biondizzle/nvfp4-megamoe-kernel
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
1,057 Commits 1 Branch 17 Tags
df10378bb58fdf043a5adede43b7d024372d3edf
Commit Graph

1057 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
biondizzle
df10378bb5 D1.4: Fix regression test for un-normalized O output (D5a) 2026-05-24 15:13:16 +00:00
biondizzle
423d97b094 D1.4: Guard LSE computation with const_expr(not normalize) - fixes BF16 type mismatch in regression test 2026-05-24 15:11:39 +00:00
biondizzle
d8e2a8f33e D1.4: Switch k_sub from cutlass.range to Python range (unrolled at trace time) 2026-05-24 15:10:28 +00:00
biondizzle
2f623a3f4b D1.4: Fix tTMrO placeholder - define only inside const_expr block 2026-05-24 14:23:22 +00:00
biondizzle
a8672c20b3 D1.4: Use cutlass.range loop for k_sub (reduce IR), guard O rescale with const_expr(n_kv_tiles>1) 2026-05-24 14:22:45 +00:00
biondizzle
402dd4567b Fix: add cutlass import to test_d1_qk512 2026-05-24 14:20:32 +00:00
biondizzle
04b8ca43ed Fix: add cpasync import to test_d1_qk512 2026-05-24 14:20:01 +00:00
biondizzle
e0cd810d39 D1.4: Add hd=512 QK-only and standalone test for compilation debugging 2026-05-24 14:19:26 +00:00
biondizzle
090acfc0ce D1.4: Reduce pv_n_tile to 128 for hd=512 to fit SMEM budget (192KB) 2026-05-24 08:07:32 +00:00
biondizzle
bc8331c9eb D1: Unrolled k_sub path (hardcoded k_sub=0,1) to avoid cutlass.range IR explosion 2026-05-24 07:03:14 +00:00
biondizzle
8cff68a28f D1: Use cutlass.range for k_sub loops (CuTeDSL immutable handle) 2026-05-24 06:43:30 +00:00
biondizzle
14c9000997 D1: Fix kvh scoping - define before loops, consume V via pipeline 2026-05-24 06:42:26 +00:00
biondizzle
553ee7be57 D1: Fix kvb→kvh typo in PV GEMM 2026-05-24 06:41:25 +00:00
biondizzle
9c0dbab280 D1: Remove qh.commit() - pipeline handles commit internally 2026-05-24 06:40:10 +00:00
biondizzle
5c267fd2ad D1: TMA producer uses acquire_and_advance + commit (no wait_and_advance) 2026-05-24 06:38:15 +00:00
biondizzle
3e00c8e1bd D1: Use same pipeline API as working code (acquire_and_advance) for k_sub path 2026-05-24 06:36:19 +00:00
biondizzle
fcc69a5c56 D1: Add PipelineState for k_sub TMA path 2026-05-24 05:02:17 +00:00
biondizzle
22fedc4ed9 D1: Fix pipeline API for K sub-tile path (producer_acquire/commit) 2026-05-24 04:59:41 +00:00
biondizzle
e93dabe43c D1: K sub-tile MMA path using pipeline barriers 2026-05-24 04:57:08 +00:00
biondizzle
fd28718483 D1: Fix TMA copies in k_sub path (no mbarrier, use cp_async wait) 2026-05-24 04:53:46 +00:00
biondizzle
170b483c2f D1: Add K sub-tile loop for hd=512 (const_expr guarded, hd≤256 path unchanged) 2026-05-24 04:51:51 +00:00
biondizzle
bc5240c740 D1: Debug TMA partition shapes at hd=512 2026-05-24 04:43:12 +00:00
biondizzle
2e732ce3a7 D1: K sub-tiling - qk_mma_tiler K-dim = k_tile=256, SMEM fits at hd=512 2026-05-24 04:41:12 +00:00
