Merge branch 'amd-develop' into amd-master

4d914af3 · Jun Liu · 223a2abe · 4b798833 · 4d914af3 · 4d914af3
Commit 4d914af3 authored Oct 31, 2024 by Jun Liu
20 changed files
--- a/include/ck_tile/ops/epilogue/default_2d_epilogue.hpp
+++ b/include/ck_tile/ops/epilogue/default_2d_epilogue.hpp
--- a/include/ck_tile/ops/epilogue/dynamic_quant_epilogue.hpp
+++ b/include/ck_tile/ops/epilogue/dynamic_quant_epilogue.hpp
--- a/include/ck_tile/ops/fmha.hpp
+++ b/include/ck_tile/ops/fmha.hpp
--- a/include/ck_tile/ops/fmha/kernel/fmha_fwd_kernel.hpp
+++ b/include/ck_tile/ops/fmha/kernel/fmha_fwd_kernel.hpp
--- a/include/ck_tile/ops/fmha/kernel/fmha_fwd_splitkv_kernel.hpp
+++ b/include/ck_tile/ops/fmha/kernel/fmha_fwd_splitkv_kernel.hpp
--- a/include/ck_tile/ops/fmha/kernel/fmha_fwd_splitkv_tile_partitioner.hpp
+++ b/include/ck_tile/ops/fmha/kernel/fmha_fwd_splitkv_tile_partitioner.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_combine_pipeline.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_combine_pipeline.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_combine_pipeline_default_policy.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_combine_pipeline_default_policy.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_pipeline_qr_ks_vs.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_pipeline_qr_ks_vs.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_pipeline_qr_ks_vs_default_policy.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_fwd_splitkv_pipeline_qr_ks_vs_default_policy.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_problem.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_problem.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qr_ks_vs.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qr_ks_vs.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qr_ks_vs_async.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qr_ks_vs_async.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qr_ks_vs_fp8.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qr_ks_vs_fp8.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qs_ks_vs.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qs_ks_vs.hpp
--- a/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qx_ks_vs_custom_policy.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/block_fmha_pipeline_qx_ks_vs_custom_policy.hpp
--- a/include/ck_tile/ops/fmha/pipeline/tile_fmha_shape.hpp
+++ b/include/ck_tile/ops/fmha/pipeline/tile_fmha_shape.hpp
--- a/include/ck_tile/ops/gemm.hpp
+++ b/include/ck_tile/ops/gemm.hpp
--- a/include/ck_tile/ops/gemm/block/block_gemm_areg_bgmem_creg_v1.hpp
+++ b/include/ck_tile/ops/gemm/block/block_gemm_areg_bgmem_creg_v1.hpp
--- a/include/ck_tile/ops/gemm/block/block_gemm_areg_breg_creg_v1.hpp
+++ b/include/ck_tile/ops/gemm/block/block_gemm_areg_breg_creg_v1.hpp