[PATCH] D87456: AMDGPU/GlobalISel Check for NoNaNsFPMath in isKnownNeverSNaN

[Petar Avramovic via Phabricator via llvm-commits]

Petar.Avramovic added a comment.

In D87456#2267858 <https://reviews.llvm.org/D87456#2267858>, @foad wrote:

>> This will be fixed by in D87511 <https://reviews.llvm.org/D87511>.
>
> Wouldn't it make more sense to change the order of the patches in the stack so that D87511 <https://reviews.llvm.org/D87511> is applied first and this one is rebaed on top of it?

Yes, but it is this way because of the regression test. Change is visible in output when there is something that accesses function attributes.



================
Comment at: llvm/lib/CodeGen/GlobalISel/IRTranslator.cpp:2658
   const TargetMachine &TM = MF->getTarget();
+  TM.resetTargetOptions(F);
   SL = std::make_unique<GISelSwitchLowering>(this, FuncInfo);
----------------
arsenm wrote:
> Should copy the comment from the DAG
>   // Reset the target options before resetting the optimization
>   // level below.
>   // FIXME: This is a horrible hack and should be processed via
>   // codegen looking at the optimization level explicitly when
>   // it wants to look at it.
This does not seem relevant since global-isel does not reset opt level.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D87456/new/

https://reviews.llvm.org/D87456