[llvm] [X86] Adding lowerings for vector ISD::LRINT and ISD::LLRINT (PR #90065)

[Craig Topper via llvm-commits]

topperc wrote:

> > > > gcc requires `-O1 -fno-math-errno -m32 -mfpmath=sse -fno-trapping-math`
> > > 
> > > 
> > > `-fno-trapping-math` is not required. I think it's on by default.
> > 
> > 
> > -fno-trapping-math is not default on gcc. I was trying to find the subset of -ffast-math that gcc requires. it appears to be fno-math-errno and fno-trapping-math.
> 
> Yes, `-fno-trapping-math` is not default on gcc. But we don't need it here https://godbolt.org/z/TsKTP4TxK

Oops you're right. I don't know what I did earlier. I guess it was the -mfpmath=sse that fixed it.

https://github.com/llvm/llvm-project/pull/90065