Merge branch 'amd-develop' into amd-master

b924e330 · Jun Liu · 72c9f129 · 9c0811f3 · b924e330 · b924e330
Commit b924e330 authored Oct 03, 2024 by Jun Liu
20 changed files
--- a/include/ck_tile/ops/gemm/kernel/gemm_tile_partitioner.hpp
+++ b/include/ck_tile/ops/gemm/kernel/gemm_tile_partitioner.hpp
--- a/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_agmem_bgmem_creg_v1.hpp
+++ b/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_agmem_bgmem_creg_v1.hpp
--- a/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_agmem_bgmem_creg_v1_default_policy.hpp
+++ b/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_agmem_bgmem_creg_v1_default_policy.hpp
--- a/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_agmem_bgmem_creg_v2.hpp
+++ b/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_agmem_bgmem_creg_v2.hpp
@@ -4,6 +4,7 @@
 #pragma once
 #include "ck_tile/core.hpp"
+#include "ck_tile/ops/gemm/pipeline/block_gemm_pipeline_agmem_bgmem_creg_v2_default_policy.hpp"
 namespace ck_tile {

--- a/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_problem.hpp
+++ b/include/ck_tile/ops/gemm/pipeline/block_gemm_pipeline_problem.hpp
--- a/include/ck_tile/ops/gemm/pipeline/tile_gemm_shape.hpp
+++ b/include/ck_tile/ops/gemm/pipeline/tile_gemm_shape.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/avg_pool2d_bwd.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/avg_pool2d_bwd.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm_universal.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_universal.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_comp_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_comp_xdl.inc
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_inter_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_inter_xdl.inc
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_intra_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_mem_intra_xdl.inc
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl.inc
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl_merged_groups.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_xdl_merged_groups.inc
--- a/library/include/ck/library/tensor_operation_instance/gpu/max_pool_bwd.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/max_pool_bwd.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/pool2d_fwd.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/pool2d_fwd.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/pool3d_fwd.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/pool3d_fwd.hpp
--- a/library/src/tensor_operation_instance/gpu/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/avg_pool2d_bwd/CMakeLists.txt
+++ b/library/src/tensor_operation_instance/gpu/avg_pool2d_bwd/CMakeLists.txt
--- a/library/src/tensor_operation_instance/gpu/avg_pool2d_bwd/device_avg_pool2d_bwd_nhwc_bf16_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/avg_pool2d_bwd/device_avg_pool2d_bwd_nhwc_bf16_instance.cpp