kaseyc commented on Issue #2601:
Hi Chris, thank you for the review!
Added the test to the x64 backend and tried to follow the naming scheme in that directory,
I noticed that the existing i128 compile test also covers these features, is it desirable to have overlapped compile and run tests?The failing checks are due to a security advisory for the raw-cpuid crate.
Lastly, for my edification: is the x64 backed the lowering framework, as opposed to the older legalizing one?
cfallin commented on Issue #2601:
Thanks!
is it desirable to have overlapped compile and run tests?
Yes, it's fine; nominally at least, they're testing different things (requiring a specific lowering, vs. ensuring that the lowering actually executes to produce desired results). IMHO we should flesh this out a bit more in the future and add run tests for everything that's required by CLIF semantics, but that's for a later time!
failing checks are due to a security advisory
I'll go ahead and do a quick PR to bump the version, then we should be able to merge main into this branch to make cargo-deny happy.
Lastly, for my edification: is the x64 backed the lowering framework, as opposed to the older legalizing one?
Yes, exactly. We sometimes call it a
Machinst
backend or "new-style backend" and it's based on lowering to VCode.
cfallin commented on Issue #2601:
@kaseyc I've fixed the cargo-deny issue in #2603; you should be able to merge in
main
now to get a green CI.
kaseyc commented on Issue #2601:
Rebased into head.
cfallin commented on Issue #2601:
Thanks!
Last updated: Nov 22 2024 at 16:03 UTC