1271 Commits

Author	SHA1	Message	Date
Qianfeng Zhang	d8f0862ff8	Tiny update in GetMaxVectorSize()	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	d0fab4c34c	Rename and add scripts for testing hdim96	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	cb6fef75ca	Enable hdim96 instances	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	69c7921dfa	Fix with regard to define stride in MakeKLdsBlockDescriptor()	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	8590f4d71c	Update in the implementation of GetAlignmentQ/GetAlignmentK/GetAlignmentV	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	346c667470	Further correction with regard to using n0_loops and k1_loops	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	8120a86ce6	Add kN0Sub to separate the n0_loop and k1_loop tile size for more flexible tuning	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	84949d4812	Simplify the codes in block_gemm	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	9e252c1ab7	Further clarification in using kSubQKHeaddim and kQKHeaddim	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	b5178551de	Clarify the using of kSubQKHeaddim and kQKHeaddim	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	e4e22cb2d9	Simplifying the codes with regard to k_lds_wite_windows and k_lds_read_windows in the pipelines	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	6ac36b9459	Tiny fix in GetQKBlockGemm	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	ff54459c23	Enable the using of WarpTile-32x32x16 and add scripts to verify	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	c4fc7b28c8	Add static_assert and comments in the with_softmax pipelines	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	c98688d5ad	Force both Gemm0 and Gemm1 to use mfma-16x16x32 on gfx950	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	fa5b077b91	Use 16x16x32 for Gemm1 on MI350 and adjust the NumPrefetchK for with_softmax trload pipeline	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	311fb8a379	Add hstu_attention_api.hpp to explicitly mark the API interfaces and update REAMD.md	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	cf2172b580	Use explicit partition_index to ensure warp_id is allocated on vpgr when accessing LDS tile_window	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	45590fb8b1	Remove useless codes in the two trload pipelines	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	ec9d1fe253	Separate Traits from Problem while being used for defining the pipeline	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	7655ffd2fb	Remove the k_element_func and v_element_func from the pipeline since they are not used	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	2ebe159050	Update to the two trload pipeline to load whole Q-tile once through LDS on mi350	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	0253442ef9	Simplify the codes in block_gemm_areg_bsmem_creg_v2_hack_1	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	0e1444a73c	Simplify the codes in block_gemm_areg_bsmem_trload_creg	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	76b515e092	[Performance] Change the tile settings for mi350/trload no_softmax pipeline to enable to use mfma-16x16x32 for Gemm-1	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	f608f29401	Improve the softmax+trload pipeline by using kN0=64 and prefetch only two k tiles	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	7d9a95dbe0	Tiny fix in trload with_softmax/no_softmax pipeline	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	e7bfee6043	Improve both the with_softmax and no_softmax pipelines	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	3e122a551d	Add kUseTrLoad = false in non-trload pipeline	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	660604635d	Clarifying the using of CK_TILE_HOST and CK_TILE_HOST_DEVICE trying to save compiling time	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	3565656005	Change in updating max_uih_seqlen in the example	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	639332b5de	Use supplement_array_by_last_element(num_targets, ) in example	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	19a887cffa	Use supplement_array_by_last_element() in example to simplify the codes	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	9537b958ab	Update to README.md	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	2ae759c5b3	Add scripts for testing the using of separate sequence lengths for k/v	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	a926db031a	Support separate sequence lengths for q and kv	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	bcefc197d8	Use separate pipelines for using or not-using softmax situations	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	4b3901e989	Implementation of hstu attention pipeline using trload for v on mi350	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	5db875df2c	Fix in the comments	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	563229a9e4	Update to gemm_0's CBlockDistribution encoding so that it is compatible with gemm_1's ABlockDistribution encoding	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	4e9d4f7487	Using separate tile settings for no-softmax and with-softmax hstu attention situations	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	91afdfee40	Update to benchmark scripts to consider for using softmax	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	be3b27edd3	Add support of softmax in hstu attention	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	ec4f174ac4	Add template parameter to gemm_0 MakeCBlockTile() for the need of defining PcompBlockTileType	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	da50eea674	Move scaling by attn_scale to inside the main-loop	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	3639fb8e38	Let IsTokenPairInsideMask() return bool type	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	a71c996049	Add instances to consider for adding softmax support	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	0abb52004a	Remove K0 from tile setting since it is not used	2026-06-23 09:27:58 +00:00
Qianfeng Zhang	de198549ad	Change to pipeline so that it is easier to add support of using softmax	2026-06-23 09:27:58 +00:00
Qianfeng Zhang	7e79736df7	Remove using IGLP method for instruction scheduling for kUseLocal true path	2026-06-23 09:27:58 +00:00