kernbench2

Files

T

ywkang 4a55ae5c0b gqa(adr-0065): P6 — close dispatch-ratio loop (ADR-0064 Test #9 direction)

Extend the opt2 R-sensitivity test to assert the opt3/opt2 dispatch ratio increases as R decreases (4.20x -> 5.22x -> 5.45x at R=0.25/0.0625/0.03125) — the cost is FIXED-dominated (command-count-driven), per ADR-0064 Test #9. Test-only; no production change (the P0 cost model + P2 recipe already make the ratio measurable).

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

2026-06-10 21:04:45 -07:00

test_gqa_decode_mc.py

gqa: tile-granular Ring KV (P3c) + rename to gqa_attention_* + ADR-0060/62/63/64 → Accepted

2026-06-10 16:20:00 -07:00

test_gqa_decode_opt2.py

gqa(adr-0065): P6 — close dispatch-ratio loop (ADR-0064 Test #9 direction)

2026-06-10 21:04:45 -07:00

test_gqa_decode_sp.py

gqa: tile-granular Ring KV (P3c) + rename to gqa_attention_* + ADR-0060/62/63/64 → Accepted

2026-06-10 16:20:00 -07:00

test_gqa_decode.py

gqa: tile-granular Ring KV (P3c) + rename to gqa_attention_* + ADR-0060/62/63/64 → Accepted