Alex Shinn

LGTM, sorry for the delay.

The difference is the _proc variants define a procedure that calls out to the opcode, so you're getting an extra level of indirection. The source location is coming from there.

What changed after 0.7.3 was the flags for using static libs and for including clibs.c were separated to allow the flexibility of specifying your own libs. The [manual](https://synthcode.com/scheme/chibi/#h2_Installation) clearly states...

I had copy and pasted that command directly from the manual, however it was the latest copy of the manual: https://github.com/ashinn/chibi-scheme/blob/master/doc/chibi.scrbl#L82 which was updated after the last release (0.10).

So EOF is somehow becoming positive on 9front?

`i` is `sexp_sint_t` which may indeed be larger than `int`, so without the cast you'd expect it to promote EOF to the larger size. This _should_ work properly, so it...

It makes sense for iterators and higher-order functions to be call/cc safe, because they are generic and should compose with continuations. This is not the case for `array-copy`, where the...

> What "call/cc [safety]" actually means is that (using continuations) one cannot detect mutation of previously allocated locations. Thanks for the definition. It's important to be precise in specifying new...

I'm curious as to what this efficient implementation is. The implementation in SRFI 133 is not call/cc safe: https://github.com/scheme-requests-for-implementation/srfi-133/blob/master/vectors/vectors-impl.scm#L823

This probably performs well in benchmarks for small vectors, but you risk blowing the stack. Building an intermediate list is safer, but I realized we can just copy the vector...