[SelectionDAG] Optimization of BITREVERSE legalization for power-of-2 integer scalar/vector types

[RKSimon]

Summary

An extension of D19978, this patch replaces the default BITREVERSE evaluation of individual bit masks+shifts with block mask+shifts when we have integer elements of power-of-2 bits in size.

After calling BSWAP to reverse the order of the constituent bytes (which typically follows a similar approach), every neighbouring 4-bits, 2-bits and finally 1-bit pairs are masked off and swapped over with shifts.

In doing so we can significantly reduce the number of operations required.

[spatel]

I haven't looked at bitreverse before this, so James or someone else should review too. A few nits inline, but otherwise LGTM.

Is there something that guarantees that an i2 or i4 type doesn't get in here? Should we assert that VT is i8 or larger?

BTW, there's a bug in the LangRef:
"The llvm.bitreverse.iN intrinsic returns an i16 value"

lib/CodeGen/SelectionDAG/LegalizeDAG.cpp

43–1

Open code?

45–46

If I'm reading this sentence correctly, it should be:
"If we can, perform"

test/CodeGen/X86/vector-bitreverse.ll

-1304–1

I wish we printed hex asm comments for 8-bit immediates too. :)

[RKSimon]

Updated based on Sanjay's comments.

[RKSimon]

James - any comments?

[RKSimon]

ping

[jmolloy]

Hi,

Apologies, I was away on vacation.

Difficult to see in the x86 test noise, but do we have any tests that check the original, unoptimized behaviour? (like using i7 or i24)?

Cheers,

James

[RKSimon]

Updated with examples of non-legal types (i4 and i24)

[RKSimon]

Readded the aarch64 bitreverse tests

[RKSimon]

ping?

[RKSimon]

ping2

[jmolloy]

Sorry Simon,

LGTM.

lib/CodeGen/SelectionDAG/LegalizeDAG.cpp

 #include "llvm/Target/TargetSubtargetInfo.h"
 using namespace llvm;
 
 #define DEBUG_TYPE "legalizedag"
 
 namespace {
 
 struct FloatSignAsInt;
 
 //===----------------------------------------------------------------------===//
 /// This takes an arbitrary SelectionDAG as input and
 /// hacks on it until the target machine can handle it.  This involves
 /// eliminating value sizes the machine cannot handle (promoting small sizes to
 /// large sizes or splitting up large values into small values) as well as
 /// eliminating operations the machine cannot handle.
 ///
 /// This code also does a small amount of optimization and recognition of idioms
 /// as part of its processing.  For example, if a target does not support a
SDValue SelectionDAGLegalize::PromoteLegalFP_TO_INT(SDValue LegalOp, EVT DestVT,
   // Okay, we found the operation and type to use.
   SDValue Operation = DAG.getNode(OpToUse, dl, NewOutTy, LegalOp);
 
   // Truncate the result of the extended FP_TO_*INT operation to the desired
   // size.
   return DAG.getNode(ISD::TRUNCATE, dl, DestVT, Operation);
 }
 
-/// Open code the operations for BITREVERSE.
+/// Legalize a BITREVERSE scalar/vector operation as a series of mask + shifts.
 SDValue SelectionDAGLegalize::ExpandBITREVERSE(SDValue Op, const SDLoc &dl) {
   EVT VT = Op.getValueType();
   EVT SHVT = TLI.getShiftAmountTy(VT, DAG.getDataLayout());
   unsigned Sz = VT.getScalarSizeInBits();
 
-  SDValue Tmp, Tmp2;
+  SDValue Tmp, Tmp2, Tmp3;
+
+  // If we can, perform BSWAP first and then the mask+swap the i4, then i2
+  // and finally the i1 pairs.
+  // TODO: We can easily support i4/i2 legal types if any target ever does.
+  if (Sz >= 8 && isPowerOf2_32(Sz)) {
+    // Create the masks - repeating the pattern every byte.
+    APInt MaskHi4(Sz, 0), MaskHi2(Sz, 0), MaskHi1(Sz, 0);
+    APInt MaskLo4(Sz, 0), MaskLo2(Sz, 0), MaskLo1(Sz, 0);
+    for (unsigned J = 0; J != Sz; J += 8) {
+      MaskHi4 = MaskHi4.Or(APInt(Sz, 0xF0ull << J));
+      MaskLo4 = MaskLo4.Or(APInt(Sz, 0x0Full << J));
+      MaskHi2 = MaskHi2.Or(APInt(Sz, 0xCCull << J));
+      MaskLo2 = MaskLo2.Or(APInt(Sz, 0x33ull << J));
+      MaskHi1 = MaskHi1.Or(APInt(Sz, 0xAAull << J));
+      MaskLo1 = MaskLo1.Or(APInt(Sz, 0x55ull << J));
+    }
+
+    // BSWAP if the type is wider than a single byte.
+    Tmp = (Sz > 8 ? DAG.getNode(ISD::BSWAP, dl, VT, Op) : Op);
+
+    // swap i4: ((V & 0xF0) >> 4) | ((V & 0x0F) << 4)
+    Tmp2 = DAG.getNode(ISD::AND, dl, VT, Tmp, DAG.getConstant(MaskHi4, dl, VT));
+    Tmp3 = DAG.getNode(ISD::AND, dl, VT, Tmp, DAG.getConstant(MaskLo4, dl, VT));
+    Tmp2 = DAG.getNode(ISD::SRL, dl, VT, Tmp2, DAG.getConstant(4, dl, VT));
+    Tmp3 = DAG.getNode(ISD::SHL, dl, VT, Tmp3, DAG.getConstant(4, dl, VT));
+    Tmp = DAG.getNode(ISD::OR, dl, VT, Tmp2, Tmp3);
+
+    // swap i2: ((V & 0xCC) >> 2) | ((V & 0x33) << 2)
+    Tmp2 = DAG.getNode(ISD::AND, dl, VT, Tmp, DAG.getConstant(MaskHi2, dl, VT));
+    Tmp3 = DAG.getNode(ISD::AND, dl, VT, Tmp, DAG.getConstant(MaskLo2, dl, VT));
+    Tmp2 = DAG.getNode(ISD::SRL, dl, VT, Tmp2, DAG.getConstant(2, dl, VT));
+    Tmp3 = DAG.getNode(ISD::SHL, dl, VT, Tmp3, DAG.getConstant(2, dl, VT));
+    Tmp = DAG.getNode(ISD::OR, dl, VT, Tmp2, Tmp3);
+
+    // swap i1: ((V & 0xAA) >> 1) | ((V & 0x55) << 1)
+    Tmp2 = DAG.getNode(ISD::AND, dl, VT, Tmp, DAG.getConstant(MaskHi1, dl, VT));
+    Tmp3 = DAG.getNode(ISD::AND, dl, VT, Tmp, DAG.getConstant(MaskLo1, dl, VT));
+    Tmp2 = DAG.getNode(ISD::SRL, dl, VT, Tmp2, DAG.getConstant(1, dl, VT));
+    Tmp3 = DAG.getNode(ISD::SHL, dl, VT, Tmp3, DAG.getConstant(1, dl, VT));
+    Tmp = DAG.getNode(ISD::OR, dl, VT, Tmp2, Tmp3);
+    return Tmp;
+  }
+
   Tmp = DAG.getConstant(0, dl, VT);
   for (unsigned I = 0, J = Sz-1; I < Sz; ++I, --J) {
     if (I < J)
       Tmp2 =
           DAG.getNode(ISD::SHL, dl, VT, Op, DAG.getConstant(J - I, dl, SHVT));
     else
       Tmp2 =
           DAG.getNode(ISD::SRL, dl, VT, Op, DAG.getConstant(I - J, dl, SHVT));
 
     APInt Shift(Sz, 1);
     Shift = Shift.shl(J);
     Tmp2 = DAG.getNode(ISD::AND, dl, VT, Tmp2, DAG.getConstant(Shift, dl, VT));
     Tmp = DAG.getNode(ISD::OR, dl, VT, Tmp, Tmp2);
   }
 
   return Tmp;
 }
 
 /// Open code the operations for BSWAP of the specified operation.
 SDValue SelectionDAGLegalize::ExpandBSWAP(SDValue Op, const SDLoc &dl) {
   EVT VT = Op.getValueType();
   EVT SHVT = TLI.getShiftAmountTy(VT, DAG.getDataLayout());
   SDValue Tmp1, Tmp2, Tmp3, Tmp4, Tmp5, Tmp6, Tmp7, Tmp8;
-  switch (VT.getSimpleVT().SimpleTy) {
+  switch (VT.getSimpleVT().getScalarType().SimpleTy) {
   default: llvm_unreachable("Unhandled Expand type in BSWAP!");
   case MVT::i16:
     Tmp2 = DAG.getNode(ISD::SHL, dl, VT, Op, DAG.getConstant(8, dl, SHVT));
     Tmp1 = DAG.getNode(ISD::SRL, dl, VT, Op, DAG.getConstant(8, dl, SHVT));
     return DAG.getNode(ISD::OR, dl, VT, Tmp1, Tmp2);
   case MVT::i32:
     Tmp4 = DAG.getNode(ISD::SHL, dl, VT, Op, DAG.getConstant(24, dl, SHVT));
     Tmp3 = DAG.getNode(ISD::SHL, dl, VT, Op, DAG.getConstant(8, dl, SHVT));

