[PATCH] D137655: Expand fminimum/fmaximum into fminnum/fmaxnum + NaN check

[Matt Arsenault via Phabricator via llvm-commits]

arsenm added inline comments.


================
Comment at: llvm/lib/Target/NVPTX/NVPTXISelLowering.cpp:2227-2228
+  SDValue NonPropagatingResult = DAG.getNode(NT, SL, VT, {LHS, RHS});
+  SDValue NaN = DAG.getConstantFP(std::numeric_limits<double>::quiet_NaN(), SL,
+                                  VT);
+  return DAG.getSelectCC(SL, LHS, RHS, NaN, NonPropagatingResult, ISD::SETUO);
----------------
arsenm wrote:
> Should go through APFloat 
Can also move this to generic code and check which of the variants are legal 


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D137655/new/

https://reviews.llvm.org/D137655