Merge remote-tracking branch 'origin/develop' into wavelet_model

aa5859e4 · Chao Liu · 9bd6cc0e · 5ee30459 · aa5859e4 · aa5859e4
Commit aa5859e4 authored Aug 13, 2022 by Chao Liu
20 changed files
--- a/example/12_reduce/reduce_blockwise_two_call.cpp
+++ b/example/12_reduce/reduce_blockwise_two_call.cpp
--- a/example/12_reduce/reduce_example_common.hpp
+++ b/example/12_reduce/reduce_example_common.hpp
--- a/example/12_reduce/reduce_multiblock_atomic_add.cpp
+++ b/example/12_reduce/reduce_multiblock_atomic_add.cpp
--- a/example/12_reduce/reduce_multiblock_atomic_add_impl.hpp
+++ b/example/12_reduce/reduce_multiblock_atomic_add_impl.hpp
--- a/example/13_pool2d_fwd/pool2d_fwd_common.hpp
+++ b/example/13_pool2d_fwd/pool2d_fwd_common.hpp
--- a/example/14_gemm_xdl_requant_relu_requant/gemm_xdl_requant_relu_requant_int8.cpp
+++ b/example/14_gemm_xdl_requant_relu_requant/gemm_xdl_requant_relu_requant_int8.cpp
--- a/example/15_grouped_gemm/grouped_gemm_xdl_fp16.cpp
+++ b/example/15_grouped_gemm/grouped_gemm_xdl_fp16.cpp
--- a/example/16_gemm_multi_d_multi_reduces/CMakeLists.txt
+++ b/example/16_gemm_multi_d_multi_reduces/CMakeLists.txt
+add_example_executable(example_gemm_add_add_mean_meansquare_xdl_fp16 gemm_add_add_mean_meansquare_xdl_fp16.cpp)
+add_example_executable(example_gemm_mean_meansquare_xdl_fp16 gemm_mean_meansquare_xdl_fp16.cpp)
+
+#exclude GEMM+max exampe from testing, since there is random failure on gfx908
+#https://github.com/ROCmSoftwarePlatform/composable_kernel/issues/358
+#TODO: fix the failure and re-enable this test
+add_example_executable_no_testing(example_gemm_max_xdl_fp16 gemm_max_xdl_fp16.cpp)
--- a/example/16_gemm_multi_d_multi_reduces/gemm_add_add_mean_meansquare_xdl_fp16.cpp
+++ b/example/16_gemm_multi_d_multi_reduces/gemm_add_add_mean_meansquare_xdl_fp16.cpp
--- a/example/16_gemm_multi_d_multi_reduces/gemm_max_xdl_fp16.cpp
+++ b/example/16_gemm_multi_d_multi_reduces/gemm_max_xdl_fp16.cpp
--- a/example/16_gemm_multi_d_multi_reduces/gemm_mean_meansquare_xdl_fp16.cpp
+++ b/example/16_gemm_multi_d_multi_reduces/gemm_mean_meansquare_xdl_fp16.cpp
--- a/example/16_gemm_reduce/CMakeLists.txt
+++ b/example/16_gemm_reduce/CMakeLists.txt
-add_example_executable(example_gemm_reduce_xdl_max_fp16 gemm_reduce_xdl_max_fp16.cpp)
-add_example_executable(example_gemm_reduce_xdl_mean_squaremean_fp16 gemm_reduce_xdl_mean_squaremean_fp16.cpp)
--- a/example/16_gemm_reduce/gemm_reduce_xdl_mean_squaremean_fp16.cpp
+++ b/example/16_gemm_reduce/gemm_reduce_xdl_mean_squaremean_fp16.cpp
--- a/example/17_convnd_bwd_data/CMakeLists.txt
+++ b/example/17_convnd_bwd_data/CMakeLists.txt
--- a/example/17_convnd_bwd_data_xdl/README.md
+++ b/example/17_convnd_bwd_data_xdl/README.md
--- a/example/17_convnd_bwd_data/convnd_bwd_data_common.hpp
+++ b/example/17_convnd_bwd_data/convnd_bwd_data_common.hpp
--- a/example/17_convnd_bwd_data/convnd_bwd_data_xdl_fp16.cpp
+++ b/example/17_convnd_bwd_data/convnd_bwd_data_xdl_fp16.cpp
--- a/example/17_convnd_bwd_data_xdl/CMakeLists.txt
+++ b/example/17_convnd_bwd_data_xdl/CMakeLists.txt
--- a/example/17_convnd_bwd_data_xdl/convnd_bwd_data_xdl.cpp
+++ b/example/17_convnd_bwd_data_xdl/convnd_bwd_data_xdl.cpp
--- a/example/18_batched_gemm_reduce/batched_gemm_reduce_xdl_fp16.cpp
+++ b/example/18_batched_gemm_reduce/batched_gemm_reduce_xdl_fp16.cpp