[PATCH] D57044: [AArch64] OOptimize floating point materialization

[Eli Friedman via Phabricator via llvm-commits]

efriedma added a comment.

Do we have test coverage for the various f16 cases somewhere?  I'd like to make sure we don't crash or miscompile for f16 inf.



================
Comment at: lib/Target/AArch64/AArch64ISelLowering.cpp:5442
+  // If we can not materialize in immediate field for fmov, check if the
+  // value can be encoded as the immediate operand of a logical instruction.
+  if (!IsLegal && (VT == MVT::f64 || VT == MVT::f32))
----------------
Maybe clarify precisely what sequence you expect will be emitted here?


================
Comment at: lib/Target/AArch64/AArch64ISelLowering.cpp:5448
+  LLVM_DEBUG(dbgs() << (IsLegal ? "Legal " : "Illegal ") << VT.getEVTString()
+                    << " imm value: "; Imm.dump(););
+  return IsLegal;
----------------
"dbgs() << Imm" should work for APInt, I think.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D57044/new/

https://reviews.llvm.org/D57044