diff --git a/lib/Target/X86/AsmPrinter/X86MCInstLower.cpp b/lib/Target/X86/AsmPrinter/X86MCInstLower.cpp
index 88370e630bc..c851ca3fc80 100644
--- a/lib/Target/X86/AsmPrinter/X86MCInstLower.cpp
+++ b/lib/Target/X86/AsmPrinter/X86MCInstLower.cpp
@@ -287,7 +287,9 @@ void X86MCInstLower::Lower(const MachineInstr *MI, MCInst &OutMI) const {
     LowerUnaryToTwoAddr(OutMI, X86::MMX_PCMPEQDrr); break;
   case X86::FsFLD0SS:     LowerUnaryToTwoAddr(OutMI, X86::PXORrr); break;
   case X86::FsFLD0SD:     LowerUnaryToTwoAddr(OutMI, X86::PXORrr); break;
-  case X86::V_SET0:       LowerUnaryToTwoAddr(OutMI, X86::XORPSrr); break;
+  case X86::V_SET0PS:     LowerUnaryToTwoAddr(OutMI, X86::XORPSrr); break;
+  case X86::V_SET0PD:     LowerUnaryToTwoAddr(OutMI, X86::XORPDrr); break;
+  case X86::V_SET0PI:     LowerUnaryToTwoAddr(OutMI, X86::PXORrr); break;
   case X86::V_SETALLONES: LowerUnaryToTwoAddr(OutMI, X86::PCMPEQDrr); break;
 
   case X86::MOV16r0:
diff --git a/lib/Target/X86/X86InstrInfo.cpp b/lib/Target/X86/X86InstrInfo.cpp
index d9b09609609..a6d9c045aa5 100644
--- a/lib/Target/X86/X86InstrInfo.cpp
+++ b/lib/Target/X86/X86InstrInfo.cpp
@@ -2514,7 +2514,9 @@ MachineInstr* X86InstrInfo::foldMemoryOperandImpl(MachineFunction &MF,
     Alignment = (*LoadMI->memoperands_begin())->getAlignment();
   else
     switch (LoadMI->getOpcode()) {
-    case X86::V_SET0:
+    case X86::V_SET0PS:
+    case X86::V_SET0PD:
+    case X86::V_SET0PI:
     case X86::V_SETALLONES:
       Alignment = 16;
       break;
@@ -2544,11 +2546,13 @@ MachineInstr* X86InstrInfo::foldMemoryOperandImpl(MachineFunction &MF,
 
   SmallVector<MachineOperand,X86AddrNumOperands> MOs;
   switch (LoadMI->getOpcode()) {
-  case X86::V_SET0:
+  case X86::V_SET0PS:
+  case X86::V_SET0PD:
+  case X86::V_SET0PI:
   case X86::V_SETALLONES:
   case X86::FsFLD0SD:
   case X86::FsFLD0SS: {
-    // Folding a V_SET0 or V_SETALLONES as a load, to ease register pressure.
+    // Folding a V_SET0P? or V_SETALLONES as a load, to ease register pressure.
     // Create a constant-pool entry and operands to load from it.
 
