2 ms·
Yes, I was referring to request to provide copy of -noboundscheck. You stated that nobody seemed interested in turning off safety features, so I provided a coun
by sanxiyn 11y ago
Yes, I was referring to request to provide copy of -noboundscheck. You stated that nobody seemed interested in turning off safety features, so I provided a counterexample.
As a matter of fact, I am interested in -noboundscheck switch. The motivation is so that people can easily confirm bound checks overhead is zero by themselves, because it seems many people find this hard to believe. I didn't submit a pull request implementing this (it is easy to implement) because I thought it was fait accompli. Was I mistaken and you will support such pull request?
- pcwalton 11y agoIt'd need to go through an RFC process, but I would be neutral and not opposed to such an RFC.
- kibwen 11y agoI actually intend to benchmark this myself in the future and write a comprehensive blog post on my findings. Personally, I would oppose any RFC that attempted to add such a compiler flag, especially if its only purpose was to prove a point.