merge develop to attn-train-develop-qloop

76f2b6cd · danyao12 · 9b4c780a · 1ee99dca · 76f2b6cd · 76f2b6cd
Commit 76f2b6cd authored Jul 14, 2023 by danyao12
20 changed files
--- a/example/29_batched_gemm_bias_e_permute/batched_gemm_bias_e_permute_wmma_fp16.cpp
+++ b/example/29_batched_gemm_bias_e_permute/batched_gemm_bias_e_permute_wmma_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_conv_fwd_multiple_d/CMakeLists.txt
+++ b/example/30_grouped_conv_fwd_multiple_d/CMakeLists.txt
--- a/example/30_grouped_conv_fwd_multiple_d/common.hpp
+++ b/example/30_grouped_conv_fwd_multiple_d/common.hpp
--- a/example/30_grouped_conv_fwd_multiple_d/common_wmma.hpp
+++ b/example/30_grouped_conv_fwd_multiple_d/common_wmma.hpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_wmma_fp16.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_wmma_fp16.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_wmma_int8.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_wmma_int8.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_bf16.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_bf16.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_fp16.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_fp16.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_fp32.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_fp32.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_int4.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_int4.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_int8.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_bias_relu_add_xdl_int8.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_xdl_fp16.cpp
+++ b/example/30_grouped_conv_fwd_multiple_d/grouped_conv_fwd_xdl_fp16.cpp
--- a/example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_bias_relu_add_example.inc
+++ b/example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_bias_relu_add_example.inc
--- a/example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_bias_relu_add_wmma_example.inc
+++ b/example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_bias_relu_add_wmma_example.inc
--- a/example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_example.inc
+++ b/example/30_grouped_conv_fwd_multiple_d/run_grouped_conv_fwd_example.inc
--- a/example/31_batched_gemm_gemm/CMakeLists.txt
+++ b/example/31_batched_gemm_gemm/CMakeLists.txt
--- a/example/31_batched_gemm_gemm/batched_gemm_gemm_xdl_bf16.cpp
+++ b/example/31_batched_gemm_gemm/batched_gemm_gemm_xdl_bf16.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/31_batched_gemm_gemm/batched_gemm_gemm_xdl_fp32.cpp
+++ b/example/31_batched_gemm_gemm/batched_gemm_gemm_xdl_fp32.cpp