best performing kernel for GEMV codex problem with M=1 with inverted B matrix

930b2872 · Harisankar Sadasivan · a1e17d18 · a4f72a31 · 930b2872 · 930b2872
Commit 930b2872 authored Oct 11, 2023 by Harisankar Sadasivan
20 changed files
--- a/include/ck/utility/amd_xdlops.hpp
+++ b/include/ck/utility/amd_xdlops.hpp
--- a/include/ck/utility/dynamic_buffer.hpp
+++ b/include/ck/utility/dynamic_buffer.hpp
--- a/include/ck/utility/is_detected.hpp
+++ b/include/ck/utility/is_detected.hpp
--- a/include/ck/utility/sequence.hpp
+++ b/include/ck/utility/sequence.hpp
--- a/include/ck/utility/tuple.hpp
+++ b/include/ck/utility/tuple.hpp
--- a/include/ck/utility/type_convert.hpp
+++ b/include/ck/utility/type_convert.hpp
--- a/library/include/ck/library/reference_tensor_operation/cpu/reference_column_to_image.hpp
+++ b/library/include/ck/library/reference_tensor_operation/cpu/reference_column_to_image.hpp
--- a/library/include/ck/library/reference_tensor_operation/cpu/reference_conv_bwd_weight.hpp
+++ b/library/include/ck/library/reference_tensor_operation/cpu/reference_conv_bwd_weight.hpp
--- a/library/include/ck/library/reference_tensor_operation/cpu/reference_conv_fwd.hpp
+++ b/library/include/ck/library/reference_tensor_operation/cpu/reference_conv_fwd.hpp
--- a/library/include/ck/library/reference_tensor_operation/cpu/reference_gemm.hpp
+++ b/library/include/ck/library/reference_tensor_operation/cpu/reference_gemm.hpp
--- a/library/include/ck/library/reference_tensor_operation/cpu/reference_image_to_column.hpp
+++ b/library/include/ck/library/reference_tensor_operation/cpu/reference_image_to_column.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/batchnorm_backward.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/batchnorm_backward.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/batchnorm_forward.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/batchnorm_forward.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/batchnorm_infer.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/batchnorm_infer.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/conv_tensor_rearrange.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/conv_tensor_rearrange.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/conv_tensor_rearrange/device_column_to_image_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/conv_tensor_rearrange/device_column_to_image_instance.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/image_to_column/device_image_to_column_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/image_to_column/device_image_to_column_instance.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm_bilinear.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_bilinear.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm_splitk.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_splitk.hpp