biondizzle 1ad243f095 CRITICAL FIX: keep GMEM iteration dim free in tBgK/tVgV slice ...

The slice (None,0,None,0) was hardcoding the GMEM iteration dim to 0,
meaning TMA always loaded K/V from tile 0 regardless of kt.
Changed to (None,None,None,0) to keep gmem_iter free,
then index with (None, kt, None) in the TMA copy loop.

This is the root cause of multi-tile failure: TMA was always reading
the first 128 tokens for ALL KV tiles.

2026-05-22 15:52:06 +00:00

..

archive

Restructure: cutedsl/ -> dsv4/ with proper layering

2026-05-21 17:30:44 +00:00

e2e

Restructure: cutedsl/ -> dsv4/ with proper layering

2026-05-21 17:30:44 +00:00

integration

Restructure: cutedsl/ -> dsv4/ with proper layering

2026-05-21 17:30:44 +00:00

unit

Revert "debug: test 12w identity softmax with n=256 to verify multi-tile pipeline"

2026-05-22 10:25:48 +00:00

fmha_v3_stage_c_example1.py

CRITICAL FIX: keep GMEM iteration dim free in tBgK/tVgV slice

2026-05-22 15:52:06 +00:00

native_stage_c_patch.py

more stuff

2026-05-22 08:57:38 +00:00

requirements.txt

test: add standalone layer 0 comparison test (no vLLM, no Docker)

2026-05-16 02:13:18 +00:00

run_test.sh

fix: use setup.py install for CUTLASS extension build

2026-05-16 02:21:17 +00:00