[libc-commits] [libc] [llvm] [APFloat] Add exp function for APFloat::IEEESsingle using expf implementation from LLVM libc. (PR #143959)
Joshua Cranmer via libc-commits
libc-commits at lists.llvm.org
Thu Jun 26 12:03:17 PDT 2025
https://github.com/jcranmer-intel commented:
Doing a local build, and manually tweaking the test to set denormal flushing on, did give me a test failure, so that at least does satisfy my concerns about tests failing if the compiler has fast-math inadvertently turned on.
Asides from the comment below, I don't really have any blocking concerns here. But there's still some lingering issues I have about how this gets extended to unsupported floating-point types or how to actually use these functions in optimizations or constant evaluation if it's not a required dependency.
https://github.com/llvm/llvm-project/pull/143959
More information about the libc-commits
mailing list