Sam Tobin-Hochstadt

Results 486 comments of Sam Tobin-Hochstadt

Upon further examination, I think rlib.rrandom is basically the same thing as the Racket code.

Note that mixed-exactness complex number don't exist, which also argues in favor of this. Sam

Probably this is more what you want: ``` #lang racket (provide (struct-out foo)) (struct foo (a)) (define-syntax y (make-rename-transformer (syntax-property #'foo 'not-free-identifier=? #t))) ```

If you've installed Racket there should be a parallel scribble binary. There is also a raco scribble command that should work as well if you can run raco.

The problem is that `brew install racket` prints a deprecation warning and tells people to install `minimal-racket` instead (that's what's described in the issue).

Let's start by figuring out what Racket does here, since I know Matthew has optimized this a lot.

I think that none of the papers are the illuminating here. The high-order bit though is that Racket has a separate continuation-mark stack. We could probably emulate that by tracking...

Racket caches the results of lookup halfway on every lookup.

from @cfbolz: GC: there is a way to get statistics from the GC, via rpython.rlib.rgc: rgc.get_stats(rgc.TOTAL_MEMORY) this returns an int. The downside is that it doesn't work untranslated, so you...

For my edification, what patterns in Schemify output get helped by this?