Update include path to break the remod's cyclic dep issue (#2978)

* Update include path to break the cyclic dep issue * Use ck_tile::permute_vectors_i4x4_b in tile engine --------- Co-authored-by: Damien Lejeune <damien.lejeune@amd.com> Co-authored-by: Adam Osewski <19374865+aosewski@users.noreply.github.com>
2026-05-03 21:21:22 +00:00 · 2025-10-13 13:24:47 +02:00
parent e9f0cc83a8
commit 46c10c316d
25 changed files with 51 additions and 61 deletions
--- a/include/ck_tile/host.hpp
+++ b/include/ck_tile/host.hpp
@@ -16,6 +16,7 @@
 #include "ck_tile/host/host_tensor.hpp"
 #include "ck_tile/host/joinable_thread.hpp"
 #include "ck_tile/host/kernel_launch.hpp"
+#include "ck_tile/host/permute_pk_int4.hpp"
 #include "ck_tile/host/ranges.hpp"
 #include "ck_tile/host/reference/reference_batched_dropout.hpp"
 #include "ck_tile/host/reference/reference_batched_dropout_randval.hpp"
--- a/include/ck_tile/ops/add_rmsnorm2d_rdquant.hpp
+++ b/include/ck_tile/ops/add_rmsnorm2d_rdquant.hpp
@@ -9,5 +9,7 @@
 #include "ck_tile/ops/add_rmsnorm2d_rdquant/pipeline/add_rmsnorm2d_rdquant_fwd_pipeline_problem.hpp"
 #include "ck_tile/ops/add_rmsnorm2d_rdquant/pipeline/add_rmsnorm2d_rdquant_fwd_pipeline_three_pass.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/batched_transpose.hpp
+++ b/include/ck_tile/ops/batched_transpose.hpp
@@ -12,5 +12,7 @@
 #include "ck_tile/ops/batched_transpose/pipeline/batched_transpose_policy.hpp"
 #include "ck_tile/ops/batched_transpose/pipeline/batched_transpose_problem.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/common.hpp
+++ b/include/ck_tile/ops/common.hpp
@@ -4,6 +4,7 @@
 #pragma once

 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
-#include "ck_tile/ops/common/streamk_common.hpp"
--- a/include/ck_tile/ops/common/load_interleaved_pk_type.hpp
+++ b/include/ck_tile/ops/common/load_interleaved_pk_type.hpp
@@ -4,7 +4,7 @@
 #pragma once

 #include "ck_tile/core/config.hpp"
-#include "ck_tile/ops/elementwise.hpp"
+#include "ck_tile/ops/elementwise/unary_element_wise_operation.hpp"

