engineering/lumbda
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions
lumbda/proof
History
russell@unturf.com e700273136 Add superinstructions LOOK+1, LOOK-1 for 25% faster loops 
Fused opcodes: LOOK_ADD1 (lookup + increment) and LOOK_SUB1
(lookup + decrement) emitted directly by compiler for (+ sym 1)
and (- sym 1) patterns. Eliminates one dispatch per loop iteration.

sum-to(50000) ratio improved from 59x to 45x vs Python.
ackermann(3,4) steady at 83x. 571 tests green.

Also defines LOOK_LOOK, CONST_EQ_JF, LOOK_CONST_CALL2
superinstruction opcodes (VM handlers ready, compiler emission
for remaining patterns deferred to next pass).
2026-04-14 13:53:08 -04:00
..
.github/workflows Add superinstructions LOOK+1, LOOK-1 for 25% faster loops 2026-04-14 13:53:08 -04:00
lean Add superinstructions LOOK+1, LOOK-1 for 25% faster loops 2026-04-14 13:53:08 -04:00
.gitignore Add superinstructions LOOK+1, LOOK-1 for 25% faster loops 2026-04-14 13:53:08 -04:00
benchmark.sh Add EML universality proof — verify arXiv:2603.21852v2 2026-04-13 19:43:18 -04:00
benchmark_results.md Add proof friction benchmark, update README 2026-04-14 13:31:05 -04:00
eml_proof.lsp Add EML universality proof — verify arXiv:2603.21852v2 2026-04-13 19:43:18 -04:00
eml_proof.py Add EML universality proof — verify arXiv:2603.21852v2 2026-04-13 19:43:18 -04:00