Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
8 changes: 8 additions & 0 deletions README.md
Original file line number Diff line number Diff line change
Expand Up @@ -177,3 +177,11 @@ but the template jit pattern is great for a first attempt.
a two-pass register allocator would buy us another 1.5x or so on a few benchmarks. it would also triple the size of the codegen module and the point of jitvm is "small enough to fit in your head". the current allocator is intentionally bad: it picks the next free callee-saved register, falls back to spilling, and the spill code is dumb.

if you want to see what a serious jit looks like, read the V8 baseline compiler. this is the version where you can read every line in an afternoon.

## See also

- [github.com/f4rkh4d/jitvm/issues](https://github.com/f4rkh4d/jitvm/issues) for bug reports and feature requests

## issues

bug reports and feature requests are welcome at [github.com/f4rkh4d/jitvm/issues](https://github.com/f4rkh4d/jitvm/issues).
1 change: 1 addition & 0 deletions ROADMAP.md
Original file line number Diff line number Diff line change
Expand Up @@ -64,3 +64,4 @@ i'm explicitly not doing:
be tuples first, then arrays-as-tuples-with-runtime-length.
- a self-hosted compiler. jitvm is written in rust and is going to stay
written in rust.

Loading