Merge branch 'develop' into jizhan/reduce_threadwise_multi_d

300337cd · letaoqin · f306d02e · 02fa2c29 · 300337cd · 300337cd
Commit 300337cd authored May 30, 2024 by letaoqin
20 changed files
--- a/example/61_contraction_multi_ABD/contraction_multi_ABD_xdl_fp16.cpp
+++ b/example/61_contraction_multi_ABD/contraction_multi_ABD_xdl_fp16.cpp
--- a/example/62_convnd_activ/binary/CMakeLists.txt
+++ b/example/62_convnd_activ/binary/CMakeLists.txt
--- a/example/62_convnd_activ/binary/convnd_bwd_weight_xdl_bilinear_residual_fp16.cpp
+++ b/example/62_convnd_activ/binary/convnd_bwd_weight_xdl_bilinear_residual_fp16.cpp
--- a/example/CMakeLists.txt
+++ b/example/CMakeLists.txt
--- a/example/ck_tile/01_fmha/CMakeLists.txt
+++ b/example/ck_tile/01_fmha/CMakeLists.txt
--- a/example/ck_tile/01_fmha/README.md
+++ b/example/ck_tile/01_fmha/README.md
--- a/example/ck_tile/01_fmha/bias.hpp
+++ b/example/ck_tile/01_fmha/bias.hpp
--- a/example/ck_tile/01_fmha/fmha_fwd.cpp
+++ b/example/ck_tile/01_fmha/fmha_fwd.cpp
--- a/example/ck_tile/01_fmha/fmha_fwd.hpp
+++ b/example/ck_tile/01_fmha/fmha_fwd.hpp
--- a/example/ck_tile/01_fmha/generate.py
+++ b/example/ck_tile/01_fmha/generate.py
--- a/example/ck_tile/01_fmha/mask.hpp
+++ b/example/ck_tile/01_fmha/mask.hpp
--- a/example/ck_tile/01_fmha/misc/gamc.png
+++ b/example/ck_tile/01_fmha/misc/gamc.png
--- a/example/ck_tile/01_fmha/script/benchmark.sh
+++ b/example/ck_tile/01_fmha/script/benchmark.sh
--- a/example/ck_tile/01_fmha/script/smoke_test.sh
+++ b/example/ck_tile/01_fmha/script/smoke_test.sh
--- a/example/ck_tile/01_fmha/utils.hpp
+++ b/example/ck_tile/01_fmha/utils.hpp
--- a/example/ck_tile/CMakeLists.txt
+++ b/example/ck_tile/CMakeLists.txt
--- a/example/ck_tile/remod.py
+++ b/example/ck_tile/remod.py
--- a/include/ck/ck.hpp
+++ b/include/ck/ck.hpp
--- a/include/ck/config.h.in
+++ b/include/ck/config.h.in
--- a/include/ck/host_utility/device_prop.hpp
+++ b/include/ck/host_utility/device_prop.hpp