test/CodeGen/AArch64/bitreverse.ll

 ; CHECK: rev32
 ; CHECK: ushr
   %b = call <2 x i16> @llvm.bitreverse.v2i16(<2 x i16> %a)
   ret <2 x i16> %b
 }
 
 declare i8 @llvm.bitreverse.i8(i8) readnone
 
-; Unfortunately some of the shift-and-inserts become BFIs, and some do not :(
 define i8 @g(i8 %a) {
 ; CHECK-LABEL: g:
-; CHECK-DAG: lsr [[S5:w.*]], w0, #5
-; CHECK-DAG: lsr [[S4:w.*]], w0, #4
-; CHECK-DAG: lsr [[S3:w.*]], w0, #3
-; CHECK-DAG: lsr [[S2:w.*]], w0, #2
-; CHECK-DAG: lsl [[L1:w.*]], w0, #29
-; CHECK-DAG: lsl [[L2:w.*]], w0, #19
-; CHECK-DAG: lsl [[L3:w.*]], w0, #17
-
-; CHECK-DAG: and [[T1:w.*]], [[L1]], #0x40000000
-; CHECK-DAG: bfi [[T1]], w0, #31, #1
-; CHECK-DAG: bfi [[T1]], [[S2]], #29, #1
-; CHECK-DAG: bfi [[T1]], [[S3]], #28, #1
-; CHECK-DAG: bfi [[T1]], [[S4]], #27, #1
-; CHECK-DAG: bfi [[T1]], [[S5]], #26, #1
-; CHECK-DAG: and [[T2:w.*]], [[L2]], #0x2000000
-; CHECK-DAG: and [[T3:w.*]], [[L3]], #0x1000000
-; CHECK-DAG: orr [[T4:w.*]], [[T1]], [[T2]]
-; CHECK-DAG: orr [[T5:w.*]], [[T4]], [[T3]]
-; CHECK:     lsr w0, [[T5]], #24
+; CHECK-DAG: rev [[RV:w.*]], w0
+; CHECK-DAG: and [[L4:w.*]], [[RV]], #0xf0f0f0f
+; CHECK-DAG: and [[H4:w.*]], [[RV]], #0xf0f0f0f0
+; CHECK-DAG: lsr [[S4:w.*]], [[H4]], #4
+; CHECK-DAG: orr [[R4:w.*]], [[S4]], [[L4]], lsl #4
+
+; CHECK-DAG: and [[L2:w.*]], [[R4]], #0x33333333
+; CHECK-DAG: and [[H2:w.*]], [[R4]], #0xcccccccc
+; CHECK-DAG: lsr [[S2:w.*]], [[H2]], #2
+; CHECK-DAG: orr [[R2:w.*]], [[S2]], [[L2]], lsl #2
+
+; CHECK-DAG: mov [[P1:w.*]], #1426063360
+; CHECK-DAG: mov [[N1:w.*]], #-1442840576
+; CHECK-DAG: and [[L1:w.*]], [[R2]], [[P1]]
+; CHECK-DAG: and [[H1:w.*]], [[R2]], [[N1]]
+; CHECK-DAG: lsr [[S1:w.*]], [[H1]], #1
+; CHECK-DAG: orr [[R1:w.*]], [[S1]], [[L1]], lsl #1
 
+; CHECK-DAG: lsr w0, [[R1]], #24
+; CHECK-DAG: ret
   %b = call i8 @llvm.bitreverse.i8(i8 %a)
   ret i8 %b
 }
 
 declare <8 x i8> @llvm.bitreverse.v8i8(<8 x i8>) readnone
 
 define <8 x i8> @g_vec(<8 x i8> %a) {
-; Try and match as much of the sequence as precisely as possible.
+; CHECK-DAG: movi [[M1:v.*]], #15
+; CHECK-DAG: movi [[M2:v.*]], #240
+; CHECK:     and  [[A1:v.*]], v0.8b, [[M1]]
+; CHECK:     and  [[A2:v.*]], v0.8b, [[M2]]
+; CHECK-DAG: shl  [[L4:v.*]], [[A1]], #4
+; CHECK-DAG: ushr [[R4:v.*]], [[A2]], #4
+; CHECK-DAG: orr  [[V4:v.*]], [[R4]], [[L4]]
+
+; CHECK-DAG: movi [[M3:v.*]], #51
+; CHECK-DAG: movi [[M4:v.*]], #204
+; CHECK:     and  [[A3:v.*]], [[V4]], [[M3]]
+; CHECK:     and  [[A4:v.*]], [[V4]], [[M4]]
+; CHECK-DAG: shl  [[L2:v.*]], [[A3]], #2
+; CHECK-DAG: ushr [[R2:v.*]], [[A4]], #2
+; CHECK-DAG: orr  [[V2:v.*]], [[R2]], [[L2]]
+
+; CHECK-DAG: movi [[M5:v.*]], #85
+; CHECK-DAG: movi [[M6:v.*]], #170
+; CHECK:     and  [[A5:v.*]], [[V2]], [[M5]]
+; CHECK:     and  [[A6:v.*]], [[V2]], [[M6]]
+; CHECK-DAG: shl  [[L1:v.*]], [[A5]], #1
+; CHECK-DAG: ushr [[R1:v.*]], [[A6]], #1
+; CHECK:     orr  [[V1:v.*]], [[R1]], [[L1]]
 
-; CHECK-LABEL: g_vec:
-; CHECK-DAG: movi [[M1:v.*]], #128
-; CHECK-DAG: movi [[M2:v.*]], #64
-; CHECK-DAG: movi [[M3:v.*]], #32
-; CHECK-DAG: movi [[M4:v.*]], #16
-; CHECK-DAG: movi [[M5:v.*]], #8{{$}}
-; CHECK-DAG: movi [[M6:v.*]], #4{{$}}
-; CHECK-DAG: movi [[M7:v.*]], #2{{$}}
-; CHECK-DAG: movi [[M8:v.*]], #1{{$}}
-; CHECK-DAG: shl  [[S1:v.*]], v0.8b, #7
-; CHECK-DAG: shl  [[S2:v.*]], v0.8b, #5
-; CHECK-DAG: shl  [[S3:v.*]], v0.8b, #3
-; CHECK-DAG: shl  [[S4:v.*]], v0.8b, #1
-; CHECK-DAG: ushr [[S5:v.*]], v0.8b, #1
-; CHECK-DAG: ushr [[S6:v.*]], v0.8b, #3
-; CHECK-DAG: ushr [[S7:v.*]], v0.8b, #5
-; CHECK-DAG: ushr [[S8:v.*]], v0.8b, #7
-; CHECK-DAG: and  [[A1:v.*]], [[S1]], [[M1]]
-; CHECK-DAG: and  [[A2:v.*]], [[S2]], [[M2]]
-; CHECK-DAG: and  [[A3:v.*]], [[S3]], [[M3]]
-; CHECK-DAG: and  [[A4:v.*]], [[S4]], [[M4]]
-; CHECK-DAG: and  [[A5:v.*]], [[S5]], [[M5]]
-; CHECK-DAG: and  [[A6:v.*]], [[S6]], [[M6]]
-; CHECK-DAG: and  [[A7:v.*]], [[S7]], [[M7]]
-; CHECK-DAG: and  [[A8:v.*]], [[S8]], [[M8]]
-
-; The rest can be ORRed together in any order; it's not worth the test
-; maintenance to match them precisely.
-; CHECK-DAG: orr
-; CHECK-DAG: orr
-; CHECK-DAG: orr
-; CHECK-DAG: orr
-; CHECK-DAG: orr
-; CHECK-DAG: orr
-; CHECK-DAG: orr
-; CHECK: ret
+; CHECK:     ret
   %b = call <8 x i8> @llvm.bitreverse.v8i8(<8 x i8> %a)
   ret <8 x i8> %b
 }

test/CodeGen/X86/bitreverse.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=i686-unknown | FileCheck %s
 
 ; These tests just check that the plumbing is in place for @llvm.bitreverse. The
 ; actual output is massive at the moment as llvm.bitreverse is not yet legal.
 
