diff --git a/lib/Analysis/BasicAliasAnalysis.cpp b/lib/Analysis/BasicAliasAnalysis.cpp
index b1578f4b88e..2ede4baaf68 100644
--- a/lib/Analysis/BasicAliasAnalysis.cpp
+++ b/lib/Analysis/BasicAliasAnalysis.cpp
@@ -66,7 +66,7 @@ using namespace llvm;
 
 /// Enable analysis of recursive PHI nodes.
 static cl::opt<bool> EnableRecPhiAnalysis("basic-aa-recphi", cl::Hidden,
-                                          cl::init(false));
+                                          cl::init(true));
 
 /// By default, even on 32-bit architectures we use 64-bit integers for
 /// calculations. This will allow us to more-aggressively decompose indexing
diff --git a/test/Analysis/BasicAA/phi-loop.ll b/test/Analysis/BasicAA/phi-loop.ll
index db3023c6560..e54752a9223 100644
--- a/test/Analysis/BasicAA/phi-loop.ll
+++ b/test/Analysis/BasicAA/phi-loop.ll
@@ -1,4 +1,4 @@
-; RUN: opt < %s -basic-aa -basic-aa-recphi=1 -gvn -S | FileCheck %s
+; RUN: opt < %s -basic-aa -gvn -S | FileCheck %s
 ;
 ; Check that section->word_ofs doesn't get reloaded in every iteration of the
 ; for loop.
diff --git a/test/Analysis/BasicAA/recphi.ll b/test/Analysis/BasicAA/recphi.ll
index dfc88937bf6..a6157f22a4f 100644
--- a/test/Analysis/BasicAA/recphi.ll
+++ b/test/Analysis/BasicAA/recphi.ll
@@ -1,4 +1,4 @@
-; RUN: opt < %s -basic-aa -aa-eval -print-all-alias-modref-info -basic-aa-recphi -disable-output 2>&1 | FileCheck %s
+; RUN: opt < %s -basic-aa -aa-eval -print-all-alias-modref-info -disable-output 2>&1 | FileCheck %s
 
 ; CHECK-LABEL: Function: simple: 5 pointers, 0 call sites
 ; CHECK:         NoAlias:      float* %src1, float* %src2
diff --git a/test/Transforms/LoopIdiom/reuse-cast.ll b/test/Transforms/LoopIdiom/reuse-cast.ll
index 0b1bfeaa273..0dda3515156 100644
--- a/test/Transforms/LoopIdiom/reuse-cast.ll
+++ b/test/Transforms/LoopIdiom/reuse-cast.ll
@@ -83,21 +83,22 @@ declare void @use.i1(i1)
 define void @reuse_cast_2(i32 %x, i32* %ptr.1.start) {
 ; CHECK-LABEL: @reuse_cast_2(
 ; CHECK-NEXT:  entry:
+; CHECK-NEXT:    [[PTR_1_START1:%.*]] = bitcast i32* [[PTR_1_START:%.*]] to i8*
 ; CHECK-NEXT:    [[STACK:%.*]] = alloca [2 x i32], align 4
 ; CHECK-NEXT:    [[CAST_TO_REUSE:%.*]] = bitcast [2 x i32]* [[STACK]] to i8*
 ; CHECK-NEXT:    [[C_0:%.*]] = icmp sgt i32 [[X:%.*]], 0
 ; CHECK-NEXT:    [[TMP0:%.*]] = bitcast [2 x i32]* [[STACK]] to i8*
 ; CHECK-NEXT:    [[PTR_2_START:%.*]] = getelementptr inbounds [2 x i32], [2 x i32]* [[STACK]], i64 0, i64 0
 ; CHECK-NEXT:    call void @use.i8(i8* [[CAST_TO_REUSE]])
+; CHECK-NEXT:    call void @llvm.memcpy.p0i8.p0i8.i64(i8* align 4 [[CAST_TO_REUSE]], i8* align 4 [[PTR_1_START1]], i64 8, i1 false)
 ; CHECK-NEXT:    br label [[LOOP:%.*]]
 ; CHECK:       loop:
 ; CHECK-NEXT:    [[IV:%.*]] = phi i32 [ 0, [[ENTRY:%.*]] ], [ [[IV_NEXT:%.*]], [[LOOP]] ]
-; CHECK-NEXT:    [[PTR_1:%.*]] = phi i32* [ [[PTR_1_START:%.*]], [[ENTRY]] ], [ [[PTR_1_NEXT:%.*]], [[LOOP]] ]
+; CHECK-NEXT:    [[PTR_1:%.*]] = phi i32* [ [[PTR_1_START]], [[ENTRY]] ], [ [[PTR_1_NEXT:%.*]], [[LOOP]] ]
 ; CHECK-NEXT:    [[PTR_2:%.*]] = phi i32* [ [[PTR_2_START]], [[ENTRY]] ], [ [[PTR_2_NEXT:%.*]], [[LOOP]] ]
 ; CHECK-NEXT:    [[PTR_1_NEXT]] = getelementptr inbounds i32, i32* [[PTR_1]], i64 1
 ; CHECK-NEXT:    [[LV:%.*]] = load i32, i32* [[PTR_1]], align 4
 ; CHECK-NEXT:    [[PTR_2_NEXT]] = getelementptr inbounds i32, i32* [[PTR_2]], i64 1
-; CHECK-NEXT:    store i32 [[LV]], i32* [[PTR_2]], align 4
 ; CHECK-NEXT:    [[IV_NEXT]] = add nuw nsw i32 [[IV]], 1
 ; CHECK-NEXT:    [[C_1:%.*]] = icmp eq i32 [[IV]], 0
 ; CHECK-NEXT:    br i1 [[C_1]], label [[LOOP]], label [[EXIT:%.*]]