Commit Graph

1027 Commits

Author SHA1 Message Date
9ea7f511b0 fix: dummy tma_p reuse tma_q for non-SMEM-P 2026-05-24 03:33:41 +00:00
00c9991396 merge 2026-05-24 03:32:54 +00:00
ae4c501903 merge: keep our fmha.py 2026-05-24 03:32:24 +00:00
edacdae017 fix: use make_tiled_tma_atom_A for P TMA 2026-05-24 03:32:15 +00:00
5e154770de D1: SMEM-P using make_tiled_copy_C + retile (correct CUTLASS pattern) 2026-05-24 03:32:03 +00:00
0f943ab48c feat: SMEM-P via gP→TMA→sP path (register→GMEM→SMEM) 2026-05-24 03:31:23 +00:00
d0aec403e4 D1: Layout diagnostic for SMEM-P 2026-05-24 03:30:29 +00:00
dd7356afc6 D1: Simplified debug test 2026-05-24 03:29:14 +00:00
1f1b16ad07 D1: Add hd=128 debug test 2026-05-24 03:28:26 +00:00
0f52c3453c D1: Fix SMEM-P (coordinate store), LSE (FP32), add TMEM-P-only test 2026-05-24 03:27:14 +00:00
52570d94cb D1: Fix SMEM-P - coordinate-indexed store (replaces make_tiled_copy_C) 2026-05-24 03:24:44 +00:00
70c9d93d28 feat: SMEM-P make_tiled_copy_C + zero-fill dest tensor 2026-05-24 03:23:53 +00:00
86b7ff8b36 Merge branch 'master' of ssh://sweetapi.com:2222/biondizzle/nvfp4-megamoe-kernel 2026-05-24 03:23:22 +00:00
e1a5077fda D1: LSE diagnostic at various hd 2026-05-24 03:23:16 +00:00
7cf9704584 feat: SMEM-P using make_tiled_copy_C(qk_mma) approach 2026-05-24 03:22:57 +00:00
efe6f4dff8 D1: Add diagnostic test (TMEM-P vs SMEM-P at various hd) 2026-05-24 03:22:23 +00:00
847372976f Merge branch 'master' of ssh://sweetapi.com:2222/biondizzle/nvfp4-megamoe-kernel 2026-05-24 03:21:06 +00:00
7142e459cc D1 test: compare un-norm O + norm using ref row_sum + LSE verification 2026-05-24 03:21:01 +00:00
fe47a5f882 fix: LSE type mismatch Float32→BFloat16 2026-05-24 03:20:26 +00:00
791cd8b9c7 merge: keep our fmha.py (coordinate-indexed SMEM-P + epilogue_tma_store) 2026-05-24 03:19:52 +00:00
6313974fba D1.5: Fix SMEM-P - use coordinate-indexed store (same proven pattern) 2026-05-24 03:19:32 +00:00
3fcb7a0a48 feat: SMEM-P with make_tiled_copy_tv + partition_S 2026-05-24 03:19:18 +00:00
153db24be2 D1.5: Always output un-normalized O + LSE (epilogue_tma_store only, no TMEM round-trip normalize) 2026-05-24 03:18:38 +00:00
d68ab348bb feat: SMEM-P using make_tiled_copy_A from PV MMA 2026-05-24 03:16:34 +00:00
b4a985631b fix: fence_proxy not fence 2026-05-24 02:45:01 +00:00
95f0898c64 merge: resolve conflict (keep our version) 2026-05-24 02:44:09 +00:00
228ec3c638 D1.5: Replace broken make_cotiled_copy SMEM-P with coordinate-indexed store 2026-05-24 02:43:42 +00:00
c58ca550ae feat: SMEM-P with make_tiled_copy_tv + manual fill 2026-05-24 02:43:12 +00:00
8faac948fc feat: SMEM-P using make_tiled_copy_tv + logical sP view 2026-05-24 02:42:27 +00:00
eda7d40df2 Merge branch 'master' of ssh://sweetapi.com:2222/biondizzle/nvfp4-megamoe-kernel 2026-05-24 02:41:39 +00:00
952c25e227 D1.5: Use tCtO_fake layout for epilogue_tma_store (needs STAGE dim) 2026-05-24 02:41:32 +00:00
0a980de7ad feat: SMEM-P using make_cotiled_copy (one-row-per-thread) 2026-05-24 02:41:19 +00:00
85eb2bc4bb D1.5: Remove duplicate tTMrO definition (keep unconditional one) 2026-05-24 02:40:40 +00:00
83077db55e merge 2026-05-24 02:40:14 +00:00
86ff386ea8 D1.5: Move tTMrO after O rescale atoms (fix tTMEM_LOADcO reference) 2026-05-24 02:39:18 +00:00
cd223e1b98 fix: reorder tTMrO definition after tTMEM_LOADcO 2026-05-24 02:36:08 +00:00
54e94d44ef fix: tTMrO scoping + restore SMEM-P coordinate write 2026-05-24 02:34:53 +00:00
6ead708c7d D1.5: Move tTMrO def before softmax loop (CuTeDSL scoping) 2026-05-24 02:32:51 +00:00
5a34865062 debug: zero-fill sP to check deadlock 2026-05-24 02:31:12 +00:00
81652629e3 D1.5: Use proven Stage C approach - normalize via TMEM round-trip + epilogue_tma_store 2026-05-24 02:30:25 +00:00
974cddbf7b test: add try/except for SMEM-P coord test 2026-05-24 02:15:07 +00:00
5fd556db63 test: use FmhaKernel for SMEM-P coord test 2026-05-24 01:59:25 +00:00
e50ba7212c test: SMEM-P coordinate verification test 2026-05-24 01:58:32 +00:00
3325d7442c fix: proper v_major from tensor 2026-05-24 01:55:37 +00:00
f2dd36cd4b fix: COL_MAJOR 2026-05-24 01:54:30 +00:00
c815560d64 fix: hardcode v_major for diag test 2026-05-24 01:54:20 +00:00
34fe43a551 diag: use FmhaKernel setup for cotiled test 2026-05-24 01:54:08 +00:00
7a9881ab82 fix: v_major derivation in cotiled diag 2026-05-24 01:53:40 +00:00
a5fbc6cebd diag: simplified cotiled layout test 2026-05-24 01:53:24 +00:00
60ff5e4f53 fix: LayoutEnum import in cotiled diag test 2026-05-24 01:52:58 +00:00