bjorn3 opened PR #6174 from bjorn3:bsc-unwinding-more-libcall-sigs
to bytecodealliance:main
:
This is necessary to be able to call them in the interpreter. All the remaining libcalls which signature() doesn't handle are never used in clif ir. Only in code compiled by a backend.
bjorn3 requested jameysharp for a review on PR #6174.
bjorn3 requested wasmtime-compiler-reviewers for a review on PR #6174.
bjorn3 updated PR #6174.
bjorn3 updated PR #6174.
bjorn3 has marked PR #6174 as ready for review.
jameysharp submitted PR review.
jameysharp submitted PR review.
jameysharp created PR review comment:
I think we have the target architecture available on
self
here and could compute the pointer size from it. I'm not too worried about it right now so you could leave a FIXME comment like you did in the interpreter if you prefer.
jameysharp created PR review comment:
I had a really hard time figuring out why the length arguments also use
pointer_type
since they aren't pointers. You are correct but maybe we can make this a little more clear.One way might be to rename that parameter; maybe
int_pointer_type
?Another way is to just add a comment. Actually I'd kind of appreciate a comment on each libcall signature anyway. Maybe describing its API in either C or Rust syntax, as a reminder, so it's easy to compare that to the ABI we're transforming it to.
bjorn3 updated PR #6174.
bjorn3 submitted PR review.
bjorn3 created PR review comment:
Added the C function signatures in comments. I don't really like
int_pointer_type
as name as I think it can be interpreted as either a type that is used forint*
pointers or for theint
type and pointers, but it is used forsize_t
and pointers.
bjorn3 updated PR #6174.
jameysharp submitted PR review.
jameysharp merged PR #6174.
Last updated: Nov 22 2024 at 16:03 UTC