Robert Allan JamesandClaude Opus 5.5
acea36ea55
- Investigation (riscv64): completions are slow, not lost -- up to 15 ms, while the ~1M-iteration bound was ~14 ms of guest time. Each timeout left the request in the device while the driver forgot it; the next request took its used entry as its own and returned before its own I/O finished, and so did every later one (394 in one boot), rewriting descriptors the device was still reading. - Fix: VBLK_TIMEOUT_MS (2 s) on the raw hardware counter; each request waits for exactly one more used entry; a timed-out request stays outstanding and its completion is drained before anything is reused, or the next request is refused untouched. Counters for timeouts, late drains and busy refusals. - After the fix, all three ISAs: 0 timeouts, 0 unaccounted completions over 1,912 I/Os; riscv64 slowest completion 7.7 ms (was 15.1). - Standard sequence on amd64/aarch64/riscv64: hash, G, ledger unchanged; 0 timeouts in the riscv64 log. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
experiments/
Design of Experiments (DoE) campaign data and analysis for the LithosAnanke kernel.
bare_metal/— the bare-metal DoE experiment: blind full-factorial 2⁴ runs measuring the L8 Jacquard mode selector's effect on the Steady-State Machine across amd64/aarch64/riscv64. Seeexperiments/bare_metal/README.mdfor the full directory layout (runs/,latest/,analysis/{charts,tables,report}).
Hosted-VM DoE campaigns (factorial/heartbeat/ablation studies) live under
docs/working/experiments/ instead — this top-level directory is
specifically the kernel/bare-metal side.