[llvm-dev] Implementing a proposed InstCombine optimization

[via llvm-dev]

> On Apr 7, 2016, at 10:06 AM, Carlos Liam <carlos at aarzee.me> wrote:
> 
> @escha:
> 
> Alive-NJ (https://github.com/rutgers-apl/alive-nj) verifies the optimization as correct for half, single and double floating-point types:

The problem is that it proves it’s correct on a certain set of semantics, but not all targets will share those semantics, which is something you have to be very careful about with floating point. In some cases “correct under IEEE” means “correct under non-IEEE”, but in other cases it doesn’t.

—escha