[compiler-rt] [compiler-rt][builtins]Fix complex division for aarch64 (PR #106664)

[David Green via llvm-commits]

davemgreen wrote:

Which "source code / other libraries" are producing the snan's?

My understanding is that snan is considered mostly useless, especially in a floating point environment where fp exceptions are not honoured. As in https://discourse.llvm.org/t/stronger-floating-point-nan-guarantees/72165/54. In my opinion it would be a shame to make every program slower for something should really not be used, (especially if there is no -fsignalling-nans).

Others might have other opinions though, perhaps with this routine being in the rt-library and giving the wrong result it is more important to fix.

https://github.com/llvm/llvm-project/pull/106664