c9448aca03
Add debug prints for SMEM-P partition layouts
2026-05-23 19:13:07 +00:00
b4b11db0fa
Fix SMEM-P: use BF16 copy atom and BF16 source with QK C-fragment layout
2026-05-23 19:12:13 +00:00
7b65adf7a3
Fix SMEM-P copy: use tcgen05.copy.St32x32bOp with Float32 and copy from rP_words (Float32) not rP_bf16
2026-05-23 19:11:08 +00:00
dc2c9ffb92
Update STAGE_D.md with current action plan - starting NVFP4-0 verification and D1.3 validation on B200
2026-05-23 19:09:56 +00:00
f0f78b804c
📋 Update STAGE_D.md: D1.3 ✅ SOLVED, D1.4 ✅ IMPLEMENTED, D1.5 🟡 complex refactor, checklist updated
2026-05-23 18:37:53 +00:00
d995cd0c5c
🎉 Mark D1.3 as SOLVED! SMEM-P rank mismatch fixed, enables hd>64 support
2026-05-23 18:26:15 +00:00
4bf3c435b5
Fix rP scope issue: use rP_bf16.iterator instead of rP.iterator
2026-05-23 09:36:22 +00:00
431b8e0abe
Fix duplicate else: line in SMEM-P block
2026-05-23 09:35:47 +00:00
d2bb02a331
SMEM-P: Use QK C-fragment layout instead of TMEM layout to fix rank mismatch
2026-05-23 09:35:24 +00:00
63c9a5ce82
Fix sP_2d definition for tSMEM_CPYsP
2026-05-23 09:34:50 +00:00
c7006b0969
Remove debug print lines referencing deleted sP_2d
2026-05-23 09:34:09 +00:00
b09c432942
Remove duplicate sP_2d line causing indentation error
2026-05-23 09:33:40 +00:00
518dce37f0
SMEM-P: Implement rank mismatch fix by reshaping source tensor
2026-05-23 09:33:24 +00:00
a3659c581d
Update STAGE_D.md checklist with current progress and lessons learned
2026-05-23 09:27:48 +00:00
c9dda47971
Add more debug prints for sP shapes
2026-05-23 09:26:30 +00:00
2283de1cfc
Add debug prints to SMEM-P path to understand rank mismatch
2026-05-23 09:25:48 +00:00
7c350e6a18
Fix SMEM-P copy rank mismatch (use rP_bf16 directly instead of group_modes)
2026-05-23 09:21:13 +00:00
cb2849bff5
D1.3: Implement SMEM-P path (write P to SMEM via tiled_smem_copy instead of zeroing sP)
2026-05-23 09:20:37 +00:00
2c36cd0d32
Stage D1: Multi-PV-tile support for hd>256 (tcgen05 MMA max N=256)
2026-05-23 09:04:01 +00:00
f556060ddf
Fix v_fmha layout to use pv_n_tile instead of head_dim for multi-PV-tile support
2026-05-23 09:02:01 +00:00
f1ad264da6
D1.4: Add pv_n_tile and n_pv_tiles for multi-PV-tile support (tcgen05 MMA max N=256)
2026-05-23 09:00:18 +00:00
2083204a90
diag: add 2-CTA check + fix LayoutEnum in MMA test
2026-05-23 08:45:26 +00:00
02daa63f67
fix: remove bad import in NVFP4 diag
2026-05-23 08:44:37 +00:00
401e24768a
fix: import ceil_div in quantize.py (was NameError at runtime)
2026-05-23 08:40:24 +00:00
28724b8c18
fix: use quantize_activation_nvfp4 in diag
2026-05-23 08:39:12 +00:00
690be100aa
fix: use correct API for NVFP4-0 diag (sf_vec_size + mma_tiler_mn)
2026-05-23 08:38:19 +00:00
0fc43eb624
fix: use Sm100BlockScaledPersistentDenseGemmKernel in diag
2026-05-23 08:30:43 +00:00
e6c9616eba
fix: use randint+view for FP4/FP8 tensors in diag
2026-05-23 08:29:16 +00:00
96567ea64d
fix: NVFP4-0 diag script — import SF_VEC_SIZE from quantize.py
2026-05-23 08:28:13 +00:00
f6f4ce8ec2
diag: NVFP4-0 primitive verification script
2026-05-23 08:26:56 +00:00
241b49b1ee
docs: add NVFP4 precision roadmap to STAGE_D.md (3 honest buckets + speculative bucket)
2026-05-23 07:39:09 +00:00
73fa8a2b70
shit carmine left dangling
2026-05-23 06:55:22 +00:00
bd2da14ca6
D1.2: TMEM budget verified on B200. Split-PV mandatory at hd=512 (MMA max N=256)
2026-05-23 06:43:01 +00:00
a951a95276
D1.2: fix probe for hd=512 (MMA max N=256, use pv_n_tile)
2026-05-23 06:41:42 +00:00
97899b42a3
D1.2: simplify TMEM budget probe, fix printf args
2026-05-23 06:40:55 +00:00
640bae606e
fix: cuda.CUstream import
2026-05-23 06:40:05 +00:00
595c9d677c
D1.2: TMEM budget probe using @cute.jit for MLIR context
2026-05-23 06:39:27 +00:00
1a781447cb
fix: remove unused v_fmha_layout from probe
2026-05-23 06:38:08 +00:00
3490601e02
D1.2: TMEM budget probe with real tensor major modes
2026-05-23 06:37:34 +00:00
8b351bd871
fix: OperandMajorMode.MN not .M
2026-05-23 06:36:39 +00:00
5b3bd4876c
fix: typo + OperandMajorMode for TMEM budget probe
2026-05-23 06:35:55 +00:00
677245fc5c
fix: LayoutEnum import from cutlass.utils
2026-05-23 06:35:03 +00:00
bd968db5d4
fix import path for tcgen05
2026-05-23 06:34:30 +00:00
afec55ad72
D1.2: TMEM budget probe script for hd=64,128,256,512
2026-05-23 06:33:26 +00:00
5012703bad
fix: add SwiGLU clamping to fused kernel (paper §4.2.3, CG-1)
...
The fused SwiGLU kernel stored swiglu_limit but never applied it.
Paper §4.2.3: gate capped at swiglu_limit, linear clamped to [-limit, +limit].
Non-fused reference path already applies clamping correctly.
Fix: add fmin/fmax clamping in FP32 before BF16 conversion.
2026-05-23 06:32:54 +00:00
580d2f6999
STAGE_D.md: restructure with correctness gaps, TMEM budget, execution order
2026-05-23 06:31:37 +00:00
df43c3232d
D1.1: Fix make_fragment_A — use sP for SMEM source pv_mma
2026-05-23 06:04:44 +00:00
80434d0284
D1.1: Fix PV A-operand construction — compile-time branch for TMEM vs SMEM
2026-05-23 06:03:27 +00:00
d36b727898
D1.1: Add SMEM-P path behind use_smem_p flag (stub: zero sP)
2026-05-23 06:01:02 +00:00
bd0b56dddd
D1.0: Replace HEAD_DIM=64 with self.head_dim constructor parameter
2026-05-23 05:55:03 +00:00
