| Commit message (Collapse) | Author | Age | Files | Lines |
| |
|
|
|
|
|
|
|
|
|
|
|
| |
We haven't implemented it yet, but there's now a design for how we check
whether a function has written its output correctly. Functions must write
to each output at the top level at least once, and never overwrite an output
register in the top-level once it's been defined.
This is conservative (it can be perfectly reasonable for functions to write
the output, reuse the register for a temporary, and then write the output
again) but easy to check.
|
| |
|
|
|
|
|
| |
Bugfix: statements defining a new register variable require an initializer
instruction.
|
| |
|
| |
|
|
|
|
|
| |
Copy (mov) instructions are unlike instructions we've encoded so far, in
that their destination is not read.
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
| |
See test-convert-function-with-arg-and-body.
|
|
|
|
|
|
|
|
|
|
|
| |
One test failing. It took enough debugging just to get to the expected
failure that it seems worth saving this snapshot.
Saw some signs that I have to remember to zero-out allocated memory. We
need a scalable solution for this.
I think parse-var-with-type needs to be rewritten. Just added a test and
a hacky fix for now.
|
|
|
|
| |
Minor tweaks to stop wasting horizontal space in the trace.
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
The problem: when I hit 'G' to go to the bottom of the trace, if the bottom
is visible on screen, the screen scrolls so the bottom of the trace
is the bottom-most line on screen. But the cursor moves to where the trace
used to end rather than the new location of the bottom of the trace (the
bottom of the screen).
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Try to make the comments consistent with the type system we'll eventually
have.
|
| |
|
|
|
|
| |
Fix CI.
|
| |
|
| |
|
| |
|
|
|
|
|
| |
Fuck, 'tmp' is a generic name, and running `clean` deleted some files I'd
stashed away.
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
|
| |
Rather surprisingly, all the treeshake tooling is done in just about 2
hours of work. From now on it'll be easier to update stats.txt. Observations:
a) Binaries are tiny compared to conventional stacks. Tens of KB.
b) ~80% of binaries are tests and unused libraries in all my apps.
c) ~75% of LoC in SubX sources are tests or comments.
|
|
|
|
|
|
|
|
|
| |
Start of a new script called treeshake to emit stats for minimal line counts
and binary sizes for all apps.
It doesn't actually do any dead-code deletion yet. But it does build and
run all apps successfully. (Except apps/mu; we'll ignore that for now.
It's probably not being disciplined about identifying internal labels.)
|
|
|
|
|
| |
Fix a bug in one test: it checks eax when the component under test returns
nothing. It's been just accidentally passing all these months.
|
| |
|
|
|
|
|
|
| |
Standardize conventions for labels within objects in the data segment.
We're going to use this in a new tool.
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
| |
All tests passing again. We have big gaping holes for type- and var-management.
We're going to work on the latter first.
|
|
|
|
|
| |
Lots of debugging to add two curly braces. I need tests for populate-mu-function-body,
or even parse-mu-block.
|