normalization/CMakeLists.txt
--- a/test/normalization/test_groupnorm_fp16.cpp
+++ b/test/normalization/test_groupnorm_fp16.cpp
--- a/test/normalization/test_groupnorm_fp32.cpp
+++ b/test/normalization/test_groupnorm_fp32.cpp
--- a/test/normalization/test_layernorm2d_fp16.cpp
+++ b/test/normalization/test_layernorm2d_fp16.cpp
--- a/test/normalization/test_layernorm2d_fp32.cpp
+++ b/test/normalization/test_layernorm2d_fp32.cpp