[PATCH] D92083: Legalize the SETCC/STRICT_FSETCCS as libcall

[Qiu Chaofan via Phabricator via llvm-commits]

qiucf added a comment.

Thanks for the work! I thought code to 'unfold' `SELECT_CC` and `BR_CC` should exist in generic code..



================
Comment at: llvm/lib/CodeGen/SelectionDAG/LegalizeDAG.cpp:4475
+    if (RHS.getNode())
+      LHS = DAG.getNode(ISD::SETCC, dl, Node->getValueType(0), LHS, RHS,
+                        DAG.getCondCode(CCCode));
----------------
Will this okay if `IsStrict == true`?


================
Comment at: llvm/lib/Target/PowerPC/PPCISelLowering.cpp:8161
+
+  // PowerPC didn't have native instruction to do the comparasion for fp128 if
+  // Power9 vector is not enabled. So, we need to do the following
----------------
`comparasion` typo?


================
Comment at: llvm/lib/Target/PowerPC/PPCISelLowering.cpp:8176
+  // Not FP, or using SPE? Not a fsel.
+  if (!CmpVT.isFloatingPoint() || !TV.getValueType().isFloatingPoint() ||
+      Subtarget.hasSPE())
----------------
This early return isn't about `P8` or `P9`?


Repository:
  rG LLVM Github Monorepo

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D92083/new/

https://reviews.llvm.org/D92083