 declare <2 x i16> @llvm.bitreverse.v2i16(<2 x i16>) readnone
 
 define <2 x i16> @test_bitreverse_v2i16(<2 x i16> %a) nounwind {
 ; CHECK-LABEL: test_bitreverse_v2i16:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    pushl %edi
-; CHECK-NEXT:    pushl %esi
-; CHECK-NEXT:    movzwl {{[0-9]+}}(%esp), %edx
-; CHECK-NEXT:    movzwl {{[0-9]+}}(%esp), %eax
-; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    shll $15, %ecx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    andl $2, %esi
-; CHECK-NEXT:    shll $13, %esi
-; CHECK-NEXT:    orl %ecx, %esi
-; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    andl $4, %ecx
-; CHECK-NEXT:    shll $11, %ecx
-; CHECK-NEXT:    orl %esi, %ecx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    andl $8, %esi
-; CHECK-NEXT:    shll $9, %esi
-; CHECK-NEXT:    orl %ecx, %esi
-; CHECK-NEXT:    movl %eax, %edi
-; CHECK-NEXT:    andl $16, %edi
-; CHECK-NEXT:    shll $7, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    andl $32, %ecx
-; CHECK-NEXT:    shll $5, %ecx
-; CHECK-NEXT:    orl %edi, %ecx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    andl $64, %esi
-; CHECK-NEXT:    shll $3, %esi
-; CHECK-NEXT:    leal (%eax,%eax), %edi
-; CHECK-NEXT:    andl $256, %edi # imm = 0x100
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shrl %esi
-; CHECK-NEXT:    andl $128, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    movl %eax, %edi
-; CHECK-NEXT:    shrl $3, %edi
-; CHECK-NEXT:    andl $64, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shrl $5, %esi
-; CHECK-NEXT:    andl $32, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    movl %eax, %edi
-; CHECK-NEXT:    shrl $7, %edi
-; CHECK-NEXT:    andl $16, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shrl $9, %esi
-; CHECK-NEXT:    andl $8, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    movl %eax, %edi
-; CHECK-NEXT:    shrl $11, %edi
-; CHECK-NEXT:    andl $4, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shrl $13, %esi
-; CHECK-NEXT:    andl $2, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    shrl $15, %eax
-; CHECK-NEXT:    orl %esi, %eax
-; CHECK-NEXT:    orl %ecx, %eax
-; CHECK-NEXT:    movl %edx, %ecx
-; CHECK-NEXT:    shll $15, %ecx
-; CHECK-NEXT:    movl %edx, %esi
-; CHECK-NEXT:    andl $2, %esi
-; CHECK-NEXT:    shll $13, %esi
-; CHECK-NEXT:    orl %ecx, %esi
-; CHECK-NEXT:    movl %edx, %ecx
-; CHECK-NEXT:    andl $4, %ecx
-; CHECK-NEXT:    shll $11, %ecx
-; CHECK-NEXT:    orl %esi, %ecx
-; CHECK-NEXT:    movl %edx, %esi
-; CHECK-NEXT:    andl $8, %esi
-; CHECK-NEXT:    shll $9, %esi
-; CHECK-NEXT:    orl %ecx, %esi
-; CHECK-NEXT:    movl %edx, %edi
-; CHECK-NEXT:    andl $16, %edi
-; CHECK-NEXT:    shll $7, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %edx, %ecx
-; CHECK-NEXT:    andl $32, %ecx
-; CHECK-NEXT:    shll $5, %ecx
-; CHECK-NEXT:    orl %edi, %ecx
-; CHECK-NEXT:    movl %edx, %esi
-; CHECK-NEXT:    andl $64, %esi
-; CHECK-NEXT:    shll $3, %esi
-; CHECK-NEXT:    leal (%edx,%edx), %edi
-; CHECK-NEXT:    andl $256, %edi # imm = 0x100
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %edx, %esi
-; CHECK-NEXT:    shrl %esi
-; CHECK-NEXT:    andl $128, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    movl %edx, %edi
-; CHECK-NEXT:    shrl $3, %edi
-; CHECK-NEXT:    andl $64, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %edx, %esi
-; CHECK-NEXT:    shrl $5, %esi
-; CHECK-NEXT:    andl $32, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    movl %edx, %edi
-; CHECK-NEXT:    shrl $7, %edi
-; CHECK-NEXT:    andl $16, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %edx, %esi
-; CHECK-NEXT:    shrl $9, %esi
-; CHECK-NEXT:    andl $8, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    movl %edx, %edi
-; CHECK-NEXT:    shrl $11, %edi
-; CHECK-NEXT:    andl $4, %edi
-; CHECK-NEXT:    orl %esi, %edi
-; CHECK-NEXT:    movl %edx, %esi
-; CHECK-NEXT:    shrl $13, %esi
-; CHECK-NEXT:    andl $2, %esi
-; CHECK-NEXT:    orl %edi, %esi
-; CHECK-NEXT:    shrl $15, %edx
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    orl %ecx, %edx
+; CHECK-NEXT:    movw {{[0-9]+}}(%esp), %cx
+; CHECK-NEXT:    movw {{[0-9]+}}(%esp), %ax
+; CHECK-NEXT:    rolw $8, %ax
+; CHECK-NEXT:    movl %eax, %edx
+; CHECK-NEXT:    andl $3855, %edx # imm = 0xF0F
+; CHECK-NEXT:    shll $4, %edx
+; CHECK-NEXT:    andl $61680, %eax # imm = 0xF0F0
+; CHECK-NEXT:    shrl $4, %eax
+; CHECK-NEXT:    orl %edx, %eax
+; CHECK-NEXT:    movl %eax, %edx
+; CHECK-NEXT:    andl $13107, %edx # imm = 0x3333
+; CHECK-NEXT:    andl $52428, %eax # imm = 0xCCCC
+; CHECK-NEXT:    shrl $2, %eax
+; CHECK-NEXT:    leal (%eax,%edx,4), %eax
+; CHECK-NEXT:    movl %eax, %edx
+; CHECK-NEXT:    andl $21845, %edx # imm = 0x5555
+; CHECK-NEXT:    andl $43690, %eax # imm = 0xAAAA
+; CHECK-NEXT:    shrl %eax
+; CHECK-NEXT:    leal (%eax,%edx,2), %eax
+; CHECK-NEXT:    rolw $8, %cx
+; CHECK-NEXT:    movl %ecx, %edx
+; CHECK-NEXT:    andl $3855, %edx # imm = 0xF0F
+; CHECK-NEXT:    shll $4, %edx
+; CHECK-NEXT:    andl $61680, %ecx # imm = 0xF0F0
+; CHECK-NEXT:    shrl $4, %ecx
+; CHECK-NEXT:    orl %edx, %ecx
+; CHECK-NEXT:    movl %ecx, %edx
+; CHECK-NEXT:    andl $13107, %edx # imm = 0x3333
+; CHECK-NEXT:    andl $52428, %ecx # imm = 0xCCCC
+; CHECK-NEXT:    shrl $2, %ecx
+; CHECK-NEXT:    leal (%ecx,%edx,4), %ecx
+; CHECK-NEXT:    movl %ecx, %edx
+; CHECK-NEXT:    andl $21845, %edx # imm = 0x5555
+; CHECK-NEXT:    andl $43690, %ecx # imm = 0xAAAA
+; CHECK-NEXT:    shrl %ecx
+; CHECK-NEXT:    leal (%ecx,%edx,2), %edx
 ; CHECK-NEXT:    # kill: %AX<def> %AX<kill> %EAX<kill>
 ; CHECK-NEXT:    # kill: %DX<def> %DX<kill> %EDX<kill>
-; CHECK-NEXT:    popl %esi
-; CHECK-NEXT:    popl %edi
 ; CHECK-NEXT:    retl
   %b = call <2 x i16> @llvm.bitreverse.v2i16(<2 x i16> %a)
   ret <2 x i16> %b
 }
 
 declare i24 @llvm.bitreverse.i24(i24) readnone
 
 define i24 @test_bitreverse_i24(i24 %a) nounwind {
 ; CHECK-LABEL: test_bitreverse_i24:
 ; CHECK:       # BB#0:
-; CHECK-NEXT:    pushl %esi
 ; CHECK-NEXT:    movl {{[0-9]+}}(%esp), %eax
+; CHECK-NEXT:    bswapl %eax
 ; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    shll $31, %ecx
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    andl $2, %edx
-; CHECK-NEXT:    shll $29, %edx
-; CHECK-NEXT:    orl %ecx, %edx
-; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    andl $4, %ecx
-; CHECK-NEXT:    shll $27, %ecx
-; CHECK-NEXT:    orl %edx, %ecx
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    andl $8, %edx
-; CHECK-NEXT:    shll $25, %edx
-; CHECK-NEXT:    orl %ecx, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    andl $16, %esi
-; CHECK-NEXT:    shll $23, %esi
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    andl $32, %ecx
-; CHECK-NEXT:    shll $21, %ecx
-; CHECK-NEXT:    orl %esi, %ecx
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    andl $64, %edx
-; CHECK-NEXT:    shll $19, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shll $17, %esi
-; CHECK-NEXT:    andl $16777216, %esi # imm = 0x1000000
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shll $15, %edx
-; CHECK-NEXT:    andl $8388608, %edx # imm = 0x800000
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shll $13, %esi
-; CHECK-NEXT:    andl $4194304, %esi # imm = 0x400000
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shll $11, %edx
-; CHECK-NEXT:    andl $2097152, %edx # imm = 0x200000
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shll $9, %esi
-; CHECK-NEXT:    andl $1048576, %esi # imm = 0x100000
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shll $7, %edx
-; CHECK-NEXT:    andl $524288, %edx # imm = 0x80000
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shll $5, %esi
-; CHECK-NEXT:    andl $262144, %esi # imm = 0x40000
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    leal (,%eax,8), %edx
-; CHECK-NEXT:    andl $131072, %edx # imm = 0x20000
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    leal (%eax,%eax), %esi
-; CHECK-NEXT:    andl $65536, %esi # imm = 0x10000
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shrl %edx
-; CHECK-NEXT:    andl $32768, %edx # imm = 0x8000
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shrl $3, %esi
-; CHECK-NEXT:    andl $16384, %esi # imm = 0x4000
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shrl $5, %edx
-; CHECK-NEXT:    andl $8192, %edx # imm = 0x2000
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shrl $7, %esi
-; CHECK-NEXT:    andl $4096, %esi # imm = 0x1000
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shrl $9, %edx
-; CHECK-NEXT:    andl $2048, %edx # imm = 0x800
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    movl %eax, %esi
-; CHECK-NEXT:    shrl $11, %esi
-; CHECK-NEXT:    andl $1024, %esi # imm = 0x400
-; CHECK-NEXT:    orl %edx, %esi
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shrl $13, %edx
-; CHECK-NEXT:    andl $512, %edx # imm = 0x200
-; CHECK-NEXT:    orl %esi, %edx
-; CHECK-NEXT:    shrl $15, %eax
-; CHECK-NEXT:    andl $256, %eax # imm = 0x100
-; CHECK-NEXT:    orl %edx, %eax
+; CHECK-NEXT:    andl $252645135, %ecx # imm = 0xF0F0F0F
+; CHECK-NEXT:    shll $4, %ecx
+; CHECK-NEXT:    andl $-252645136, %eax # imm = 0xF0F0F0F0
+; CHECK-NEXT:    shrl $4, %eax
 ; CHECK-NEXT:    orl %ecx, %eax
+; CHECK-NEXT:    movl %eax, %ecx
+; CHECK-NEXT:    andl $858993459, %ecx # imm = 0x33333333
+; CHECK-NEXT:    andl $-858993460, %eax # imm = 0xCCCCCCCC
+; CHECK-NEXT:    shrl $2, %eax
+; CHECK-NEXT:    leal (%eax,%ecx,4), %eax
+; CHECK-NEXT:    movl %eax, %ecx
+; CHECK-NEXT:    andl $1431655680, %ecx # imm = 0x55555500
+; CHECK-NEXT:    andl $-1431655936, %eax # imm = 0xAAAAAA00
+; CHECK-NEXT:    shrl %eax
+; CHECK-NEXT:    leal (%eax,%ecx,2), %eax
 ; CHECK-NEXT:    shrl $8, %eax
-; CHECK-NEXT:    popl %esi
 ; CHECK-NEXT:    retl
   %b = call i24 @llvm.bitreverse.i24(i24 %a)
   ret i24 %b
 }
 
