From 2327aa83ba340d53d4d485d6d7b8edd47f5338c9 Mon Sep 17 00:00:00 2001 From: igneum-labs <337424239+igneum-labs@users.noreply.github.com> Date: Wed, 7 Oct 2026 08:08:01 +0000 Subject: [PATCH] attack-pass F6: O-1.14 takes the rented 2019-class CPU host route, two-hour cap Co-Authored-By: Claude Fable 5.1 --- docs/analysis/attack-pass-2026-10.md | 6 ++++++ 1 file changed, 6 insertions(+) diff --git a/docs/analysis/attack-pass-2026-10.md b/docs/analysis/attack-pass-2026-10.md index b53736d6..102c6fc1 100644 --- a/docs/analysis/attack-pass-2026-10.md +++ b/docs/analysis/attack-pass-2026-10.md @@ -132,6 +132,12 @@ and shadow block), plus the O-1.14 laptop run (the Windows `igneum-pow` build on the miner's side, not the verifier's; `dr736` already FAILs at 10.51 ms cold one-core, but it is not the shipping class). Gate: the worst program under 10 ms cold on the half-core proxy and on the laptop. +O-1.14 route (coordinator, 7 October 2026, 10:1x UK): no US laptop is due, so the laptop run is replaced by a +rented 2019-class CPU host through the fleet agent, capped at two hours of rent; the Linux `igneum-pow` from the box +(the same binary as the proxies) runs `bench --warps 50` for v2, mx8, mx8+sh256x27, dr368 and dr736 (the +known-fail) on it; INCOMPLETE with the numbers so far if the cap lands first. The Windows exe was also built on the +box for the day a laptop appears (1,009,675 bytes, sha256 fbed7538...e6c9). + Result (average, verified on the box): class v4 `mx8+sh256x27` runs 4.90 to 5.06 ms per warp cold on one EPYC 9454P core (nice 19, taskset), 8.23 ms on the half-core proxy (both SMT siblings busy). Under 10 ms. Status RUNNING: the 10^5-program worst-case search and the O-1.14 laptop relay run are owed before the row reads PASS.