Consider using an embedded libm

[emberian]

For performance and consistency reasons, we might want to bundle our own libm. It further reduces our dependency on system libraries, opens the potential for LTO, and gives us something we can control and define.

A good candidate is openlibm, which has a long history.

See #5702, #1222

[emberian]

Unless there are objections, I plan on using openlibm.

[brendanzab]

I don't know enough about it to make any comment, but I do think it would be awesome if the libm functionality was rewritten in Rust. But this might be reasonable in the short term. cc. @jensnockert

[thestinger]

I think we should support alternate math libraries, but I don't think we should be embedding any more libraries.

[emberian]

There was also a discussion on IRC; I'm convinced that this is a bad idea.

[thestinger]

I'm re-opening this, because I think it's going to be necessary if we want to keep using the LLVM math intrinsics. These intrinsics assume errno is not set on errors so LLVM may reorder them in a way that would clobber another errno value before it is read.

[steveklabnik]

I'm pulling a massive triage effort to get us ready for 1.0. As part of this, I'm moving stuff that's wishlist-like to the RFCs repo, as that's where major new things should get discussed/prioritized.

This issue has been moved to the RFCs repo: rust-lang/rfcs#711