 declare i8 @llvm.bitreverse.i8(i8) readnone
 
 define i8 @test_bitreverse_i8(i8 %a) {
 ; CHECK-LABEL: test_bitreverse_i8:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    movb {{[0-9]+}}(%esp), %al
+; CHECK-NEXT:    rolb $4, %al
 ; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    shlb $7, %cl
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shlb $5, %dl
-; CHECK-NEXT:    andb $64, %dl
-; CHECK-NEXT:    movb %al, %ah
-; CHECK-NEXT:    shlb $3, %ah
-; CHECK-NEXT:    andb $32, %ah
-; CHECK-NEXT:    orb %dl, %ah
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    addb %dl, %dl
-; CHECK-NEXT:    andb $16, %dl
-; CHECK-NEXT:    orb %ah, %dl
-; CHECK-NEXT:    movb %al, %ah
-; CHECK-NEXT:    shrb %ah
-; CHECK-NEXT:    andb $8, %ah
-; CHECK-NEXT:    orb %dl, %ah
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shrb $3, %dl
-; CHECK-NEXT:    andb $4, %dl
-; CHECK-NEXT:    orb %ah, %dl
-; CHECK-NEXT:    movb %al, %ah
-; CHECK-NEXT:    shrb $5, %ah
-; CHECK-NEXT:    andb $2, %ah
-; CHECK-NEXT:    orb %dl, %ah
-; CHECK-NEXT:    shrb $7, %al
-; CHECK-NEXT:    orb %ah, %al
+; CHECK-NEXT:    andb $51, %cl
+; CHECK-NEXT:    shlb $2, %cl
+; CHECK-NEXT:    andb $-52, %al
+; CHECK-NEXT:    shrb $2, %al
+; CHECK-NEXT:    orb %cl, %al
+; CHECK-NEXT:    movl %eax, %ecx
+; CHECK-NEXT:    andb $85, %cl
+; CHECK-NEXT:    addb %cl, %cl
+; CHECK-NEXT:    andb $-86, %al
+; CHECK-NEXT:    shrb %al
 ; CHECK-NEXT:    orb %cl, %al
 ; CHECK-NEXT:    retl
   %b = call i8 @llvm.bitreverse.i8(i8 %a)
   ret i8 %b
 }
 
 declare i4 @llvm.bitreverse.i4(i4) readnone
 
 define i4 @test_bitreverse_i4(i4 %a) {
 ; CHECK-LABEL: test_bitreverse_i4:
 ; CHECK:       # BB#0:
 ; CHECK-NEXT:    movb {{[0-9]+}}(%esp), %al
+; CHECK-NEXT:    rolb $4, %al
 ; CHECK-NEXT:    movl %eax, %ecx
-; CHECK-NEXT:    shlb $7, %cl
-; CHECK-NEXT:    movl %eax, %edx
-; CHECK-NEXT:    shlb $5, %dl
-; CHECK-NEXT:    andb $64, %dl
-; CHECK-NEXT:    movb %al, %ah
-; CHECK-NEXT:    shlb $3, %ah
-; CHECK-NEXT:    andb $32, %ah
-; CHECK-NEXT:    orb %dl, %ah
-; CHECK-NEXT:    addb %al, %al
-; CHECK-NEXT:    andb $16, %al
-; CHECK-NEXT:    orb %ah, %al
+; CHECK-NEXT:    andb $51, %cl
+; CHECK-NEXT:    shlb $2, %cl
+; CHECK-NEXT:    andb $-52, %al
+; CHECK-NEXT:    shrb $2, %al
+; CHECK-NEXT:    orb %cl, %al
+; CHECK-NEXT:    movl %eax, %ecx
+; CHECK-NEXT:    andb $80, %cl
+; CHECK-NEXT:    addb %cl, %cl
+; CHECK-NEXT:    andb $-96, %al
+; CHECK-NEXT:    shrb %al
 ; CHECK-NEXT:    orb %cl, %al
 ; CHECK-NEXT:    shrb $4, %al
 ; CHECK-NEXT:    retl
   %b = call i4 @llvm.bitreverse.i4(i4 %a)
   ret i4 %b
 }
 
 ; These tests check that bitreverse(constant) calls are folded

test/CodeGen/X86/vector-bitreverse.ll

 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSE2
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+ssse3 | FileCheck %s --check-prefix=ALL --check-prefix=SSE --check-prefix=SSSE3
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefix=ALL --check-prefix=AVX  --check-prefix=AVX512  --check-prefix=AVX512F
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f,+avx512bw | FileCheck %s --check-prefix=ALL --check-prefix=AVX  --check-prefix=AVX512  --check-prefix=AVX512BW
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+xop,+avx | FileCheck %s --check-prefix=ALL --check-prefix=XOP --check-prefix=XOPAVX1
 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+xop,+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=XOP --check-prefix=XOPAVX2
 
 define i8 @test_bitreverse_i8(i8 %a) nounwind {
 ; SSE-LABEL: test_bitreverse_i8:
 ; SSE:       # BB#0:
+; SSE-NEXT:    rolb $4, %dil
 ; SSE-NEXT:    movl %edi, %eax
-; SSE-NEXT:    shlb $7, %al
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shlb $5, %cl
-; SSE-NEXT:    andb $64, %cl
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shlb $3, %dl
-; SSE-NEXT:    andb $32, %dl
-; SSE-NEXT:    orb %cl, %dl
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    addb %cl, %cl
-; SSE-NEXT:    andb $16, %cl
-; SSE-NEXT:    orb %dl, %cl
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrb %dl
-; SSE-NEXT:    andb $8, %dl
-; SSE-NEXT:    orb %cl, %dl
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrb $3, %cl
-; SSE-NEXT:    andb $4, %cl
-; SSE-NEXT:    orb %dl, %cl
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrb $5, %dl
-; SSE-NEXT:    andb $2, %dl
-; SSE-NEXT:    orb %cl, %dl
-; SSE-NEXT:    shrb $7, %dil
-; SSE-NEXT:    orb %dl, %dil
+; SSE-NEXT:    andb $51, %al
+; SSE-NEXT:    shlb $2, %al
+; SSE-NEXT:    andb $-52, %dil
+; SSE-NEXT:    shrb $2, %dil
+; SSE-NEXT:    orb %al, %dil
+; SSE-NEXT:    movl %edi, %eax
+; SSE-NEXT:    andb $85, %al
+; SSE-NEXT:    addb %al, %al
+; SSE-NEXT:    andb $-86, %dil
+; SSE-NEXT:    shrb %dil
 ; SSE-NEXT:    orb %al, %dil
 ; SSE-NEXT:    movl %edi, %eax
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: test_bitreverse_i8:
 ; AVX:       # BB#0:
+; AVX-NEXT:    rolb $4, %dil
 ; AVX-NEXT:    movl %edi, %eax
-; AVX-NEXT:    shlb $7, %al
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shlb $5, %cl
-; AVX-NEXT:    andb $64, %cl
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shlb $3, %dl
-; AVX-NEXT:    andb $32, %dl
-; AVX-NEXT:    orb %cl, %dl
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    addb %cl, %cl
-; AVX-NEXT:    andb $16, %cl
-; AVX-NEXT:    orb %dl, %cl
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrb %dl
-; AVX-NEXT:    andb $8, %dl
-; AVX-NEXT:    orb %cl, %dl
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrb $3, %cl
-; AVX-NEXT:    andb $4, %cl
-; AVX-NEXT:    orb %dl, %cl
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrb $5, %dl
-; AVX-NEXT:    andb $2, %dl
-; AVX-NEXT:    orb %cl, %dl
-; AVX-NEXT:    shrb $7, %dil
-; AVX-NEXT:    orb %dl, %dil
+; AVX-NEXT:    andb $51, %al
+; AVX-NEXT:    shlb $2, %al
+; AVX-NEXT:    andb $-52, %dil
+; AVX-NEXT:    shrb $2, %dil
+; AVX-NEXT:    orb %al, %dil
+; AVX-NEXT:    movl %edi, %eax
+; AVX-NEXT:    andb $85, %al
+; AVX-NEXT:    addb %al, %al
+; AVX-NEXT:    andb $-86, %dil
+; AVX-NEXT:    shrb %dil
 ; AVX-NEXT:    orb %al, %dil
 ; AVX-NEXT:    movl %edi, %eax
 ; AVX-NEXT:    retq
 ;
 ; XOP-LABEL: test_bitreverse_i8:
 ; XOP:       # BB#0:
 ; XOP-NEXT:    vmovd %edi, %xmm0
 ; XOP-NEXT:    vpperm {{.*}}(%rip), %xmm0, %xmm0, %xmm0
 ; XOP-NEXT:    vpextrb $0, %xmm0, %eax
 ; XOP-NEXT:    # kill: %AL<def> %AL<kill> %EAX<kill>
 ; XOP-NEXT:    retq
   %b = call i8 @llvm.bitreverse.i8(i8 %a)
   ret i8 %b
 }
 
