perf(v4): POST at a boot's pace -- v3's per-suite put-back, and a lookup that rejects by two bytes (#189)

Ruled (Captain Bob, 2026-11-14): "Find the TCG cost first". v3's whole POST took the amd64
accept from about 3 m 15 s to 8 m 45 s, and aarch64 past its 570 s (logs/20261114-091651,
STOPPED, committed). Profiled on the host (gprof, mesh-boot): 87% of the time in POST's
lines, about 280,000 opcodes a line -- the interpreter's dictionary lookup.

- POST puts the dictionary back before each of v3's suites, as v3's run_test_suite does
  ("to remove test-created words"): a mark, (POST), forgotten and made again
  (V4_POST_SUITE). Only the dictionary: SCR, BASE and the rest stay as the cases left them.
- (LOOKUP) tries each entry by its count and first character first -- one fetch of the
  name's first cell, masked -- and only then its hidden flag and the whole name. The answer
  is the same; an entry that cannot be it costs a few opcodes, not some forty.

mesh-boot: 37.7 s -> 15.7 s, POST 990 of 990. make -C v4 test: all v4 tests 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
2026-11-14 13:26:45 +00:00
parent 37469f6c54
commit 281972abb2
11 changed files with 318 additions and 273 deletions
+5
View File
@@ -23,6 +23,11 @@
#define V4_POST_KERNEL 8 /* or'd in: the case needs the kernel's words (capsules/v4/physics.4th, its
requests on port 0) and is not run on a node with no kernel -- the hosted
product (#189 Phase 7: v3's whole POST) */
#define V4_POST_SUITE 16 /* or'd in: the first case of one of v3's suites. Before it the dictionary is put
back -- what the cases before defined is forgotten, to a mark -- as v3's
run_test_suite does after each suite ("to remove test-created words"): a
dictionary grown by every case before made each lookup longer, and 990
cases did not fit a boot's time under emulation */
typedef struct {
const char *name;