[PATCH] D44909: [DAGCombine] (float)((int) f) --> ftrunc (PR36617)

[Nemanja Ivanovic via Phabricator via llvm-commits]

nemanjai added a comment.

In https://reviews.llvm.org/D44909#1048805, @scanon wrote:

> Two questions, to which I do not know the answer:
>
> (a) Are the semantics of ISD::FP_TO_[US]INT for out-of-range values specified anywhere? The do not seem to be, but maybe I'm just missing it.
>  (b) Are there test cases that make sure this transform is *not* applied for armv7 and x86 -sse4.1?


I think this is important to answer. I can see that on PPC, we use saturating FP -> Int conversions (i.e. values larger than `2^n -1` produce the maximum integral value, etc.). That behaviour will certainly change with this combine.


https://reviews.llvm.org/D44909