Dynamic tensor descriptor (#24)

* support dynamic tensor descriptor * use buffer load OOB feature for padding case * add navi support * add int8x4 inference kernel Co-authored-by: Chao Liu <chao@ixt-rack-81.local.lan> Co-authored-by: Jing Zhang <jizhan@amd.com>

Dynamic tensor descriptor (#24)
* support dynamic tensor descriptor * use buffer load OOB feature for padding case * add navi support * add int8x4 inference kernel Co-authored-by: Chao Liu <chao@ixt-rack-81.local.lan> Co-authored-by: Jing Zhang <jizhan@amd.com>
fcbb9788 · Chao Liu · GitHub · bbcb67d0 · fcbb9788 · fcbb9788
Unverified Commit fcbb9788 authored Mar 25, 2021 by Chao Liu Committed by GitHub Mar 25, 2021
20 changed files
--- a/composable_kernel/include/tensor_description/tensor_descriptor.hpp
+++ b/composable_kernel/include/tensor_description/tensor_descriptor.hpp
--- a/composable_kernel/include/tensor_description/tensor_descriptor_helper.hpp
+++ b/composable_kernel/include/tensor_description/tensor_descriptor_helper.hpp
--- a/composable_kernel/include/tensor_operation/blockwise_batched_gemm.hpp
+++ b/composable_kernel/include/tensor_operation/blockwise_batched_gemm.hpp
--- a/composable_kernel/include/tensor_operation/blockwise_dynamic_tensor_slice_transfer.hpp
+++ b/composable_kernel/include/tensor_operation/blockwise_dynamic_tensor_slice_transfer.hpp
--- a/composable_kernel/include/tensor_operation/blockwise_gemm.hpp
+++ b/composable_kernel/include/tensor_operation/blockwise_gemm.hpp
--- a/composable_kernel/include/tensor_operation/blockwise_gemm_v2.hpp
+++ b/composable_kernel/include/tensor_operation/blockwise_gemm_v2.hpp
--- a/composable_kernel/include/tensor_operation/blockwise_gemm_v3.hpp
+++ b/composable_kernel/include/tensor_operation/blockwise_gemm_v3.hpp
--- a/composable_kernel/include/tensor_operation/blockwise_generic_tensor_slice_copy.hpp
+++ b/composable_kernel/include/tensor_operation/blockwise_generic_tensor_slice_copy.hpp
--- a/composable_kernel/include/tensor_operation/gridwise_dynamic_gemm.hpp
+++ b/composable_kernel/include/tensor_operation/gridwise_dynamic_gemm.hpp
--- a/composable_kernel/include/tensor_operation/gridwise_dynamic_gemm_v2.hpp
+++ b/composable_kernel/include/tensor_operation/gridwise_dynamic_gemm_v2.hpp
--- a/composable_kernel/include/tensor_operation/gridwise_gemm.hpp
+++ b/composable_kernel/include/tensor_operation/gridwise_gemm.hpp
--- a/composable_kernel/include/tensor_operation/gridwise_tensor_contraction.hpp
+++ b/composable_kernel/include/tensor_operation/gridwise_tensor_contraction.hpp
--- a/composable_kernel/include/tensor_operation/threadwise_dynamic_tensor_slice_transfer.hpp
+++ b/composable_kernel/include/tensor_operation/threadwise_dynamic_tensor_slice_transfer.hpp
--- a/composable_kernel/include/tensor_operation/threadwise_gemm.hpp
+++ b/composable_kernel/include/tensor_operation/threadwise_gemm.hpp
--- a/composable_kernel/include/tensor_operation/threadwise_gemm_v2.hpp
+++ b/composable_kernel/include/tensor_operation/threadwise_gemm_v2.hpp
--- a/composable_kernel/include/tensor_operation/threadwise_gemm_v3.hpp
+++ b/composable_kernel/include/tensor_operation/threadwise_gemm_v3.hpp
--- a/composable_kernel/include/tensor_operation/threadwise_generic_tensor_slice_copy.hpp
+++ b/composable_kernel/include/tensor_operation/threadwise_generic_tensor_slice_copy.hpp
--- a/composable_kernel/include/utility/amd_buffer_addressing.hpp
+++ b/composable_kernel/include/utility/amd_buffer_addressing.hpp
--- a/composable_kernel/include/utility/amd_buffer_addressing_v2.hpp
+++ b/composable_kernel/include/utility/amd_buffer_addressing_v2.hpp
--- a/composable_kernel/include/utility/amd_inline_asm.hpp
+++ b/composable_kernel/include/utility/amd_inline_asm.hpp