 define i16 @test_bitreverse_i16(i16 %a) nounwind {
 ; SSE-LABEL: test_bitreverse_i16:
 ; SSE:       # BB#0:
 ; SSE-NEXT:    # kill: %EDI<def> %EDI<kill> %RDI<def>
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    andl $32768, %ecx # imm = 0x8000
+; SSE-NEXT:    rolw $8, %di
+; SSE-NEXT:    movl %edi, %eax
+; SSE-NEXT:    andl $3855, %eax # imm = 0xF0F
+; SSE-NEXT:    shll $4, %eax
+; SSE-NEXT:    andl $61680, %edi # imm = 0xF0F0
+; SSE-NEXT:    shrl $4, %edi
+; SSE-NEXT:    orl %eax, %edi
 ; SSE-NEXT:    movl %edi, %eax
-; SSE-NEXT:    shll $15, %eax
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    andl $2, %edx
-; SSE-NEXT:    shll $13, %edx
-; SSE-NEXT:    leal (%rdx,%rax), %eax
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    andl $4, %edx
-; SSE-NEXT:    shll $11, %edx
-; SSE-NEXT:    orl %edx, %eax
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    andl $8, %edx
-; SSE-NEXT:    shll $9, %edx
-; SSE-NEXT:    orl %edx, %eax
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    andl $16, %edx
-; SSE-NEXT:    shll $7, %edx
-; SSE-NEXT:    orl %edx, %eax
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    andl $32, %edx
-; SSE-NEXT:    shll $5, %edx
-; SSE-NEXT:    orl %edx, %eax
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    andl $64, %edx
-; SSE-NEXT:    shll $3, %edx
-; SSE-NEXT:    leal (%rdi,%rdi), %esi
-; SSE-NEXT:    andl $256, %esi # imm = 0x100
-; SSE-NEXT:    orl %edx, %esi
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl %edx
-; SSE-NEXT:    andl $128, %edx
-; SSE-NEXT:    orl %esi, %edx
-; SSE-NEXT:    movl %edi, %esi
-; SSE-NEXT:    shrl $3, %esi
-; SSE-NEXT:    andl $64, %esi
-; SSE-NEXT:    orl %edx, %esi
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $5, %edx
-; SSE-NEXT:    andl $32, %edx
-; SSE-NEXT:    orl %esi, %edx
-; SSE-NEXT:    movl %edi, %esi
-; SSE-NEXT:    shrl $7, %esi
-; SSE-NEXT:    andl $16, %esi
-; SSE-NEXT:    orl %edx, %esi
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $9, %edx
-; SSE-NEXT:    andl $8, %edx
-; SSE-NEXT:    orl %esi, %edx
-; SSE-NEXT:    movl %edi, %esi
-; SSE-NEXT:    shrl $11, %esi
-; SSE-NEXT:    andl $4, %esi
-; SSE-NEXT:    orl %edx, %esi
-; SSE-NEXT:    shrl $13, %edi
-; SSE-NEXT:    andl $2, %edi
-; SSE-NEXT:    orl %esi, %edi
-; SSE-NEXT:    shrl $15, %ecx
-; SSE-NEXT:    orl %edi, %ecx
-; SSE-NEXT:    orl %ecx, %eax
+; SSE-NEXT:    andl $13107, %eax # imm = 0x3333
+; SSE-NEXT:    andl $52428, %edi # imm = 0xCCCC
+; SSE-NEXT:    shrl $2, %edi
+; SSE-NEXT:    leal (%rdi,%rax,4), %eax
+; SSE-NEXT:    movl %eax, %ecx
+; SSE-NEXT:    andl $21845, %ecx # imm = 0x5555
+; SSE-NEXT:    andl $43690, %eax # imm = 0xAAAA
+; SSE-NEXT:    shrl %eax
+; SSE-NEXT:    leal (%rax,%rcx,2), %eax
 ; SSE-NEXT:    # kill: %AX<def> %AX<kill> %EAX<kill>
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: test_bitreverse_i16:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    # kill: %EDI<def> %EDI<kill> %RDI<def>
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    andl $32768, %ecx # imm = 0x8000
+; AVX-NEXT:    rolw $8, %di
 ; AVX-NEXT:    movl %edi, %eax
-; AVX-NEXT:    shll $15, %eax
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    andl $2, %edx
-; AVX-NEXT:    shll $13, %edx
-; AVX-NEXT:    leal (%rdx,%rax), %eax
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    andl $4, %edx
-; AVX-NEXT:    shll $11, %edx
-; AVX-NEXT:    orl %edx, %eax
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    andl $8, %edx
-; AVX-NEXT:    shll $9, %edx
-; AVX-NEXT:    orl %edx, %eax
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    andl $16, %edx
-; AVX-NEXT:    shll $7, %edx
-; AVX-NEXT:    orl %edx, %eax
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    andl $32, %edx
-; AVX-NEXT:    shll $5, %edx
-; AVX-NEXT:    orl %edx, %eax
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    andl $64, %edx
-; AVX-NEXT:    shll $3, %edx
-; AVX-NEXT:    leal (%rdi,%rdi), %esi
-; AVX-NEXT:    andl $256, %esi # imm = 0x100
-; AVX-NEXT:    orl %edx, %esi
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl %edx
-; AVX-NEXT:    andl $128, %edx
-; AVX-NEXT:    orl %esi, %edx
-; AVX-NEXT:    movl %edi, %esi
-; AVX-NEXT:    shrl $3, %esi
-; AVX-NEXT:    andl $64, %esi
-; AVX-NEXT:    orl %edx, %esi
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $5, %edx
-; AVX-NEXT:    andl $32, %edx
-; AVX-NEXT:    orl %esi, %edx
-; AVX-NEXT:    movl %edi, %esi
-; AVX-NEXT:    shrl $7, %esi
-; AVX-NEXT:    andl $16, %esi
-; AVX-NEXT:    orl %edx, %esi
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $9, %edx
-; AVX-NEXT:    andl $8, %edx
-; AVX-NEXT:    orl %esi, %edx
-; AVX-NEXT:    movl %edi, %esi
-; AVX-NEXT:    shrl $11, %esi
-; AVX-NEXT:    andl $4, %esi
-; AVX-NEXT:    orl %edx, %esi
-; AVX-NEXT:    shrl $13, %edi
-; AVX-NEXT:    andl $2, %edi
-; AVX-NEXT:    orl %esi, %edi
-; AVX-NEXT:    shrl $15, %ecx
-; AVX-NEXT:    orl %edi, %ecx
-; AVX-NEXT:    orl %ecx, %eax
+; AVX-NEXT:    andl $3855, %eax # imm = 0xF0F
+; AVX-NEXT:    shll $4, %eax
+; AVX-NEXT:    andl $61680, %edi # imm = 0xF0F0
+; AVX-NEXT:    shrl $4, %edi
+; AVX-NEXT:    orl %eax, %edi
+; AVX-NEXT:    movl %edi, %eax
+; AVX-NEXT:    andl $13107, %eax # imm = 0x3333
+; AVX-NEXT:    andl $52428, %edi # imm = 0xCCCC
+; AVX-NEXT:    shrl $2, %edi
+; AVX-NEXT:    leal (%rdi,%rax,4), %eax
+; AVX-NEXT:    movl %eax, %ecx
+; AVX-NEXT:    andl $21845, %ecx # imm = 0x5555
+; AVX-NEXT:    andl $43690, %eax # imm = 0xAAAA
+; AVX-NEXT:    shrl %eax
+; AVX-NEXT:    leal (%rax,%rcx,2), %eax
 ; AVX-NEXT:    # kill: %AX<def> %AX<kill> %EAX<kill>
 ; AVX-NEXT:    retq
 ;
 ; XOP-LABEL: test_bitreverse_i16:
 ; XOP:       # BB#0:
 ; XOP-NEXT:    vmovd %edi, %xmm0
 ; XOP-NEXT:    vpperm {{.*}}(%rip), %xmm0, %xmm0, %xmm0
 ; XOP-NEXT:    vmovd %xmm0, %eax
 ; XOP-NEXT:    # kill: %AX<def> %AX<kill> %EAX<kill>
 ; XOP-NEXT:    retq
   %b = call i16 @llvm.bitreverse.i16(i16 %a)
   ret i16 %b
 }
 
