perf(v4): the lookup keeps its key in B; POST sends one line per suite, not three (#189)

Captain Bob's "Find the TCG cost first", continued. aarch64 still ran out of its 570 s at 281972ab
(logs/20261114-133155, STOPPED); amd64 passed in 253 s (logs/20261114-132732).

- (LOOKUP): the count-and-first-character key in register B while the entries are walked, and the
  xt kept on the stack until an entry passes it -- about 19 opcodes for an entry that cannot be the
  word, from 28. The answer is the same.
- POST: a suite begins with one line, DECIMAL FORTH DEFINITIONS FORGET (POST); its mark is made by
  the state line of its first case run (a skipped kernel case passes it on).

mesh-boot: 15.7 s -> 12.2 s (37.7 s before this work), POST 990 of 990. make -C v4 test: all passed.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RzZgrf7V7Zk2iczBXDRbm
This commit is contained in:
Claude
2026-11-17 13:11:47 +00:00
parent 281972abb2
commit 5b752693ae
5 changed files with 28 additions and 8 deletions
Binary file not shown.
+6
View File
@@ -0,0 +1,6 @@
# STOPPED -- timed out (development boot, not the protocol)
`timeout 570 scripts/nightly.sh accept aarch64` on 281972ab (#189 Phase 7: v3's whole POST, 990 cases, after the
per-suite put-back and the lookup's two-byte test; amd64 on that commit 253 s). The 570 s ran out in POST: the log
ends in the capsules' parity lines. aarch64 took 460-473 s with the 539-case POST (logs/20261113-170343,
logs/20261113-191606). Followed by a faster lookup and fewer POST lines.