fix an invisible bug when combining repeated FP divisors

This patch fixes bugs that were exposed by the addition of fast-math-flags in the DAG:
r237046 ( http://reviews.llvm.org/rL237046 ):

1. When replacing a division node, it's not enough to RAUW.
   We should call CombineTo() to delete dead nodes and combine again.
2. Because we are changing the DAG, we can't return an empty SDValue
   after the transform. As the code comments say:

    Visitation implementation - Implement dag node combining for different node types.
    The semantics are as follows: Return Value:
      SDValue.getNode() == 0 - No change was made
      SDValue.getNode() == N - N was replaced, is dead and has been handled.
      otherwise - N should be replaced by the returned Operand.

The new test case shows no difference with or without this patch, but it will crash if
we re-apply r237046 or enable FMF via the current -enable-fmf-dag cl::opt.

Differential Revision: http://reviews.llvm.org/D9893

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@241826 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/CodeGen/SelectionDAG/DAGCombiner.cpp b/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
index 86b73d35806e287984ac2b5e5b7842025067993f..e05714342d211301486150475234ec9cd29d4eb8 100644 (file)
--- a/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
+++ b/lib/CodeGen/SelectionDAG/DAGCombiner.cpp
@@ -8374,6 +8374,9 @@ SDValue DAGCombiner::visitFDIV(SDNode *N) {
 
     if (TLI.combineRepeatedFPDivisors(Users.size())) {
       SDValue FPOne = DAG.getConstantFP(1.0, DL, VT);
+      // FIXME: This optimization requires some level of fast-math, so the
+      // created reciprocal node should at least have the 'allowReciprocal'
+      // fast-math-flag set.
       SDValue Reciprocal = DAG.getNode(ISD::FDIV, DL, VT, FPOne, N1);
 
       // Dividend / Divisor -> Dividend * Reciprocal
@@ -8382,10 +8385,14 @@ SDValue DAGCombiner::visitFDIV(SDNode *N) {
         if (Dividend != FPOne) {
           SDValue NewNode = DAG.getNode(ISD::FMUL, SDLoc(U), VT, Dividend,
                                         Reciprocal);
-          DAG.ReplaceAllUsesWith(U, NewNode.getNode());
+          CombineTo(U, NewNode);
+        } else if (U != Reciprocal.getNode()) {
+          // In the absence of fast-math-flags, this user node is always the
+          // same node as Reciprocal, but with FMF they may be different nodes.
+          CombineTo(U, Reciprocal);
         }
       }
-      return SDValue();
+      return SDValue(N, 0);  // N was replaced.
     }
   }
 

diff --git a/test/CodeGen/X86/fdiv-combine.ll b/test/CodeGen/X86/fdiv-combine.ll
index 8adfde7f3f38a07813053b437f845e000d798d7c..34eac62e367335af0916e638300c9ba22cf55782 100644 (file)
--- a/test/CodeGen/X86/fdiv-combine.ll
+++ b/test/CodeGen/X86/fdiv-combine.ll
@@ -27,5 +27,22 @@ define float @div2_arcp(float %x, float %y, float %z) #0 {
   ret float %div2
 }
 
+; If the reciprocal is already calculated, we should not
+; generate an extra multiplication by 1.0. 
+
+define double @div3_arcp(double %x, double %y, double %z) #0 {
+; CHECK-LABEL: div3_arcp:
+; CHECK:       # BB#0:
+; CHECK-NEXT:    movsd{{.*#+}} xmm2 = mem[0],zero
+; CHECK-NEXT:    divsd %xmm1, %xmm2
+; CHECK-NEXT:    mulsd %xmm2, %xmm0
+; CHECK-NEXT:    addsd %xmm2, %xmm0
+; CHECK-NEXT:    retq
+  %div1 = fdiv fast double 1.0, %y
+  %div2 = fdiv fast double %x, %y
+  %ret = fadd fast double %div2, %div1
+  ret double %ret
+}
+
 ; FIXME: If the backend understands 'arcp', then this attribute is unnecessary.
 attributes #0 = { "unsafe-fp-math"="true" }