merge develop

1dbdab56 · Jing Zhang · d2e49b23 · bac7df8f · 1dbdab56 · 1dbdab56
Commit 1dbdab56 authored Aug 18, 2022 by Jing Zhang
20 changed files
--- a/example/28_grouped_gemm_bias_e_permute/grouped_gemm_bias_e_permute_xdl_fp16.cpp
+++ b/example/28_grouped_gemm_bias_e_permute/grouped_gemm_bias_e_permute_xdl_fp16.cpp
--- a/example/29_batched_gemm_bias_e_permute/CMakeLists.txt
+++ b/example/29_batched_gemm_bias_e_permute/CMakeLists.txt
+add_example_executable(example_batched_gemm_bias_e_permute_xdl_fp16 batched_gemm_bias_e_permute_xdl_fp16.cpp)
--- a/example/29_batched_gemm_bias_e_permute/batched_gemm_bias_e_permute_xdl_fp16.cpp
+++ b/example/29_batched_gemm_bias_e_permute/batched_gemm_bias_e_permute_xdl_fp16.cpp
--- a/example/30_grouped_convnd_fwd_bias_relu/CMakeLists.txt
+++ b/example/30_grouped_convnd_fwd_bias_relu/CMakeLists.txt
-add_example_executable(example_grouped_convnd_fwd_bias_relu_xdl_fp16 grouped_convnd_fwd_bias_relu_xdl_fp16.cpp)
-target_link_libraries(example_grouped_convnd_fwd_bias_relu_xdl_fp16 PRIVATE utility)
--- a/example/30_grouped_convnd_fwd_bias_relu_add/CMakeLists.txt
+++ b/example/30_grouped_convnd_fwd_bias_relu_add/CMakeLists.txt
+add_example_executable(example_grouped_convnd_fwd_bias_relu_add_xdl_fp16 grouped_convnd_fwd_bias_relu_add_xdl_fp16.cpp)
+target_link_libraries(example_grouped_convnd_fwd_bias_relu_add_xdl_fp16 PRIVATE utility)
+
+add_example_executable(example_grouped_convnd_fwd_bias_relu_add_xdl_fp32 grouped_convnd_fwd_bias_relu_add_xdl_fp32.cpp)
+target_link_libraries(example_grouped_convnd_fwd_bias_relu_add_xdl_fp32 PRIVATE utility)
+
+add_example_executable(example_grouped_convnd_fwd_bias_relu_add_xdl_bf16 grouped_convnd_fwd_bias_relu_add_xdl_bf16.cpp)
+target_link_libraries(example_grouped_convnd_fwd_bias_relu_add_xdl_bf16 PRIVATE utility)
+
+add_example_executable(example_grouped_convnd_fwd_bias_relu_add_xdl_int8 grouped_convnd_fwd_bias_relu_add_xdl_int8.cpp)
+target_link_libraries(example_grouped_convnd_fwd_bias_relu_add_xdl_int8 PRIVATE utility)
\ No newline at end of file
--- a/example/30_grouped_convnd_fwd_bias_relu/README.md
+++ b/example/30_grouped_convnd_fwd_bias_relu/README.md
--- a/example/30_grouped_convnd_fwd_bias_relu/grouped_convnd_fwd_bias_common.hpp
+++ b/example/30_grouped_convnd_fwd_bias_relu/grouped_convnd_fwd_bias_common.hpp
--- a/example/30_grouped_convnd_fwd_bias_relu_add/grouped_convnd_fwd_bias_relu_add_xdl_bf16.cpp
+++ b/example/30_grouped_convnd_fwd_bias_relu_add/grouped_convnd_fwd_bias_relu_add_xdl_bf16.cpp
--- a/example/30_grouped_convnd_fwd_bias_relu/grouped_convnd_fwd_bias_relu_xdl_fp16.cpp
+++ b/example/30_grouped_convnd_fwd_bias_relu/grouped_convnd_fwd_bias_relu_xdl_fp16.cpp
--- a/example/30_grouped_convnd_fwd_bias_relu_add/grouped_convnd_fwd_bias_relu_add_xdl_fp32.cpp
+++ b/example/30_grouped_convnd_fwd_bias_relu_add/grouped_convnd_fwd_bias_relu_add_xdl_fp32.cpp
--- a/example/30_grouped_convnd_fwd_bias_relu_add/grouped_convnd_fwd_bias_relu_add_xdl_int8.cpp
+++ b/example/30_grouped_convnd_fwd_bias_relu_add/grouped_convnd_fwd_bias_relu_add_xdl_int8.cpp
--- a/example/31_batched_gemm_gemm/CMakeLists.txt
+++ b/example/31_batched_gemm_gemm/CMakeLists.txt
+add_example_executable(example_batched_gemm_gemm_xdl_fp16 batched_gemm_gemm_xdl_fp16.cpp)
--- a/example/31_batched_gemm_gemm/batched_gemm_gemm_xdl_fp16.cpp
+++ b/example/31_batched_gemm_gemm/batched_gemm_gemm_xdl_fp16.cpp
--- a/example/32_batched_gemm_scale_softmax_gemm/CMakeLists.txt
+++ b/example/32_batched_gemm_scale_softmax_gemm/CMakeLists.txt
+add_example_executable(example_batched_gemm_scale_softmax_gemm_xdl_fp16 batched_gemm_scale_softmax_gemm_xdl_fp16.cpp)
--- a/example/32_batched_gemm_scale_softmax_gemm/batched_gemm_scale_softmax_gemm_xdl_fp16.cpp
+++ b/example/32_batched_gemm_scale_softmax_gemm/batched_gemm_scale_softmax_gemm_xdl_fp16.cpp
--- a/example/33_multiple_reduce/CMakeLists.txt
+++ b/example/33_multiple_reduce/CMakeLists.txt
+add_example_executable(example_dual_reduce_multiblock dual_reduce_multiblock.cpp)
+add_example_executable(example_dual_reduce_threadwise dual_reduce_threadwise.cpp)
--- a/example/33_multiple_reduce/README.md
+++ b/example/33_multiple_reduce/README.md
--- a/example/33_multiple_reduce/dual_reduce_common.hpp
+++ b/example/33_multiple_reduce/dual_reduce_common.hpp
--- a/example/33_multiple_reduce/dual_reduce_multiblock.cpp
+++ b/example/33_multiple_reduce/dual_reduce_multiblock.cpp
--- a/example/33_multiple_reduce/dual_reduce_threadwise.cpp
+++ b/example/33_multiple_reduce/dual_reduce_threadwise.cpp