biondizzle
4564a264db Docs: Update STAGE_D.md, README.md status for D1 hd≤256 milestone 2026-05-24 04:32:43 +00:00
biondizzle
085d72ea8f D1: Full test with TMEM-P at hd=64,128,256,512 2026-05-24 04:07:40 +00:00
biondizzle
2b3435f97c D1: Remove debug prints, clean up 2026-05-24 04:06:26 +00:00
biondizzle
38c6486fc7 D1: const_expr for sP layout selection (CuTeDSL) 2026-05-24 04:05:17 +00:00
biondizzle
a945edea79 D1: Python if for sP layout (trace-time, not MLIR) 2026-05-24 04:04:27 +00:00
biondizzle
955b023164 D1: Tiny 4-mode sP placeholder for TMEM-P path 2026-05-24 04:03:28 +00:00
biondizzle
762f45a8a2 D1: Conditional sP allocation (saves 64KB SMEM for TMEM-P at hd=256) 2026-05-24 04:02:02 +00:00
biondizzle
2184c79732 D1: Fix sP dummy allocation 2026-05-24 04:00:19 +00:00
biondizzle
44d96f2cf8 D1: Skip sP allocation when use_smem_p=False (saves 64KB at hd=256) 2026-05-24 03:59:27 +00:00
biondizzle
8f9b6518f1 D1: Fix syntax (separate kv_stage line) 2026-05-24 03:58:12 +00:00
biondizzle
6597a1cd16 D1: Reduce kv_stage to 1 at hd>128 to avoid SMEM overflow 2026-05-24 03:55:44 +00:00
biondizzle
39367265e5 D1: FIX qk_mma_tiler K-dim = head_dim (was hardcoded to 64, broke hd>64) 2026-05-24 03:53:19 +00:00
biondizzle
13675a12f8 D1: Print qk_ik in _setup 2026-05-24 03:51:40 +00:00
biondizzle
58ed4f5e8b D1: Add more debug prints (QK/PV mode2 sizes) 2026-05-24 03:49:55 +00:00
biondizzle
96a52e63e9 D1: SMEM-P test at hd=128 2026-05-24 03:48:37 +00:00
biondizzle
ebde1d67fd D1: Add sP shape debug print 2026-05-24 03:46:27 +00:00
biondizzle
0f52c3453c D1: Fix SMEM-P (coordinate store), LSE (FP32), add TMEM-P-only test 2026-05-24 03:27:14 +00:00
biondizzle
52570d94cb D1: Fix SMEM-P - coordinate-indexed store (replaces make_tiled_copy_C) 2026-05-24 03:24:44 +00:00
biondizzle
70c9d93d28 feat: SMEM-P make_tiled_copy_C + zero-fill dest tensor 2026-05-24 03:23:53 +00:00
biondizzle
86b7ff8b36 Merge branch 'master' of ssh://sweetapi.com:2222/biondizzle/nvfp4-megamoe-kernel 2026-05-24 03:23:22 +00:00
biondizzle
e1a5077fda D1: LSE diagnostic at various hd 2026-05-24 03:23:16 +00:00
biondizzle
7cf9704584 feat: SMEM-P using make_tiled_copy_C(qk_mma) approach 2026-05-24 03:22:57 +00:00
biondizzle
efe6f4dff8 D1: Add diagnostic test (TMEM-P vs SMEM-P at various hd) 2026-05-24 03:22:23 +00:00
biondizzle
847372976f Merge branch 'master' of ssh://sweetapi.com:2222/biondizzle/nvfp4-megamoe-kernel 2026-05-24 03:21:06 +00:00
biondizzle
7142e459cc D1 test: compare un-norm O + norm using ref row_sum + LSE verification 2026-05-24 03:21:01 +00:00
biondizzle
fe47a5f882 fix: LSE type mismatch Float32→BFloat16 2026-05-24 03:20:26 +00:00
biondizzle
791cd8b9c7 merge: keep our fmha.py (coordinate-indexed SMEM-P + epilogue_tma_store) 2026-05-24 03:19:52 +00:00