 namespace ck_tile {

--- a/include/ck_tile/ops/elementwise.hpp
+++ b/include/ck_tile/ops/elementwise.hpp
@@ -10,5 +10,7 @@
 #include "ck_tile/ops/elementwise/pipeline/elementwise_shape.hpp"
 #include "ck_tile/ops/elementwise/unary_element_wise_operation.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/epilogue.hpp
+++ b/include/ck_tile/ops/epilogue.hpp
@@ -8,5 +8,7 @@
 #include "ck_tile/ops/epilogue/default_2d_epilogue.hpp"
 #include "ck_tile/ops/epilogue/dynamic_quant_epilogue.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/flatmm.hpp
+++ b/include/ck_tile/ops/flatmm.hpp
@@ -14,5 +14,7 @@
 #include "ck_tile/ops/flatmm/pipeline/flatmm_pipeline_agmem_bgmem_creg_v1_policy.hpp"
 #include "ck_tile/ops/flatmm/pipeline/tile_flatmm_shape.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/fmha.hpp
+++ b/include/ck_tile/ops/fmha.hpp
@@ -60,5 +60,7 @@
 #include "ck_tile/ops/fmha/pipeline/tile_fmha_shape.hpp"
 #include "ck_tile/ops/fmha/pipeline/tile_fmha_traits.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/fused_moe.hpp
+++ b/include/ck_tile/ops/fused_moe.hpp
@@ -16,5 +16,7 @@
 #include "ck_tile/ops/fused_moe/pipeline/moe_sorting_pipeline.hpp"
 #include "ck_tile/ops/fused_moe/pipeline/moe_sorting_policy.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/gemm.hpp
+++ b/include/ck_tile/ops/gemm.hpp
@@ -30,18 +30,18 @@
 #include "ck_tile/ops/gemm/block/block_wp_asmem_bsmem_creg_v1_custom_policy.hpp"
 #include "ck_tile/ops/gemm/kernel/batched_gemm_kernel.hpp"
 #include "ck_tile/ops/gemm/kernel/gemm_kernel.hpp"
-#include "ck_tile/ops/gemm/kernel/gemm_multi_d_kernel.hpp"
 #include "ck_tile/ops/gemm/kernel/gemm_multi_abd_kernel.hpp"
+#include "ck_tile/ops/gemm/kernel/gemm_multi_d_kernel.hpp"
 #include "ck_tile/ops/gemm/kernel/gemm_tile_partitioner.hpp"
 #include "ck_tile/ops/gemm/kernel/grouped_gemm_kernel.hpp"
 #include "ck_tile/ops/gemm/kernel/streamk_gemm_kernel.hpp"
 #include "ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp"
 #include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_base.hpp"
+#include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_async.hpp"
+#include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_async_default_policy.hpp"
 #include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_v3.hpp"
 #include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_v4.hpp"
 #include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_v4_default_policy.hpp"
-#include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_async.hpp"
-#include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_async_default_policy.hpp"
 #include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_v5.hpp"
 #include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_comp_v5_default_policy.hpp"
 #include "ck_tile/ops/gemm/pipeline/gemm_pipeline_ag_bg_cr_mem.hpp"
@@ -72,5 +72,7 @@
 #include "ck_tile/ops/gemm/warp/warp_gemm_smfmac_impl.hpp"
 #include "ck_tile/ops/gemm/warp/warp_wmma_gemm.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/gemm_quant.hpp
+++ b/include/ck_tile/ops/gemm_quant.hpp
@@ -3,8 +3,8 @@

