[PATCH] D30270: [InstCombine] don't try SimplifyDemandedInstructionBits from add/sub because it's slow and unlikely to succeed

[Sanjay Patel via Phabricator via llvm-commits]

This revision was automatically updated to reflect the committed changes.
Closed by commit rL295898: [InstCombine] don't try SimplifyDemandedInstructionBits from add/sub because… (authored by spatel).

Changed prior to commit:
  https://reviews.llvm.org/D30270?vs=89402&id=89434#toc

Repository:
  rL LLVM

https://reviews.llvm.org/D30270

Files:
  llvm/trunk/lib/Transforms/InstCombine/InstCombineAddSub.cpp


Index: llvm/trunk/lib/Transforms/InstCombine/InstCombineAddSub.cpp
===================================================================
--- llvm/trunk/lib/Transforms/InstCombine/InstCombineAddSub.cpp
+++ llvm/trunk/lib/Transforms/InstCombine/InstCombineAddSub.cpp
@@ -1078,11 +1078,6 @@
   // FIXME: Use the match above instead of dyn_cast to allow these transforms
   // for splat vectors.
   if (ConstantInt *CI = dyn_cast<ConstantInt>(RHS)) {
-    // See if SimplifyDemandedBits can simplify this.  This handles stuff like
-    // (X & 254)+1 -> (X&254)|1
-    if (SimplifyDemandedInstructionBits(I))
-      return &I;
-
     // zext(bool) + C -> bool ? C + 1 : C
     if (ZExtInst *ZI = dyn_cast<ZExtInst>(LHS))
       if (ZI->getSrcTy()->isIntegerTy(1))
@@ -1589,9 +1584,6 @@
     if (match(Op1, m_Add(m_Value(X), m_Constant(C2))))
       return BinaryOperator::CreateSub(ConstantExpr::getSub(C, C2), X);
 
-    if (SimplifyDemandedInstructionBits(I))
-      return &I;
-
     // Fold (sub 0, (zext bool to B)) --> (sext bool to B)
     if (C->isNullValue() && match(Op1, m_ZExt(m_Value(X))))
       if (X->getType()->getScalarType()->isIntegerTy(1))


-------------- next part --------------
A non-text attachment was scrubbed...
Name: D30270.89434.patch
Type: text/x-patch
Size: 1155 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/llvm-commits/attachments/20170222/05e8f947/attachment.bin>