[PATCH] D69887: [FEnv] File with properties of constrained intrinsics

[Ulrich Weigand via Phabricator via llvm-commits]

uweigand added a comment.

See inline comment.    Apart from that, this looks good to me -- I think we should go for that solution, it makes the code a lot more maintainable IMO.



================
Comment at: llvm/lib/CodeGen/SelectionDAG/LegalizeVectorTypes.cpp:42
+  {
+    // Process strictfp nodes that were not processes in this switch.
+    bool IsStrictFPNode = false;
----------------
This double switch is a bit weird.  I'm wondering if it wouldn't be more straightforward to just pass all strictfp nodes to ScalarizeVecRes_StrictFPOp and then either just handle them there (in most cases that should be straightforward as far as I can see) or call out to special handling from within that routine.

(Same for the three other cases below.)


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D69887/new/

https://reviews.llvm.org/D69887