[InstCombine] refactor ctlz/cttz folds (NFCI)

[spatel]

Summary

I suggested this change in D23134, but it was not expressed clearly enough. Maybe reading the code is easier? :)

The existing transforms for ctlz/cttz are almost identical, and the code is obviously copy/pasted.

We just need to shift the mask over for ctlz to see if our operand gives us a known constant value.

I've also added the FIXME comment that I suggested because I think this (and several other existing transforms in this file) could be moved to InstSimplify.

[spatel]

Use getHighBitsSet() to mimic the existing code; more symmetrical this way.

[majnemer]

lib/Transforms/InstCombine/InstCombineCalls.cpp

1107

Can't IsTZ be divined from the II?

1110

auto *

[spatel]

lib/Transforms/InstCombine/InstCombineCalls.cpp

1107

Yep - that looks nicer.

1110

Done.

[spatel]

Patch updated:

1. Check intrinsic ID inside the function to simplify caller and API.
2. Use auto with dyn_cast.

[spatel]

The previous draft was missing the 'auto' change.

[majnemer]

LGTM

lib/Transforms/InstCombine/InstCombineCalls.cpp

static Instruction *simplifyMaskedScatter(IntrinsicInst &II, InstCombiner &IC) {
   // If the mask is all zeros, a scatter does nothing.
   auto *ConstMask = dyn_cast<Constant>(II.getArgOperand(3));
   if (ConstMask && ConstMask->isNullValue())
     return IC.eraseInstFromFunction(II);
 
   return nullptr;
 }
 
+static Value *foldCttzCtlz(IntrinsicInst &II, InstCombiner &IC, bool IsTZ) {
+  Value *Op0 = II.getArgOperand(0);
+  // FIXME: Try to simplify vectors of integers.
+  IntegerType *IT = dyn_cast<IntegerType>(Op0->getType());
+  if (!IT)
+    return nullptr;
+
+  unsigned BitWidth = IT->getBitWidth();
+  APInt KnownZero(BitWidth, 0);
+  APInt KnownOne(BitWidth, 0);
+  IC.computeKnownBits(Op0, KnownZero, KnownOne, 0, &II);
+
+  // Create a mask for bits above (ctlz) or below (cttz) the first known one.
+  unsigned NumMaskBits =
+      IsTZ ? KnownOne.countTrailingZeros() : KnownOne.countLeadingZeros();
+  APInt Mask = IsTZ ? APInt::getLowBitsSet(BitWidth, NumMaskBits)
+                    : APInt::getHighBitsSet(BitWidth, NumMaskBits);
+
+  // If all bits above (ctlz) or below (cttz) the first known one are known
+  // zero, this value is constant.
+  // FIXME: This should be in InstSimplify because we're replacing an
+  // instruction with a constant.
+  if ((Mask & KnownZero) == Mask)
+    return ConstantInt::get(IT, APInt(BitWidth, NumMaskBits));
+
+  return nullptr;
+}
+
 // TODO: If the x86 backend knew how to convert a bool vector mask back to an
 // XMM register mask efficiently, we could transform all x86 masked intrinsics
 // to LLVM masked intrinsics and remove the x86 masked intrinsic defs.
 static Instruction *simplifyX86MaskedLoad(IntrinsicInst &II, InstCombiner &IC) {
   Value *Ptr = II.getOperand(0);
   Value *Mask = II.getOperand(1);
   Constant *ZeroVec = Constant::getNullValue(II.getType());
 
if (ConstantInt *Power = dyn_cast<ConstantInt>(II->getArgOperand(1))) {
       if (Power->isOne())
         return replaceInstUsesWith(CI, II->getArgOperand(0));
       // powi(x, -1) -> 1/x
       if (Power->isAllOnesValue())
         return BinaryOperator::CreateFDiv(ConstantFP::get(CI.getType(), 1.0),
                                           II->getArgOperand(0));
     }
     break;
-  case Intrinsic::cttz: {
-    // If all bits below the first known one are known zero,
-    // this value is constant.
-    IntegerType *IT = dyn_cast<IntegerType>(II->getArgOperand(0)->getType());
-    // FIXME: Try to simplify vectors of integers.
-    if (!IT) break;
-    uint32_t BitWidth = IT->getBitWidth();
-    APInt KnownZero(BitWidth, 0);
-    APInt KnownOne(BitWidth, 0);
-    computeKnownBits(II->getArgOperand(0), KnownZero, KnownOne, 0, II);
-    unsigned TrailingZeros = KnownOne.countTrailingZeros();
-    APInt Mask(APInt::getLowBitsSet(BitWidth, TrailingZeros));
-    if ((Mask & KnownZero) == Mask)
-      return replaceInstUsesWith(CI, ConstantInt::get(IT,
-                                 APInt(BitWidth, TrailingZeros)));
 
-    }
+  case Intrinsic::cttz:
+    if (Value *V = foldCttzCtlz(*II, *this, true))
+      return replaceInstUsesWith(*II, V);
     break;
-  case Intrinsic::ctlz: {
-    // If all bits above the first known one are known zero,
-    // this value is constant.
-    IntegerType *IT = dyn_cast<IntegerType>(II->getArgOperand(0)->getType());
-    // FIXME: Try to simplify vectors of integers.
-    if (!IT) break;
-    uint32_t BitWidth = IT->getBitWidth();
-    APInt KnownZero(BitWidth, 0);
-    APInt KnownOne(BitWidth, 0);
-    computeKnownBits(II->getArgOperand(0), KnownZero, KnownOne, 0, II);
-    unsigned LeadingZeros = KnownOne.countLeadingZeros();
-    APInt Mask(APInt::getHighBitsSet(BitWidth, LeadingZeros));
-    if ((Mask & KnownZero) == Mask)
-      return replaceInstUsesWith(CI, ConstantInt::get(IT,
-                                 APInt(BitWidth, LeadingZeros)));
-
-    }
+  case Intrinsic::ctlz:
+    if (Value *V = foldCttzCtlz(*II, *this, false))
+      return replaceInstUsesWith(*II, V);
     break;
 
   case Intrinsic::uadd_with_overflow:
   case Intrinsic::sadd_with_overflow:
   case Intrinsic::umul_with_overflow:
   case Intrinsic::smul_with_overflow:
     if (isa<Constant>(II->getArgOperand(0)) &&
         !isa<Constant>(II->getArgOperand(1))) {