3749 Commits

This Branch

This Branch

All Branches

Author	SHA1	Message	Date
Qianfeng Zhang	6cf17bc827	Miscellaneous small updates and corrections	2026-06-28 13:59:44 +00:00
Qianfeng Zhang	52eff34d21	Add scripts for testing hstu attention fwd with drop-out	2026-06-28 13:59:44 +00:00
Qianfeng Zhang	35234a492d	Fix in kernel and HstuBlockMask interface for kHasDropout == true	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	83676066af	Fix in GetSmemSizeDropout()	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	95d3ecfb14	Update for considering kHasDropout used with minfull_attn_seqlen > 0	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	4250341a92	Fix identified issues for implementing kHasDropout == true	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	c3df01a519	Add instances and enable dispatching for kHasDropout == true	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	124f072158	Add p_drop parameter to example and drop-out logic to the reference code	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	58fbfe766e	Update to kernel and host interface for generating random numbers	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	8a4ec6382d	Add kernel and host interface for generating random numbers	2026-06-28 13:59:43 +00:00
Qianfeng Zhang	d5c872f504	Removing the class derivation to simplify struct HstuAttentionFwdCommonDropoutKargs	2026-06-24 09:03:39 +00:00
Qianfeng Zhang	0720bc48be	Fix using multiplies	2026-06-23 10:20:17 +00:00
Qianfeng Zhang	ad769baea4	Fix make_kernel() template parameters	2026-06-23 09:38:37 +00:00
Qianfeng Zhang	0b0684aff2	Revert "[ck_tile] Add get_partition_index_v2 which uses warp_id in vgpr and to be used by tile_windows on lds-based tensor_view" ... This reverts commit 2b0f3791a6.	2026-06-23 09:31:31 +00:00
Qianfeng Zhang	1805c985a0	Update to example_hstu_attention_fwd.cpp	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	dc1d433351	Rename generate_instances.py to generate_fwd_instances.py	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	503a017c42	Remove the using of kSubQKHeaddim	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	a81f32331f	Add restriction on the relationship between HstuAttention<xxx>Problem and HstuAttention<xxx>TileSettingClass	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	79ebc7479d	Add static_assert() in HstuAttentionFwdTileSettingClass	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	7a67ae4dd3	Update to the comments in reference_hstu_attention_bwd.hpp	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	7d2e575fed	Fix the comments in reference_hstu_attention_fwd.hpp	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	d07b37c097	Remove un-used element-wise functions passed through pipelines' operator() interfaces	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	cc7e216fa6	Rename GetKVBlockGemm to GetPVTBlockGemm	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	06547efd90	Remove the kHasBias==true instances to save building time	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	f41777fbd2	Renaming BUILD_HSTU_FOR_GFX95_ONLY to BUILD_HSTU_FOR_GFX95	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	295136e48b	Update the README.md according to the summary by claude code	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	a1ad9fc312	Fix the using of num_targets[] in run_group_hstu_attention	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	62627db768	Update to the comments in reference_hstu_attention_bwd.hpp	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	08873f0d50	Renaming in the dispatching codes and generate_instances.py scripts	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	c7de3af246	Split hstu_attention_util.hpp into host_util.hpp and kernel_util.hpp	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	89d6f5aa92	Remove un-needed includings from some hpp and cpp files	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	6f4f3eac48	Tiny update in generate_instances.py	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	673207ce59	Fix header file mapping bug in generate_instances.py	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	f73341de37	Some renaming in kernel and pipeline	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	42a3bfbab7	Update and fix for leeked changes and make the scripts be able to test/benchmark kStoreLSE cases	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	b17b41a1e6	Enable the kernel dispatching path from is_training & use_softmax to kStoreLSE	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	4414019296	Add instances and kStoreLSE template in dispatcher class to support outputting lse for fwd training	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	8f83a2841f	Set lse tensor dim strides in example	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	02e5c23f9c	Replace template kUseSoftmax/kStoreLSE by boolean parameters in reference fwd codes to save compiling time	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	bd126618d1	Add support for outputing lse in the example and reference hstu attention forward implementation	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	fa0a9c1656	Add support for preparing lse_dram_window in hstu fwd kernel	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	fd9af72c9f	Kernel use types declared in the problem rather than the pipeline	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	ee5bd0ebba	Tiny simplification with defining the Bias related Kargs	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	f41b0176d3	Add parameters used by storing lse in the fwd and fwd_splitkv_combine kernel to prepare for supporting training	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	270d073c88	Move num_splits/o_acc_ptr/l_acc_ptr out from HstuAttention<xxx>FwdParams struct	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	b2561b88e4	Add kStoreLSE template parameter to the problems	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	2a86bfb6f5	Implement host reference operator for hstu attention backward	2026-06-23 09:28:00 +00:00
Qianfeng Zhang	ec58f92f05	Rename the reference interfaces and the files	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	7d317adf37	Update to MakeLSEaccDramTileDistribution trying to assign more threads to MThreadPerWarp so that block_tile_reduce_sync() work on less KThreadPerWarp	2026-06-23 09:27:59 +00:00
Qianfeng Zhang	30b5d7bd01	Use buffer_view to create lse_acc_dram_naive so that out_of_boundary loading value can be specified (be -inf)	2026-06-23 09:27:59 +00:00