     // Medium and large mode can't fold loads this way.
@@ -3675,6 +3679,7 @@ static const unsigned ReplaceableInstrs[][3] = {
   { X86::ANDPSrr,    X86::ANDPDrr,   X86::PANDrr    },
   { X86::ORPSrm,     X86::ORPDrm,    X86::PORrm     },
   { X86::ORPSrr,     X86::ORPDrr,    X86::PORrr     },
+  { X86::V_SET0PS,   X86::V_SET0PD,  X86::V_SET0PI  },
   { X86::XORPSrm,    X86::XORPDrm,   X86::PXORrm    },
   { X86::XORPSrr,    X86::XORPDrr,   X86::PXORrr    },
 };
diff --git a/lib/Target/X86/X86InstrSSE.td b/lib/Target/X86/X86InstrSSE.td
index f74ca9dd7bf..5a87314c324 100644
--- a/lib/Target/X86/X86InstrSSE.td
+++ b/lib/Target/X86/X86InstrSSE.td
@@ -1115,15 +1115,19 @@ def STMXCSR : PSI<0xAE, MRM3m, (outs), (ins i32mem:$dst),
 // load of an all-zeros value if folding it would be beneficial.
 // FIXME: Change encoding to pseudo!
 let isReMaterializable = 1, isAsCheapAsAMove = 1, canFoldAsLoad = 1,
-    isCodeGenOnly = 1 in
-def V_SET0 : PSI<0x57, MRMInitReg, (outs VR128:$dst), (ins), "",
+    isCodeGenOnly = 1 in {
+def V_SET0PS : PSI<0x57, MRMInitReg, (outs VR128:$dst), (ins), "",
+                 [(set VR128:$dst, (v4f32 immAllZerosV))]>;
+def V_SET0PD : PDI<0x57, MRMInitReg, (outs VR128:$dst), (ins), "",
+                 [(set VR128:$dst, (v2f64 immAllZerosV))]>;
+let ExeDomain = SSEPackedInt in
+def V_SET0PI : PDI<0xEF, MRMInitReg, (outs VR128:$dst), (ins), "",
                  [(set VR128:$dst, (v4i32 immAllZerosV))]>;
+}
 
-def : Pat<(v2i64 immAllZerosV), (V_SET0)>;
-def : Pat<(v8i16 immAllZerosV), (V_SET0)>;
-def : Pat<(v16i8 immAllZerosV), (V_SET0)>;
-def : Pat<(v2f64 immAllZerosV), (V_SET0)>;
-def : Pat<(v4f32 immAllZerosV), (V_SET0)>;
+def : Pat<(v2i64 immAllZerosV), (V_SET0PI)>;
+def : Pat<(v8i16 immAllZerosV), (V_SET0PI)>;
+def : Pat<(v16i8 immAllZerosV), (V_SET0PI)>;
 
 def : Pat<(f32 (vector_extract (v4f32 VR128:$src), (iPTR 0))),
           (f32 (EXTRACT_SUBREG (v4f32 VR128:$src), x86_subreg_ss))>;
@@ -3026,14 +3030,14 @@ let Predicates = [HasSSE2] in {
 let AddedComplexity = 15 in {
 // Zeroing a VR128 then do a MOVS{S|D} to the lower bits.
 def : Pat<(v2f64 (X86vzmovl (v2f64 (scalar_to_vector FR64:$src)))),
-          (MOVSDrr (v2f64 (V_SET0)), FR64:$src)>;
+          (MOVSDrr (v2f64 (V_SET0PS)), FR64:$src)>;
 def : Pat<(v4f32 (X86vzmovl (v4f32 (scalar_to_vector FR32:$src)))),
-          (MOVSSrr (v4f32 (V_SET0)), FR32:$src)>;
+          (MOVSSrr (v4f32 (V_SET0PS)), FR32:$src)>;
 def : Pat<(v4f32 (X86vzmovl (v4f32 VR128:$src))),
-          (MOVSSrr (v4f32 (V_SET0)),
+          (MOVSSrr (v4f32 (V_SET0PS)),
                    (f32 (EXTRACT_SUBREG (v4f32 VR128:$src), x86_subreg_ss)))>;
 def : Pat<(v4i32 (X86vzmovl (v4i32 VR128:$src))),
-          (MOVSSrr (v4i32 (V_SET0)),
+          (MOVSSrr (v4i32 (V_SET0PI)),
                    (EXTRACT_SUBREG (v4i32 VR128:$src), x86_subreg_ss))>;
 }
 
diff --git a/test/CodeGen/X86/vec_return.ll b/test/CodeGen/X86/vec_return.ll
index 66762b4a060..676be9b7179 100644
--- a/test/CodeGen/X86/vec_return.ll
+++ b/test/CodeGen/X86/vec_return.ll
@@ -1,5 +1,5 @@
 ; RUN: llc < %s -march=x86 -mattr=+sse2 > %t
-; RUN: grep xorps %t | count 1
+; RUN: grep pxor %t | count 1
 ; RUN: grep movaps %t | count 1
 ; RUN: not grep shuf %t
 
