Merge remote-tracking branch 'origin/develop' into ck_tile/layernorm_fusion

55cdf2b9 · carlushuang · 4b59b5c9 · b098b71b · 55cdf2b9 · 55cdf2b9
Commit 55cdf2b9 authored Oct 27, 2024 by carlushuang
20 changed files
--- a/include/ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_pipeline_two_pass.hpp
+++ b/include/ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_pipeline_two_pass.hpp
--- a/include/ck_tile/ops/reduce/block/block_reduce.hpp
+++ b/include/ck_tile/ops/reduce/block/block_reduce.hpp
--- a/include/ck_tile/ops/softmax.hpp
+++ b/include/ck_tile/ops/softmax.hpp
--- a/include/ck_tile/ops/softmax/block/block_softmax_2d.hpp
+++ b/include/ck_tile/ops/softmax/block/block_softmax_2d.hpp
--- a/include/ck_tile/ops/softmax/block/block_softmax_2d_problem.hpp
+++ b/include/ck_tile/ops/softmax/block/block_softmax_2d_problem.hpp
--- a/include/ck_tile/ops/topk.hpp
+++ b/include/ck_tile/ops/topk.hpp
--- a/include/ck_tile/ops/topk/block/block_topk_stream_2d.hpp
+++ b/include/ck_tile/ops/topk/block/block_topk_stream_2d.hpp
--- a/include/ck_tile/ops/topk/block/block_topk_stream_2d_problem.hpp
+++ b/include/ck_tile/ops/topk/block/block_topk_stream_2d_problem.hpp
--- a/include/ck_tile/ops/topk_softmax.hpp
+++ b/include/ck_tile/ops/topk_softmax.hpp
--- a/include/ck_tile/ops/topk_softmax/kernel/topk_softmax_kernel.hpp
+++ b/include/ck_tile/ops/topk_softmax/kernel/topk_softmax_kernel.hpp
--- a/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_pipeline.hpp
+++ b/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_pipeline.hpp
--- a/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_policy.hpp
+++ b/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_policy.hpp
--- a/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_problem.hpp
+++ b/include/ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_problem.hpp
--- a/library/include/ck/library/reference_tensor_operation/gpu/reference_gemm.hpp
+++ b/library/include/ck/library/reference_tensor_operation/gpu/reference_gemm.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm_multiply_multiply.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_multiply_multiply.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_weight/device_grouped_conv_bwd_weight_xdl_instance.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_dynamic_op_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_dynamic_op_instance.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_xdl.inc
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_backward_weight_xdl.inc
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_dynamic_op.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_convolution_forward_dynamic_op.hpp