Go to file
Miguel M b899ddd63a Final commit, in principle.
Got things running at about ~1ns per iteration, seems good enough.
Adopted a linear strategy to evaluating the bytecode, rather than a
recursive or even imperative evaluation strategy; this also lets me
elide the offsets and store the bytecode in half the size. Looking
forward to finding out that formulas are evaluated wrongly, but couldn't
find a counterexample. Also restructured things a bit to avoid multiple
alocations when evaluating by this strategy.
2023-02-24 19:41:51 +00:00
.github/workflows vm working 2023-02-22 16:58:37 +00:00
benches Final commit, in principle. 2023-02-24 19:41:51 +00:00
pytest Final commit, in principle. 2023-02-24 19:41:51 +00:00
src Final commit, in principle. 2023-02-24 19:41:51 +00:00
.gitignore vm working 2023-02-22 16:58:37 +00:00
Cargo.lock Final commit, in principle. 2023-02-24 19:41:51 +00:00
Cargo.toml Final commit, in principle. 2023-02-24 19:41:51 +00:00
Makefile added tests, produced a flamegraph. VM seems to be the bottleneck 2023-02-24 14:50:26 +00:00
linear_vm_flamegraph.svg Final commit, in principle. 2023-02-24 19:41:51 +00:00
pyproject.toml added tests, produced a flamegraph. VM seems to be the bottleneck 2023-02-24 14:50:26 +00:00
recursive_vm_flamegraph.svg Final commit, in principle. 2023-02-24 19:41:51 +00:00
time.ps1 Final commit, in principle. 2023-02-24 19:41:51 +00:00