From 4c610e05a4094100b8e007e53042dfe9a3abf65a Mon Sep 17 00:00:00 2001 From: Robert Allan James Date: Thu, 24 Sep 2026 08:22:57 -0400 Subject: [PATCH] Revert the MU-WAIT-ALL '100 WAIT' change (Captain Bob); keep its logs The fix belongs in the kernel's compudynamics, not in the DoE capsule steering its own scheduling. Logs from the stopped relaunch and the check run stay committed. Co-Authored-By: Claude Opus 5.5 --- FABRIC-3.8.md | 7 +++++++ capsules/multiuser-doe.4th | 2 +- 2 files changed, 8 insertions(+), 1 deletion(-) diff --git a/FABRIC-3.8.md b/FABRIC-3.8.md index 0d425fba..6940430f 100644 --- a/FABRIC-3.8.md +++ b/FABRIC-3.8.md @@ -1389,3 +1389,10 @@ explicit instruction. - No yield word; capsule-only change. - **Check** (`logs/20260924-081340/amd64/`, own boot, images reset after): a cfg-0 trial ran in **115 s against ~210 s before**, `fail=0`, 2/2 dead. + - **REVERTED the same day, Captain Bob:** *"not liking this much ... redistribute work when + hera is the major consumer"*, then *"1, revert the WAIT commit"*. + - `100 WAIT` had the application steer its own scheduling, by faking work so the kernel + would back off. The fix belongs in the kernel's compudynamics (option 1: charge by + share), not in the DoE capsule. + - Only the one capsule line is reverted. The logs and this entry stay, because logs are + never deleted. diff --git a/capsules/multiuser-doe.4th b/capsules/multiuser-doe.4th index ad55c6db..a9d366a9 100644 --- a/capsules/multiuser-doe.4th +++ b/capsules/multiuser-doe.4th @@ -120,7 +120,7 @@ Block 5052 I MU-SEND-WORKER LOOP ; : MU-WAIT-ALL ( -- ) - BEGIN MU-DONE-COUNT @ MU-NW @ < WHILE 100 WAIT REPEAT ; + BEGIN MU-DONE-COUNT @ MU-NW @ < WHILE REPEAT ; Block 5053 : MU-KILL-ALL ( -- )