merge navi31_rel

ccaea50e · Jing Zhang · 0b914465 · 10127959 · ccaea50e · ccaea50e
Commit ccaea50e authored Mar 08, 2024 by Jing Zhang
20 changed files
--- a/example/01_gemm/CMakeLists.txt
+++ b/example/01_gemm/CMakeLists.txt
--- a/example/01_gemm/common.hpp
+++ b/example/01_gemm/common.hpp
--- a/example/01_gemm/gemm_xdl_fp8.cpp
+++ b/example/01_gemm/gemm_xdl_fp8.cpp
--- a/example/01_gemm/gemm_xdl_fp8_bf8.cpp
+++ b/example/01_gemm/gemm_xdl_fp8_bf8.cpp
--- a/example/01_gemm/run_gemm_example.inc
+++ b/example/01_gemm/run_gemm_example.inc
--- a/example/29_batched_gemm_bias_e_permute/CMakeLists.txt
+++ b/example/29_batched_gemm_bias_e_permute/CMakeLists.txt
--- 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/32_batched_gemm_scale_softmax_gemm/CMakeLists.txt
+++ b/example/32_batched_gemm_scale_softmax_gemm/CMakeLists.txt
--- a/example/32_batched_gemm_scale_softmax_gemm/cross_attention_forward_wmma_fp16.cpp
+++ b/example/32_batched_gemm_scale_softmax_gemm/cross_attention_forward_wmma_fp16.cpp
--- a/example/32_batched_gemm_scale_softmax_gemm/run_batched_gemm_scale_softmax_gemm_permute_wmma.inc
+++ b/example/32_batched_gemm_scale_softmax_gemm/run_batched_gemm_scale_softmax_gemm_permute_wmma.inc
--- a/example/32_batched_gemm_scale_softmax_gemm/run_cross_attention_wmma.inc
+++ b/example/32_batched_gemm_scale_softmax_gemm/run_cross_attention_wmma.inc
--- a/example/32_batched_gemm_scale_softmax_gemm/run_grouped_query_attention_forward_wmma.inc
+++ b/example/32_batched_gemm_scale_softmax_gemm/run_grouped_query_attention_forward_wmma.inc
--- a/example/32_batched_gemm_scale_softmax_gemm/run_multi_query_attention_forward_wmma.inc
+++ b/example/32_batched_gemm_scale_softmax_gemm/run_multi_query_attention_forward_wmma.inc
--- a/example/32_batched_gemm_scale_softmax_gemm/run_self_attention_wmma.inc
+++ b/example/32_batched_gemm_scale_softmax_gemm/run_self_attention_wmma.inc
--- a/example/32_batched_gemm_scale_softmax_gemm/self_attention_forward_wmma_fp16.cpp
+++ b/example/32_batched_gemm_scale_softmax_gemm/self_attention_forward_wmma_fp16.cpp
--- a/example/64_fpAintB_gemm/CMakeLists.txt
+++ b/example/64_fpAintB_gemm/CMakeLists.txt
--- a/include/ck/tensor_operation/gpu/device/impl/device_batched_gemm_softmax_gemm_permute_wmma_cshuffle.hpp
+++ b/include/ck/tensor_operation/gpu/device/impl/device_batched_gemm_softmax_gemm_permute_wmma_cshuffle.hpp
--- a/include/ck/tensor_operation/gpu/device/impl/device_fpAintB_gemm_wmma.hpp
+++ b/include/ck/tensor_operation/gpu/device/impl/device_fpAintB_gemm_wmma.hpp
--- a/include/ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp
+++ b/include/ck/tensor_operation/gpu/device/impl/device_gemm_multiple_d_xdl_cshuffle.hpp
--- a/include/ck/tensor_operation/gpu/device/impl/device_grouped_query_attention_forward_wmma.hpp
+++ b/include/ck/tensor_operation/gpu/device/impl/device_grouped_query_attention_forward_wmma.hpp