[PATCH] D144265: DAG: Handle lowering unordered compare with inf

Matt Arsenault via Phabricator via llvm-commits llvm-commits at lists.llvm.org
Sun Jun 4 07:08:32 PDT 2023


arsenm added inline comments.


================
Comment at: llvm/lib/CodeGen/SelectionDAG/TargetLowering.cpp:8034
 SDValue TargetLowering::expandIS_FPCLASS(EVT ResultVT, SDValue Op,
-                                         FPClassTest TestMask,
+                                         const FPClassTest OrigTestMask,
                                          SDNodeFlags Flags, const SDLoc &DL,
----------------
RKSimon wrote:
> Maybe drop the const - I expect the OrigTestMask rename will have caught any changes in the function
And the const will keep it that way 


================
Comment at: llvm/test/CodeGen/X86/is_fpclass.ll:258
+; CHECK-32-NEXT:    sahf
+; CHECK-32-NEXT:    setb %al
 ; CHECK-32-NEXT:    retl
----------------
RKSimon wrote:
> Do you know if we have any legality/TLI callbacks that we could use to fallback to UseFP = false to avoid these regressions?
Is this a regression? I was thinking it might be beneficial to leave all the FP in the f registers rather than copying to integer. 

I don't know what I would check for here, the excess instructions seems to just be from the weird stackiness of x87. 


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D144265/new/

https://reviews.llvm.org/D144265



More information about the llvm-commits mailing list