[DAGCombine] Fix demanded bits computation for exact shifts.

Fixes a miscompilation of MultiSource/Benchmarks/MallocBench/gs

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@240796 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/CodeGen/SelectionDAG/TargetLowering.cpp b/lib/CodeGen/SelectionDAG/TargetLowering.cpp
index b40025b41f7c2af898a7f87a88936e6723222833..fc746588e693fdf06b7e657827db268b4c556c10 100644 (file)
--- a/lib/CodeGen/SelectionDAG/TargetLowering.cpp
+++ b/lib/CodeGen/SelectionDAG/TargetLowering.cpp
@@ -700,6 +700,13 @@ bool TargetLowering::SimplifyDemandedBits(SDValue Op,
       if (ShAmt >= BitWidth)
         break;
 
+      APInt InDemandedMask = (NewMask << ShAmt);
+
+      // If the shift is exact, then it does demand the low bits (and knows that
+      // they are zero).
+      if (cast<BinaryWithFlagsSDNode>(Op)->Flags.hasExact())
+        InDemandedMask |= APInt::getLowBitsSet(BitWidth, ShAmt);
+
       // If this is ((X << C1) >>u ShAmt), see if we can simplify this into a
       // single shift.  We can do this if the top bits (which are shifted out)
       // are never demanded.
@@ -722,7 +729,7 @@ bool TargetLowering::SimplifyDemandedBits(SDValue Op,
       }
 
       // Compute the new bits that are at the top now.
-      if (SimplifyDemandedBits(InOp, (NewMask << ShAmt),
+      if (SimplifyDemandedBits(InOp, InDemandedMask,
                                KnownZero, KnownOne, TLO, Depth+1))
         return true;
       assert((KnownZero & KnownOne) == 0 && "Bits known to be one AND zero?");
@@ -753,6 +760,11 @@ bool TargetLowering::SimplifyDemandedBits(SDValue Op,
 
       APInt InDemandedMask = (NewMask << ShAmt);
 
+      // If the shift is exact, then it does demand the low bits (and knows that
+      // they are zero).
+      if (cast<BinaryWithFlagsSDNode>(Op)->Flags.hasExact())
+        InDemandedMask |= APInt::getLowBitsSet(BitWidth, ShAmt);
+
       // If any of the demanded bits are produced by the sign extension, we also
       // demand the input sign bit.
       APInt HighBits = APInt::getHighBitsSet(BitWidth, ShAmt);

diff --git a/test/CodeGen/X86/xor.ll b/test/CodeGen/X86/xor.ll
index 829be41e5127956e9a9a5605ba4a88694c0ac9a4..0f6a601d954de66779c01bfed19cdc9b71b9354a 100644 (file)
--- a/test/CodeGen/X86/xor.ll
+++ b/test/CodeGen/X86/xor.ll
@@ -193,3 +193,22 @@ define i32 @test11(i32 %b) {
 ; X32: movl    $-2, %[[REG:.*]]
 ; X32: roll    %{{.*}}, %[[REG]]
 }
+
+%struct.ref_s = type { %union.v, i16, i16 }
+%union.v = type { i64 }
+
+define %struct.ref_s* @test12(%struct.ref_s* %op, i64 %osbot, i64 %intval) {
+  %neg = shl i64 %intval, 32
+  %sext = xor i64 %neg, -4294967296
+  %idx.ext = ashr exact i64 %sext, 32
+  %add.ptr = getelementptr inbounds %struct.ref_s, %struct.ref_s* %op, i64 %idx.ext
+  ret %struct.ref_s* %add.ptr
+; X64-LABEL: test12:
+; X64: shlq    $32, %[[REG:.*]]
+; X64-NOT: not
+; X64: sarq    $28, %[[REG]]
+; X32-LABEL: test12:
+; X32: leal
+; X32-NOT: not
+; X32: shll    $2, %eax
+}