 define i32 @test_bitreverse_i32(i32 %a) nounwind {
 ; SSE-LABEL: test_bitreverse_i32:
 ; SSE:       # BB#0:
 ; SSE-NEXT:    # kill: %EDI<def> %EDI<kill> %RDI<def>
+; SSE-NEXT:    bswapl %edi
+; SSE-NEXT:    movl %edi, %eax
+; SSE-NEXT:    andl $252645135, %eax # imm = 0xF0F0F0F
+; SSE-NEXT:    shll $4, %eax
+; SSE-NEXT:    andl $-252645136, %edi # imm = 0xF0F0F0F0
+; SSE-NEXT:    shrl $4, %edi
+; SSE-NEXT:    orl %eax, %edi
 ; SSE-NEXT:    movl %edi, %eax
-; SSE-NEXT:    shll $31, %eax
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    andl $2, %ecx
-; SSE-NEXT:    shll $29, %ecx
-; SSE-NEXT:    leal (%rcx,%rax), %eax
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    andl $4, %ecx
-; SSE-NEXT:    shll $27, %ecx
-; SSE-NEXT:    orl %ecx, %eax
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    andl $8, %ecx
-; SSE-NEXT:    shll $25, %ecx
-; SSE-NEXT:    orl %ecx, %eax
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    andl $16, %ecx
-; SSE-NEXT:    shll $23, %ecx
-; SSE-NEXT:    orl %ecx, %eax
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    andl $32, %ecx
-; SSE-NEXT:    shll $21, %ecx
-; SSE-NEXT:    orl %ecx, %eax
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    andl $64, %ecx
-; SSE-NEXT:    shll $19, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shll $17, %edx
-; SSE-NEXT:    andl $16777216, %edx # imm = 0x1000000
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shll $15, %ecx
-; SSE-NEXT:    andl $8388608, %ecx # imm = 0x800000
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shll $13, %edx
-; SSE-NEXT:    andl $4194304, %edx # imm = 0x400000
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shll $11, %ecx
-; SSE-NEXT:    andl $2097152, %ecx # imm = 0x200000
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shll $9, %edx
-; SSE-NEXT:    andl $1048576, %edx # imm = 0x100000
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shll $7, %ecx
-; SSE-NEXT:    andl $524288, %ecx # imm = 0x80000
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shll $5, %edx
-; SSE-NEXT:    andl $262144, %edx # imm = 0x40000
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    leal (,%rdi,8), %ecx
-; SSE-NEXT:    andl $131072, %ecx # imm = 0x20000
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    leal (%rdi,%rdi), %edx
-; SSE-NEXT:    andl $65536, %edx # imm = 0x10000
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl %ecx
-; SSE-NEXT:    andl $32768, %ecx # imm = 0x8000
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $3, %edx
-; SSE-NEXT:    andl $16384, %edx # imm = 0x4000
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl $5, %ecx
-; SSE-NEXT:    andl $8192, %ecx # imm = 0x2000
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $7, %edx
-; SSE-NEXT:    andl $4096, %edx # imm = 0x1000
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl $9, %ecx
-; SSE-NEXT:    andl $2048, %ecx # imm = 0x800
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $11, %edx
-; SSE-NEXT:    andl $1024, %edx # imm = 0x400
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl $13, %ecx
-; SSE-NEXT:    andl $512, %ecx # imm = 0x200
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $15, %edx
-; SSE-NEXT:    andl $256, %edx # imm = 0x100
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl $17, %ecx
-; SSE-NEXT:    andl $128, %ecx
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $19, %edx
-; SSE-NEXT:    andl $64, %edx
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl $21, %ecx
-; SSE-NEXT:    andl $32, %ecx
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $23, %edx
-; SSE-NEXT:    andl $16, %edx
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl $25, %ecx
-; SSE-NEXT:    andl $8, %ecx
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    movl %edi, %edx
-; SSE-NEXT:    shrl $27, %edx
-; SSE-NEXT:    andl $4, %edx
-; SSE-NEXT:    orl %ecx, %edx
-; SSE-NEXT:    movl %edi, %ecx
-; SSE-NEXT:    shrl $29, %ecx
-; SSE-NEXT:    andl $2, %ecx
-; SSE-NEXT:    orl %edx, %ecx
-; SSE-NEXT:    shrl $31, %edi
-; SSE-NEXT:    orl %ecx, %edi
-; SSE-NEXT:    orl %edi, %eax
+; SSE-NEXT:    andl $858993459, %eax # imm = 0x33333333
+; SSE-NEXT:    andl $-858993460, %edi # imm = 0xCCCCCCCC
+; SSE-NEXT:    shrl $2, %edi
+; SSE-NEXT:    leal (%rdi,%rax,4), %eax
+; SSE-NEXT:    movl %eax, %ecx
+; SSE-NEXT:    andl $1431655765, %ecx # imm = 0x55555555
+; SSE-NEXT:    andl $-1431655766, %eax # imm = 0xAAAAAAAA
+; SSE-NEXT:    shrl %eax
+; SSE-NEXT:    leal (%rax,%rcx,2), %eax
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: test_bitreverse_i32:
 ; AVX:       # BB#0:
 ; AVX-NEXT:    # kill: %EDI<def> %EDI<kill> %RDI<def>
+; AVX-NEXT:    bswapl %edi
+; AVX-NEXT:    movl %edi, %eax
+; AVX-NEXT:    andl $252645135, %eax # imm = 0xF0F0F0F
+; AVX-NEXT:    shll $4, %eax
+; AVX-NEXT:    andl $-252645136, %edi # imm = 0xF0F0F0F0
+; AVX-NEXT:    shrl $4, %edi
+; AVX-NEXT:    orl %eax, %edi
 ; AVX-NEXT:    movl %edi, %eax
-; AVX-NEXT:    shll $31, %eax
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    andl $2, %ecx
-; AVX-NEXT:    shll $29, %ecx
-; AVX-NEXT:    leal (%rcx,%rax), %eax
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    andl $4, %ecx
-; AVX-NEXT:    shll $27, %ecx
-; AVX-NEXT:    orl %ecx, %eax
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    andl $8, %ecx
-; AVX-NEXT:    shll $25, %ecx
-; AVX-NEXT:    orl %ecx, %eax
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    andl $16, %ecx
-; AVX-NEXT:    shll $23, %ecx
-; AVX-NEXT:    orl %ecx, %eax
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    andl $32, %ecx
-; AVX-NEXT:    shll $21, %ecx
-; AVX-NEXT:    orl %ecx, %eax
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    andl $64, %ecx
-; AVX-NEXT:    shll $19, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shll $17, %edx
-; AVX-NEXT:    andl $16777216, %edx # imm = 0x1000000
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shll $15, %ecx
-; AVX-NEXT:    andl $8388608, %ecx # imm = 0x800000
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shll $13, %edx
-; AVX-NEXT:    andl $4194304, %edx # imm = 0x400000
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shll $11, %ecx
-; AVX-NEXT:    andl $2097152, %ecx # imm = 0x200000
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shll $9, %edx
-; AVX-NEXT:    andl $1048576, %edx # imm = 0x100000
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shll $7, %ecx
-; AVX-NEXT:    andl $524288, %ecx # imm = 0x80000
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shll $5, %edx
-; AVX-NEXT:    andl $262144, %edx # imm = 0x40000
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    leal (,%rdi,8), %ecx
-; AVX-NEXT:    andl $131072, %ecx # imm = 0x20000
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    leal (%rdi,%rdi), %edx
-; AVX-NEXT:    andl $65536, %edx # imm = 0x10000
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl %ecx
-; AVX-NEXT:    andl $32768, %ecx # imm = 0x8000
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $3, %edx
-; AVX-NEXT:    andl $16384, %edx # imm = 0x4000
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl $5, %ecx
-; AVX-NEXT:    andl $8192, %ecx # imm = 0x2000
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $7, %edx
-; AVX-NEXT:    andl $4096, %edx # imm = 0x1000
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl $9, %ecx
-; AVX-NEXT:    andl $2048, %ecx # imm = 0x800
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $11, %edx
-; AVX-NEXT:    andl $1024, %edx # imm = 0x400
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl $13, %ecx
-; AVX-NEXT:    andl $512, %ecx # imm = 0x200
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $15, %edx
-; AVX-NEXT:    andl $256, %edx # imm = 0x100
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl $17, %ecx
-; AVX-NEXT:    andl $128, %ecx
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $19, %edx
-; AVX-NEXT:    andl $64, %edx
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl $21, %ecx
-; AVX-NEXT:    andl $32, %ecx
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $23, %edx
-; AVX-NEXT:    andl $16, %edx
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl $25, %ecx
-; AVX-NEXT:    andl $8, %ecx
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    movl %edi, %edx
-; AVX-NEXT:    shrl $27, %edx
-; AVX-NEXT:    andl $4, %edx
-; AVX-NEXT:    orl %ecx, %edx
-; AVX-NEXT:    movl %edi, %ecx
-; AVX-NEXT:    shrl $29, %ecx
-; AVX-NEXT:    andl $2, %ecx
-; AVX-NEXT:    orl %edx, %ecx
-; AVX-NEXT:    shrl $31, %edi
-; AVX-NEXT:    orl %ecx, %edi
-; AVX-NEXT:    orl %edi, %eax
+; AVX-NEXT:    andl $858993459, %eax # imm = 0x33333333
+; AVX-NEXT:    andl $-858993460, %edi # imm = 0xCCCCCCCC
+; AVX-NEXT:    shrl $2, %edi
+; AVX-NEXT:    leal (%rdi,%rax,4), %eax
+; AVX-NEXT:    movl %eax, %ecx
+; AVX-NEXT:    andl $1431655765, %ecx # imm = 0x55555555
+; AVX-NEXT:    andl $-1431655766, %eax # imm = 0xAAAAAAAA
+; AVX-NEXT:    shrl %eax
+; AVX-NEXT:    leal (%rax,%rcx,2), %eax
 ; AVX-NEXT:    retq
 ;
 ; XOP-LABEL: test_bitreverse_i32:
 ; XOP:       # BB#0:
 ; XOP-NEXT:    vmovd %edi, %xmm0
 ; XOP-NEXT:    vpperm {{.*}}(%rip), %xmm0, %xmm0, %xmm0
 ; XOP-NEXT:    vmovd %xmm0, %eax
 ; XOP-NEXT:    retq
   %b = call i32 @llvm.bitreverse.i32(i32 %a)
   ret i32 %b
 }
 
