playhorizon's CI runner has CPUs 4-7, and remote.sh pinned agents' runs there
too: CPUs 4-7 sat at 100% while 0-3 idled, and window harnesses in pull
requests' parity jobs lost ticks. Runs share a lock per CPU set; an
--exclusive run (a benchmark) holds it alone. world_reopen also waits for
the world to load before its first count, as after each reopen.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The command ran on after remote.sh was stopped or its connection dropped,
holding the name's lock, and a game or daemon it left behind outlived the
run. It now runs in a scope named after the run (remote-<name>.scope),
stopped when the command ends, when the ssh session ends first, and before
the next run under that name.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The host fetches the commit from GitHub (nothing else is sent), keeps one
checkout and target per worktree, runs the command niced in its CI slice on
CPUs 4-7 with scratch homes, prepares vanilla data for harness commands as
CI does, and streams the output back. One run per name at a time; no lock
leaks into daemons the run starts.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>