diff --git a/test/CodeGen/X86/vec_shuffle-7.ll b/test/CodeGen/X86/vec_shuffle-7.ll
index 4cdca09c72f..64bd6a3c83b 100644
--- a/test/CodeGen/X86/vec_shuffle-7.ll
+++ b/test/CodeGen/X86/vec_shuffle-7.ll
@@ -1,5 +1,5 @@
 ; RUN: llc < %s -march=x86 -mattr=+sse2 -o %t
-; RUN: grep xorps %t | count 1
+; RUN: grep pxor %t | count 1
 ; RUN: not grep shufps %t
 
 define void @test() {
diff --git a/test/CodeGen/X86/vec_shuffle-9.ll b/test/CodeGen/X86/vec_shuffle-9.ll
index fc16a26b615..07195869b8c 100644
--- a/test/CodeGen/X86/vec_shuffle-9.ll
+++ b/test/CodeGen/X86/vec_shuffle-9.ll
@@ -1,7 +1,7 @@
 ; RUN: llc < %s -march=x86 -mattr=+sse2 | FileCheck %s
 
 define <4 x i32> @test(i8** %ptr) {
-; CHECK: xorps
+; CHECK: pxor
 ; CHECK: punpcklbw
 ; CHECK: punpcklwd
 
diff --git a/test/CodeGen/X86/vec_zero.ll b/test/CodeGen/X86/vec_zero.ll
index ae5af586cdc..4d1f05629b4 100644
--- a/test/CodeGen/X86/vec_zero.ll
+++ b/test/CodeGen/X86/vec_zero.ll
@@ -1,5 +1,6 @@
-; RUN: llc < %s -march=x86 -mattr=+sse2 | grep xorps | count 2
+; RUN: llc < %s -march=x86 -mattr=+sse2 | FileCheck %s
 
+; CHECK: xorps
 define void @foo(<4 x float>* %P) {
         %T = load <4 x float>* %P               ; <<4 x float>> [#uses=1]
         %S = fadd <4 x float> zeroinitializer, %T                ; <<4 x float>> [#uses=1]
@@ -7,6 +8,7 @@ define void @foo(<4 x float>* %P) {
         ret void
 }
 
+; CHECK: pxor
 define void @bar(<4 x i32>* %P) {
         %T = load <4 x i32>* %P         ; <<4 x i32>> [#uses=1]
         %S = add <4 x i32> zeroinitializer, %T          ; <<4 x i32>> [#uses=1]
diff --git a/test/CodeGen/X86/vec_zero_cse.ll b/test/CodeGen/X86/vec_zero_cse.ll
index 296378c6e9f..3b15d4cc407 100644
--- a/test/CodeGen/X86/vec_zero_cse.ll
+++ b/test/CodeGen/X86/vec_zero_cse.ll
@@ -1,5 +1,4 @@
-; RUN: llc < %s -relocation-model=static -march=x86 -mcpu=yonah | grep pxor | count 1
-; RUN: llc < %s -relocation-model=static -march=x86 -mcpu=yonah | grep xorps | count 1
+; RUN: llc < %s -relocation-model=static -march=x86 -mcpu=yonah | grep pxor | count 2
 ; RUN: llc < %s -relocation-model=static -march=x86 -mcpu=yonah | grep pcmpeqd | count 2
 
 @M1 = external global <1 x i64>
diff --git a/test/CodeGen/X86/xor.ll b/test/CodeGen/X86/xor.ll
index 9bfff8a06a8..f270d9d56e2 100644
--- a/test/CodeGen/X86/xor.ll
+++ b/test/CodeGen/X86/xor.ll
@@ -7,7 +7,7 @@ define <4 x i32> @test1() nounwind {
 	ret <4 x i32> %tmp
         
 ; X32: test1:
-; X32:	xorps	%xmm0, %xmm0
+; X32:	pxor	%xmm0, %xmm0
 ; X32:	ret
 }