 define i64 @test_bitreverse_i64(i64 %a) nounwind {
 ; SSE-LABEL: test_bitreverse_i64:
 ; SSE:       # BB#0:
-; SSE-NEXT:    leaq (%rdi,%rdi), %rax
-; SSE-NEXT:    movabsq $4294967296, %rcx # imm = 0x100000000
+; SSE-NEXT:    bswapq %rdi
+; SSE-NEXT:    movabsq $1085102592571150095, %rax # imm = 0xF0F0F0F0F0F0F0F
+; SSE-NEXT:    andq %rdi, %rax
+; SSE-NEXT:    shlq $4, %rax
+; SSE-NEXT:    movabsq $-1085102592571150096, %rcx # imm = 0xF0F0F0F0F0F0F0F0
+; SSE-NEXT:    andq %rdi, %rcx
+; SSE-NEXT:    shrq $4, %rcx
+; SSE-NEXT:    orq %rax, %rcx
+; SSE-NEXT:    movabsq $3689348814741910323, %rax # imm = 0x3333333333333333
+; SSE-NEXT:    andq %rcx, %rax
+; SSE-NEXT:    movabsq $-3689348814741910324, %rdx # imm = 0xCCCCCCCCCCCCCCCC
+; SSE-NEXT:    andq %rcx, %rdx
+; SSE-NEXT:    shrq $2, %rdx
+; SSE-NEXT:    leaq (%rdx,%rax,4), %rax
+; SSE-NEXT:    movabsq $6148914691236517205, %rcx # imm = 0x5555555555555555
 ; SSE-NEXT:    andq %rax, %rcx
-; SSE-NEXT:    movq %rdi, %rax
-; SSE-NEXT:    shlq $63, %rax
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $2, %rdx
-; SSE-NEXT:    shlq $61, %rdx
-; SSE-NEXT:    leaq (%rdx,%rax), %rax
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $4, %rdx
-; SSE-NEXT:    shlq $59, %rdx
-; SSE-NEXT:    orq %rdx, %rax
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $8, %rdx
-; SSE-NEXT:    shlq $57, %rdx
-; SSE-NEXT:    orq %rdx, %rax
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $16, %rdx
-; SSE-NEXT:    shlq $55, %rdx
-; SSE-NEXT:    orq %rdx, %rax
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $32, %rdx
-; SSE-NEXT:    shlq $53, %rdx
-; SSE-NEXT:    orq %rdx, %rax
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $64, %rdx
-; SSE-NEXT:    shlq $51, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $128, %rsi
-; SSE-NEXT:    shlq $49, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $256, %rdx # imm = 0x100
-; SSE-NEXT:    shlq $47, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $512, %rsi # imm = 0x200
-; SSE-NEXT:    shlq $45, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $1024, %rdx # imm = 0x400
-; SSE-NEXT:    shlq $43, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $2048, %rsi # imm = 0x800
-; SSE-NEXT:    shlq $41, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $4096, %rdx # imm = 0x1000
-; SSE-NEXT:    shlq $39, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $8192, %rsi # imm = 0x2000
-; SSE-NEXT:    shlq $37, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $16384, %rdx # imm = 0x4000
-; SSE-NEXT:    shlq $35, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $32768, %rsi # imm = 0x8000
-; SSE-NEXT:    shlq $33, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $65536, %rdx # imm = 0x10000
-; SSE-NEXT:    shlq $31, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $131072, %rsi # imm = 0x20000
-; SSE-NEXT:    shlq $29, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $262144, %rdx # imm = 0x40000
-; SSE-NEXT:    shlq $27, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $524288, %rsi # imm = 0x80000
-; SSE-NEXT:    shlq $25, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $1048576, %rdx # imm = 0x100000
-; SSE-NEXT:    shlq $23, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $2097152, %rsi # imm = 0x200000
-; SSE-NEXT:    shlq $21, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $4194304, %rdx # imm = 0x400000
-; SSE-NEXT:    shlq $19, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $8388608, %rsi # imm = 0x800000
-; SSE-NEXT:    shlq $17, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $16777216, %rdx # imm = 0x1000000
-; SSE-NEXT:    shlq $15, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $33554432, %rsi # imm = 0x2000000
-; SSE-NEXT:    shlq $13, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $67108864, %rdx # imm = 0x4000000
-; SSE-NEXT:    shlq $11, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $134217728, %rsi # imm = 0x8000000
-; SSE-NEXT:    shlq $9, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $268435456, %rdx # imm = 0x10000000
-; SSE-NEXT:    shlq $7, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    movq %rdi, %rsi
-; SSE-NEXT:    andq $536870912, %rsi # imm = 0x20000000
-; SSE-NEXT:    shlq $5, %rsi
-; SSE-NEXT:    orq %rdx, %rsi
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    andq $1073741824, %rdx # imm = 0x40000000
-; SSE-NEXT:    shlq $3, %rdx
-; SSE-NEXT:    orq %rsi, %rdx
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq %rcx
-; SSE-NEXT:    andl $-2147483648, %ecx # imm = 0x80000000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $3, %rdx
-; SSE-NEXT:    andl $1073741824, %edx # imm = 0x40000000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $5, %rcx
-; SSE-NEXT:    andl $536870912, %ecx # imm = 0x20000000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $7, %rdx
-; SSE-NEXT:    andl $268435456, %edx # imm = 0x10000000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $9, %rcx
-; SSE-NEXT:    andl $134217728, %ecx # imm = 0x8000000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $11, %rdx
-; SSE-NEXT:    andl $67108864, %edx # imm = 0x4000000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $13, %rcx
-; SSE-NEXT:    andl $33554432, %ecx # imm = 0x2000000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $15, %rdx
-; SSE-NEXT:    andl $16777216, %edx # imm = 0x1000000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $17, %rcx
-; SSE-NEXT:    andl $8388608, %ecx # imm = 0x800000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $19, %rdx
-; SSE-NEXT:    andl $4194304, %edx # imm = 0x400000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $21, %rcx
-; SSE-NEXT:    andl $2097152, %ecx # imm = 0x200000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $23, %rdx
-; SSE-NEXT:    andl $1048576, %edx # imm = 0x100000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $25, %rcx
-; SSE-NEXT:    andl $524288, %ecx # imm = 0x80000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $27, %rdx
-; SSE-NEXT:    andl $262144, %edx # imm = 0x40000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $29, %rcx
-; SSE-NEXT:    andl $131072, %ecx # imm = 0x20000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $31, %rdx
-; SSE-NEXT:    andl $65536, %edx # imm = 0x10000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $33, %rcx
-; SSE-NEXT:    andl $32768, %ecx # imm = 0x8000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $35, %rdx
-; SSE-NEXT:    andl $16384, %edx # imm = 0x4000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $37, %rcx
-; SSE-NEXT:    andl $8192, %ecx # imm = 0x2000
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $39, %rdx
-; SSE-NEXT:    andl $4096, %edx # imm = 0x1000
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $41, %rcx
-; SSE-NEXT:    andl $2048, %ecx # imm = 0x800
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $43, %rdx
-; SSE-NEXT:    andl $1024, %edx # imm = 0x400
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $45, %rcx
-; SSE-NEXT:    andl $512, %ecx # imm = 0x200
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $47, %rdx
-; SSE-NEXT:    andl $256, %edx # imm = 0x100
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $49, %rcx
-; SSE-NEXT:    andl $128, %ecx
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $51, %rdx
-; SSE-NEXT:    andl $64, %edx
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $53, %rcx
-; SSE-NEXT:    andl $32, %ecx
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $55, %rdx
-; SSE-NEXT:    andl $16, %edx
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $57, %rcx
-; SSE-NEXT:    andl $8, %ecx
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    movq %rdi, %rdx
-; SSE-NEXT:    shrq $59, %rdx
-; SSE-NEXT:    andl $4, %edx
-; SSE-NEXT:    orq %rcx, %rdx
-; SSE-NEXT:    movq %rdi, %rcx
-; SSE-NEXT:    shrq $61, %rcx
-; SSE-NEXT:    andl $2, %ecx
-; SSE-NEXT:    orq %rdx, %rcx
-; SSE-NEXT:    shrq $63, %rdi
-; SSE-NEXT:    orq %rcx, %rdi
-; SSE-NEXT:    orq %rdi, %rax
+; SSE-NEXT:    movabsq $-6148914691236517206, %rdx # imm = 0xAAAAAAAAAAAAAAAA
+; SSE-NEXT:    andq %rax, %rdx
