[PATCH] D105432: [Analysis] Add simple cost model for strict (in-order) reductions

[David Sherwood via Phabricator via llvm-commits]

david-arm added inline comments.


================
Comment at: llvm/include/llvm/Analysis/TargetTransformInfo.h:1152
+  static bool isOrderedReduction(unsigned Opcode, FastMathFlags FPFlags) {
+    if ((Opcode == Instruction::FAdd || Opcode == Instruction::FMul) &&
+        !FPFlags.allowReassoc())
----------------
sdesmalen wrote:
> Why does this need to check the opcode?
Sadly this is a result of passing using FastMathFlags to determine the algorithm. The flags are usually empty for integer operations, which means the allow reassoc flag will not be set. If we don't check the opcode then we end up using strict reductions for all integer operations.


CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D105432/new/

https://reviews.llvm.org/D105432