From a400667ab9e7d1edfee7639d8ec1b8312155dfc2 Mon Sep 17 00:00:00 2001 From: bong-water-water-bong Date: Sat, 14 Nov 2026 22:35:04 -0400 Subject: [PATCH] third_party/llama.cpp: pin to 669178cb0, one-token GDN decode updates the recurrent state in place (~+4 % at 4 streams) 8c12dfa63 -> 669178cb0, two commits: 669178cb0 hrx: one-token GDN decode updates the recurrent state in place (llama.cpp#390) 6d6cb0f32 hrx: strict refuses a misplacement, not a placement the code calls by design (engine #2450, llama.cpp#389) #390 drops the 12.6 MB-per-layer "new state -> cache" copy at Qwen3.8-27B x 4 streams, whose write-back drained during the next weight stream; c4 (-np 4 -kvu) 40.3 -> 42.0 agg tok/s, decode-path KLD identical to base to the last digit. registry/architectures.json records the new pin (registry_build.py --check-pins rc 0). Co-Authored-By: Claude Opus 5.5 --- registry/architectures.json | 2 +- third_party/llama.cpp | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/registry/architectures.json b/registry/architectures.json index 0a661bd11..f6ad38271 100644 --- a/registry/architectures.json +++ b/registry/architectures.json @@ -1,7 +1,7 @@ { "about": "HF architecture -> GGUF architecture and the backends whose code accepts it. Generated by tools/registry_build.py from the pinned sources; do not edit.", "sources": { - "llama.cpp (hrx)": "8c12dfa634ac0040d81ccdc50d41c4e3af5ffb7b" + "llama.cpp (hrx)": "669178cb0a2fae7e3fbdf8d0a49c32aaf414f5dd" }, "counts": { "hrx": 350, diff --git a/third_party/llama.cpp b/third_party/llama.cpp index 8c12dfa63..669178cb0 160000 --- a/third_party/llama.cpp +++ b/third_party/llama.cpp @@ -1 +1 @@ -Subproject commit 8c12dfa634ac0040d81ccdc50d41c4e3af5ffb7b +Subproject commit 669178cb0a2fae7e3fbdf8d0a49c32aaf414f5dd