Merge branch 'develop' into ck_tile/fmha-fwd-splitkv-minor-opt

4ceba063 · Po Yen Chen · 25e10153 · 6df5fe2a · 4ceba063 · 4ceba063
Commit 4ceba063 authored Jan 03, 2025 by Po Yen Chen
20 changed files
--- a/include/ck/utility/static_buffer.hpp
+++ b/include/ck/utility/static_buffer.hpp
--- a/include/ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_pipeline_default_policy.hpp
+++ b/include/ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_pipeline_default_policy.hpp
--- a/include/ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_pipeline_one_pass.hpp
+++ b/include/ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_pipeline_one_pass.hpp
--- 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/layernorm2d/pipeline/layernorm2d_fwd_traits.hpp
+++ b/include/ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_traits.hpp
--- a/include/ck_tile/ops/welford.hpp
+++ b/include/ck_tile/ops/welford.hpp
--- a/include/ck_tile/ops/welford/block/block_welford.hpp
+++ b/include/ck_tile/ops/welford/block/block_welford.hpp
--- a/include/ck_tile/ops/welford/block/block_welford_problem.hpp
+++ b/include/ck_tile/ops/welford/block/block_welford_problem.hpp
--- a/include/ck_tile/ops/welford/thread/thread_welford.hpp
+++ b/include/ck_tile/ops/welford/thread/thread_welford.hpp
--- a/include/ck_tile/ref/naive_attention.hpp
+++ b/include/ck_tile/ref/naive_attention.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/tensor_operation_instance/device_operation_instance_factory.hpp
+++ b/library/include/ck/library/tensor_operation_instance/device_operation_instance_factory.hpp
--- a/library/include/ck/library/tensor_operation_instance/gpu/gemm_b_scale.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_b_scale.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/gemm_universal_streamk.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/gemm_universal_streamk.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