 #pragma once

-#include "ck_tile/ops/gemm_quant/block/block_universal_gemm_as_aquant_bs_cr.hpp"
 #include "ck_tile/ops/gemm_quant/block/block_universal_gemm_ar_flatbr_bquant_cr.hpp"
+#include "ck_tile/ops/gemm_quant/block/block_universal_gemm_as_aquant_bs_cr.hpp"
 #include "ck_tile/ops/gemm_quant/block/block_universal_gemm_as_bs_bquant_cr.hpp"
 #include "ck_tile/ops/gemm_quant/kernel/gemm_quant_kernel.hpp"
 #include "ck_tile/ops/gemm_quant/kernel/grouped_gemm_quant_kernel.hpp"
@@ -15,11 +15,13 @@
 #include "ck_tile/ops/gemm_quant/pipeline/gemm_bquant_pipeline_ag_bg_cr_base.hpp"
 #include "ck_tile/ops/gemm_quant/pipeline/gemm_bquant_pipeline_ag_bg_cr_policy.hpp"
 #include "ck_tile/ops/gemm_quant/pipeline/gemm_bquant_pipeline_ag_bg_cr_v3.hpp"
-#include "ck_tile/ops/gemm_quant/pipeline/gemm_wp_bquant_pipeline_ag_bg_cr_base_policy.hpp"
-#include "ck_tile/ops/gemm_quant/pipeline/gemm_wp_bquant_pipeline_ag_bg_cr_v2.hpp"
 #include "ck_tile/ops/gemm_quant/pipeline/gemm_group_quant_utils.hpp"
 #include "ck_tile/ops/gemm_quant/pipeline/gemm_quant_pipeline_problem.hpp"
+#include "ck_tile/ops/gemm_quant/pipeline/gemm_wp_bquant_pipeline_ag_bg_cr_base_policy.hpp"
+#include "ck_tile/ops/gemm_quant/pipeline/gemm_wp_bquant_pipeline_ag_bg_cr_v2.hpp"
 #include "ck_tile/ops/gemm_quant/pipeline/tile_gemm_quant_traits.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/grouped_convolution.hpp
+++ b/include/ck_tile/ops/grouped_convolution.hpp
@@ -12,5 +12,7 @@
 #include "ck_tile/ops/grouped_convolution/utils/transform_conv_bwd_weight_to_gemm.hpp"
 #include "ck_tile/ops/grouped_convolution/utils/transform_conv_fwd_to_gemm.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/image_to_column.hpp
+++ b/include/ck_tile/ops/image_to_column.hpp
@@ -7,5 +7,7 @@
 #include "ck_tile/ops/image_to_column/pipeline/block_image_to_column_problem.hpp"
 #include "ck_tile/ops/image_to_column/pipeline/tile_image_to_column_shape.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/layernorm2d.hpp
+++ b/include/ck_tile/ops/layernorm2d.hpp
@@ -10,5 +10,7 @@
 #include "ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_pipeline_two_pass.hpp"
 #include "ck_tile/ops/layernorm2d/pipeline/layernorm2d_fwd_traits.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/norm_reduce.hpp
+++ b/include/ck_tile/ops/norm_reduce.hpp
@@ -7,5 +7,7 @@
 #include "ck_tile/ops/norm_reduce/block/block_norm_reduce_problem.hpp"
 #include "ck_tile/ops/norm_reduce/thread/thread_welford.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/permute.hpp
+++ b/include/ck_tile/ops/permute.hpp
@@ -6,5 +6,7 @@
 #include "ck_tile/ops/permute/kernel/generic_permute_kernel.hpp"
 #include "ck_tile/ops/permute/pipeline/generic_petmute_problem.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/reduce.hpp
+++ b/include/ck_tile/ops/reduce.hpp
@@ -11,5 +11,7 @@
 #include "ck_tile/ops/reduce/pipeline/reduce2d_problem.hpp"
 #include "ck_tile/ops/reduce/pipeline/reduce2d_shape.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/rmsnorm2d.hpp
+++ b/include/ck_tile/ops/rmsnorm2d.hpp
@@ -11,5 +11,7 @@
 #include "ck_tile/ops/rmsnorm2d/pipeline/rmsnorm2d_fwd_pipeline_two_pass.hpp"
 #include "ck_tile/ops/rmsnorm2d/pipeline/rmsnorm2d_fwd_traits.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/smoothquant.hpp
+++ b/include/ck_tile/ops/smoothquant.hpp
@@ -10,5 +10,7 @@
 #include "ck_tile/ops/smoothquant/pipeline/smoothquant_pipeline_problem.hpp"
 #include "ck_tile/ops/smoothquant/pipeline/smoothquant_pipeline_two_pass.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/softmax.hpp
+++ b/include/ck_tile/ops/softmax.hpp
@@ -6,5 +6,7 @@
 #include "ck_tile/ops/softmax/block/block_softmax_2d.hpp"
 #include "ck_tile/ops/softmax/block/block_softmax_2d_problem.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/topk.hpp
+++ b/include/ck_tile/ops/topk.hpp
@@ -6,5 +6,7 @@
 #include "ck_tile/ops/topk/block/block_topk_stream_2d.hpp"
 #include "ck_tile/ops/topk/block/block_topk_stream_2d_problem.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"
--- a/include/ck_tile/ops/topk_softmax.hpp
+++ b/include/ck_tile/ops/topk_softmax.hpp
@@ -8,5 +8,7 @@
 #include "ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_policy.hpp"
 #include "ck_tile/ops/topk_softmax/pipeline/topk_softmax_warp_per_row_problem.hpp"
 #include "ck_tile/ops/common/generic_2d_block_shape.hpp"
+#include "ck_tile/ops/common/load_interleaved_pk_type.hpp"
+#include "ck_tile/ops/common/streamk_common.hpp"
 #include "ck_tile/ops/common/tensor_layout.hpp"
 #include "ck_tile/ops/common/utils.hpp"