From 9103661098dc2e69013c2e3e126ab57b84840715 Mon Sep 17 00:00:00 2001 From: Yangwook Date: Thu, 4 Jun 2026 23:09:14 -0700 Subject: [PATCH] gqa(adr): mark DDD-0060 as synced in ADR-0060 SB (EN+KO) DDD-0060 was rewritten to the two-kernel composite-hybrid design, so SB item 1 ('DDD not yet synced') is stale. Update EN + KO mirror to record the DDD as synced (ADR authoritative, DDD = impl how-to). Docs only. Co-Authored-By: Claude Opus 4.8 (1M context) --- .../ADR-0060-algo-gqa-fused-attention-ahbm-ko.md | 9 ++++----- .../ADR-0060-algo-gqa-fused-attention-ahbm.md | 12 +++++------- 2 files changed, 9 insertions(+), 12 deletions(-) diff --git a/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm-ko.md b/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm-ko.md index 65226cb..2db6016 100644 --- a/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm-ko.md +++ b/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm-ko.md @@ -913,11 +913,10 @@ exercise하려 존재하고, `tl.trans`는 reshape-not-transpose이며, `bf16` 검증 패스가 필요하다. (묻지 않고) 작업 합의에 따라 여기 기록한다 — 권고는 내가 예측한 기본값이며; 검토 시 수정. -1. **DDD-0060이 아직 미동기화.** Detailed Design Document는 여전히 옛 - `tl.load_async` 더블버퍼 경로와 primitive `tl.dot` 내부 루프를 기술한다(그 - §4.3/§5/§10). 하이브리드(composite GEMM, lazy load, K 사전 transpose)로 - 갱신해야 한다. DDD가 파생 how-to이고 큰 재작성이라 *검토용으로 남김*; ADR이 이제 - 권위 있는 기록이다. **권고:** 구현 시작 전 후속으로 DDD 동기화. +1. **DDD-0060 동기화 완료 (이번 개정).** Detailed Design Document를 두 커널 + composite-hybrid + CUBE-Group SP 설계에 맞춰 재작성함(file plan, placement, + phase plan P1–P8, verification, risks). ADR이 여전히 권위 있는 설계 기록이고; + DDD는 구현 how-to로서 rationale는 여기(및 §B)를 참조한다. 2. **composite GEMM의 K operand 방향.** Q·Kᵀ는 `b = [d, TILE]`가 필요하나 KV cache는 K를 `[S_rank, d]`로 저장한다. `tl.trans`는 메타데이터 전용이고 diff --git a/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm.md b/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm.md index 9154ab8..675d87d 100644 --- a/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm.md +++ b/docs/adr-proposed/ADR-0060-algo-gqa-fused-attention-ahbm.md @@ -1011,13 +1011,11 @@ design; each needs a verification pass during implementation. Recorded here (rather than asked) per the working agreement — the recommendation is my predicted default; revise on review. -1. **DDD-0060 is not yet synced.** The Detailed Design Document still - describes the old `tl.load_async` double-buffer path and primitive - `tl.dot` inner loop (its §4.3/§5/§10). It must be updated to the hybrid - (composite GEMMs, lazy load, K pre-transposed). *Left for review* - because the DDD is a derived how-to and a large rewrite; the ADR is now - the authoritative record. **Recommend:** sync DDD as a follow-up before - implementation starts. +1. **DDD-0060 synced (this revision).** The Detailed Design Document has + been rewritten to match the two-kernel composite-hybrid + CUBE-Group SP + design (file plan, placement, phase plan P1–P8, verification, risks). The + ADR remains the authoritative design record; the DDD is the + implementation how-to and points back here (and to §B) for rationale. 2. **K operand orientation for the composite GEMM.** Q·Kᵀ needs `b = [d, TILE]`, but the KV cache stores K as `[S_rank, d]`. `tl.trans` is