From ae842df5ae90de9e52a14ba99d7a87f41e35e568 Mon Sep 17 00:00:00 2001 From: Richard Loyer Date: Sun, 24 May 2026 17:17:24 +0200 Subject: [PATCH] Chantier 2 clos: CONT low-ROI (opsync gonfle, reshape structurel) -> 3 prioritaire --- docs/02_cont_overhead.md | 2 ++ 1 file changed, 2 insertions(+) create mode 100644 docs/02_cont_overhead.md diff --git a/docs/02_cont_overhead.md b/docs/02_cont_overhead.md new file mode 100644 index 0000000..d46bf60 --- /dev/null +++ b/docs/02_cont_overhead.md @@ -0,0 +1,2 @@ +# Chantier 2 — CONT/copies prefill: ROI faible, non engage +416 CONT = ~17/couche GDN (reshape/permute DeltaNet, qwen3next.cpp). 4.6s mesure OPSYNC=serialise; async (defaut) les copies recouvrent le compute -> part reelle << gonfle. cont l.165 = TODO upstream (CUDA non-contig unary), pas removable safe. prefill deja 46-50, pas de bottleneck CONT prouve. Decision: low-ROI, skip. Le vrai gain prefill = GDN sur HTP (chantier 3), pas tailler des copies.