Rebased the hipTENSOR development branch with the contraction branch

5d015452 · Chaitanya Inumella · b7fa6bb1 · ed3feb4d · 5d015452 · 5d015452
Commit 5d015452 authored Jul 06, 2022 by Chaitanya Inumella
20 changed files
--- a/example/19_binary_elementwise/elementwise_add_4d.cpp
+++ b/example/19_binary_elementwise/elementwise_add_4d.cpp
--- a/example/20_convnd_bwd_weight_xdl/CMakeLists.txt
+++ b/example/20_convnd_bwd_weight_xdl/CMakeLists.txt
 add_example_executable(example_convnd_bwd_weight_xdl convnd_bwd_weight_xdl.cpp)
-target_link_libraries(example_convnd_bwd_weight_xdl PRIVATE conv_util)
+add_example_executable(example_convnd_bwd_weight_xdl_bf16_splitk convnd_bwd_weight_xdl_bf16_splitk.cpp)
\ No newline at end of file
+target_link_libraries(example_convnd_bwd_weight_xdl PRIVATE conv_util)
+target_link_libraries(example_convnd_bwd_weight_xdl_bf16_splitk PRIVATE conv_util)
\ No newline at end of file
--- a/example/20_convnd_bwd_weight_xdl/convnd_bwd_weight_xdl.cpp
+++ b/example/20_convnd_bwd_weight_xdl/convnd_bwd_weight_xdl.cpp
--- a/example/20_convnd_bwd_weight_xdl/convnd_bwd_weight_xdl_bf16_splitk.cpp
+++ b/example/20_convnd_bwd_weight_xdl/convnd_bwd_weight_xdl_bf16_splitk.cpp
--- a/example/21_gemm_layernorm/CMakeLists.txt
+++ b/example/21_gemm_layernorm/CMakeLists.txt
+add_example_executable(example_gemm_bias_relu_add_layernorm_xdl_fp16 gemm_bias_relu_add_layernorm_xdl_fp16.cpp)
 add_example_executable(example_gemm_layernorm_xdl_fp16 gemm_layernorm_xdl_fp16.cpp)
+add_example_executable(example_gemm_xdl_layernorm_single_kernel_fp16 gemm_xdl_layernorm_single_kernel_fp16.cpp)
--- a/example/21_gemm_layernorm/gemm_bias_relu_add_layernorm_xdl_fp16.cpp
+++ b/example/21_gemm_layernorm/gemm_bias_relu_add_layernorm_xdl_fp16.cpp
--- a/example/21_gemm_layernorm/gemm_layernorm_xdl_fp16.cpp
+++ b/example/21_gemm_layernorm/gemm_layernorm_xdl_fp16.cpp
--- a/example/02_gemm_alpha_beta/gemm_xdl_alpha_beta.cpp
+++ b/example/02_gemm_alpha_beta/gemm_xdl_alpha_beta.cpp
--- a/example/22_cgemm/cgemm_xdl_fp16.cpp
+++ b/example/22_cgemm/cgemm_xdl_fp16.cpp
--- a/example/23_contraction/CMakeLists.txt
+++ b/example/23_contraction/CMakeLists.txt
-add_example_executable(example_contraction_xdl_fp32 contraction_xdl_fp32.cpp)
--- a/example/23_contraction/contraction_xdl_fp32.cpp
+++ b/example/23_contraction/contraction_xdl_fp32.cpp
--- a/example/23_softmax/CMakeLists.txt
+++ b/example/23_softmax/CMakeLists.txt
+add_example_executable(example_softmax_blockwise softmax_blockwise.cpp)
\ No newline at end of file
--- a/example/23_softmax/README.md
+++ b/example/23_softmax/README.md
--- a/example/23_softmax/softmax_blockwise.cpp
+++ b/example/23_softmax/softmax_blockwise.cpp
--- a/example/24_batched_gemm_c_permute/CMakeLists.txt
+++ b/example/24_batched_gemm_c_permute/CMakeLists.txt
--- a/example/24_batched_gemm_c_permute/batched_gemm_c_permute_xdl_fp16.cpp
+++ b/example/24_batched_gemm_c_permute/batched_gemm_c_permute_xdl_fp16.cpp
--- a/example/25_gemm_bias_c_permute/CMakeLists.txt
+++ b/example/25_gemm_bias_c_permute/CMakeLists.txt
--- a/example/25_gemm_bias_c_permute/gemm_bias_c_permute_xdl_fp16.cpp
+++ b/example/25_gemm_bias_c_permute/gemm_bias_c_permute_xdl_fp16.cpp
--- a/example/26_contraction/CMakeLists.txt
+++ b/example/26_contraction/CMakeLists.txt
--- a/example/23_contraction/README.md
+++ b/example/23_contraction/README.md