+; SSE-NEXT:    shrq %rdx
+; SSE-NEXT:    leaq (%rdx,%rcx,2), %rax
 ; SSE-NEXT:    retq
 ;
 ; AVX-LABEL: test_bitreverse_i64:
 ; AVX:       # BB#0:
-; AVX-NEXT:    leaq (%rdi,%rdi), %rax
-; AVX-NEXT:    movabsq $4294967296, %rcx # imm = 0x100000000
+; AVX-NEXT:    bswapq %rdi
+; AVX-NEXT:    movabsq $1085102592571150095, %rax # imm = 0xF0F0F0F0F0F0F0F
+; AVX-NEXT:    andq %rdi, %rax
+; AVX-NEXT:    shlq $4, %rax
+; AVX-NEXT:    movabsq $-1085102592571150096, %rcx # imm = 0xF0F0F0F0F0F0F0F0
+; AVX-NEXT:    andq %rdi, %rcx
+; AVX-NEXT:    shrq $4, %rcx
+; AVX-NEXT:    orq %rax, %rcx
+; AVX-NEXT:    movabsq $3689348814741910323, %rax # imm = 0x3333333333333333
+; AVX-NEXT:    andq %rcx, %rax
+; AVX-NEXT:    movabsq $-3689348814741910324, %rdx # imm = 0xCCCCCCCCCCCCCCCC
+; AVX-NEXT:    andq %rcx, %rdx
+; AVX-NEXT:    shrq $2, %rdx
+; AVX-NEXT:    leaq (%rdx,%rax,4), %rax
+; AVX-NEXT:    movabsq $6148914691236517205, %rcx # imm = 0x5555555555555555
 ; AVX-NEXT:    andq %rax, %rcx
-; AVX-NEXT:    movq %rdi, %rax
-; AVX-NEXT:    shlq $63, %rax
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $2, %rdx
-; AVX-NEXT:    shlq $61, %rdx
-; AVX-NEXT:    leaq (%rdx,%rax), %rax
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $4, %rdx
-; AVX-NEXT:    shlq $59, %rdx
-; AVX-NEXT:    orq %rdx, %rax
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $8, %rdx
-; AVX-NEXT:    shlq $57, %rdx
-; AVX-NEXT:    orq %rdx, %rax
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $16, %rdx
-; AVX-NEXT:    shlq $55, %rdx
-; AVX-NEXT:    orq %rdx, %rax
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $32, %rdx
-; AVX-NEXT:    shlq $53, %rdx
-; AVX-NEXT:    orq %rdx, %rax
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $64, %rdx
-; AVX-NEXT:    shlq $51, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $128, %rsi
-; AVX-NEXT:    shlq $49, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $256, %rdx # imm = 0x100
-; AVX-NEXT:    shlq $47, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $512, %rsi # imm = 0x200
-; AVX-NEXT:    shlq $45, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $1024, %rdx # imm = 0x400
-; AVX-NEXT:    shlq $43, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $2048, %rsi # imm = 0x800
-; AVX-NEXT:    shlq $41, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $4096, %rdx # imm = 0x1000
-; AVX-NEXT:    shlq $39, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $8192, %rsi # imm = 0x2000
-; AVX-NEXT:    shlq $37, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $16384, %rdx # imm = 0x4000
-; AVX-NEXT:    shlq $35, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $32768, %rsi # imm = 0x8000
-; AVX-NEXT:    shlq $33, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $65536, %rdx # imm = 0x10000
-; AVX-NEXT:    shlq $31, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $131072, %rsi # imm = 0x20000
-; AVX-NEXT:    shlq $29, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $262144, %rdx # imm = 0x40000
-; AVX-NEXT:    shlq $27, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $524288, %rsi # imm = 0x80000
-; AVX-NEXT:    shlq $25, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $1048576, %rdx # imm = 0x100000
-; AVX-NEXT:    shlq $23, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $2097152, %rsi # imm = 0x200000
-; AVX-NEXT:    shlq $21, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $4194304, %rdx # imm = 0x400000
-; AVX-NEXT:    shlq $19, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $8388608, %rsi # imm = 0x800000
-; AVX-NEXT:    shlq $17, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $16777216, %rdx # imm = 0x1000000
-; AVX-NEXT:    shlq $15, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $33554432, %rsi # imm = 0x2000000
-; AVX-NEXT:    shlq $13, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $67108864, %rdx # imm = 0x4000000
-; AVX-NEXT:    shlq $11, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $134217728, %rsi # imm = 0x8000000
-; AVX-NEXT:    shlq $9, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $268435456, %rdx # imm = 0x10000000
-; AVX-NEXT:    shlq $7, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    movq %rdi, %rsi
-; AVX-NEXT:    andq $536870912, %rsi # imm = 0x20000000
-; AVX-NEXT:    shlq $5, %rsi
-; AVX-NEXT:    orq %rdx, %rsi
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    andq $1073741824, %rdx # imm = 0x40000000
-; AVX-NEXT:    shlq $3, %rdx
-; AVX-NEXT:    orq %rsi, %rdx
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq %rcx
-; AVX-NEXT:    andl $-2147483648, %ecx # imm = 0x80000000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $3, %rdx
-; AVX-NEXT:    andl $1073741824, %edx # imm = 0x40000000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $5, %rcx
-; AVX-NEXT:    andl $536870912, %ecx # imm = 0x20000000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $7, %rdx
-; AVX-NEXT:    andl $268435456, %edx # imm = 0x10000000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $9, %rcx
-; AVX-NEXT:    andl $134217728, %ecx # imm = 0x8000000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $11, %rdx
-; AVX-NEXT:    andl $67108864, %edx # imm = 0x4000000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $13, %rcx
-; AVX-NEXT:    andl $33554432, %ecx # imm = 0x2000000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $15, %rdx
-; AVX-NEXT:    andl $16777216, %edx # imm = 0x1000000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $17, %rcx
-; AVX-NEXT:    andl $8388608, %ecx # imm = 0x800000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $19, %rdx
-; AVX-NEXT:    andl $4194304, %edx # imm = 0x400000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $21, %rcx
-; AVX-NEXT:    andl $2097152, %ecx # imm = 0x200000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $23, %rdx
-; AVX-NEXT:    andl $1048576, %edx # imm = 0x100000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $25, %rcx
-; AVX-NEXT:    andl $524288, %ecx # imm = 0x80000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $27, %rdx
-; AVX-NEXT:    andl $262144, %edx # imm = 0x40000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $29, %rcx
-; AVX-NEXT:    andl $131072, %ecx # imm = 0x20000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $31, %rdx
-; AVX-NEXT:    andl $65536, %edx # imm = 0x10000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $33, %rcx
-; AVX-NEXT:    andl $32768, %ecx # imm = 0x8000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $35, %rdx
-; AVX-NEXT:    andl $16384, %edx # imm = 0x4000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $37, %rcx
-; AVX-NEXT:    andl $8192, %ecx # imm = 0x2000
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $39, %rdx
-; AVX-NEXT:    andl $4096, %edx # imm = 0x1000
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $41, %rcx
-; AVX-NEXT:    andl $2048, %ecx # imm = 0x800
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $43, %rdx
-; AVX-NEXT:    andl $1024, %edx # imm = 0x400
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $45, %rcx
-; AVX-NEXT:    andl $512, %ecx # imm = 0x200
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $47, %rdx
-; AVX-NEXT:    andl $256, %edx # imm = 0x100
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $49, %rcx
-; AVX-NEXT:    andl $128, %ecx
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $51, %rdx
-; AVX-NEXT:    andl $64, %edx
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $53, %rcx
-; AVX-NEXT:    andl $32, %ecx
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $55, %rdx
-; AVX-NEXT:    andl $16, %edx
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $57, %rcx
-; AVX-NEXT:    andl $8, %ecx
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    movq %rdi, %rdx
-; AVX-NEXT:    shrq $59, %rdx
-; AVX-NEXT:    andl $4, %edx
-; AVX-NEXT:    orq %rcx, %rdx
-; AVX-NEXT:    movq %rdi, %rcx
-; AVX-NEXT:    shrq $61, %rcx
-; AVX-NEXT:    andl $2, %ecx
-; AVX-NEXT:    orq %rdx, %rcx
-; AVX-NEXT:    shrq $63, %rdi
-; AVX-NEXT:    orq %rcx, %rdi
-; AVX-NEXT:    orq %rdi, %rax
+; AVX-NEXT:    movabsq $-6148914691236517206, %rdx # imm = 0xAAAAAAAAAAAAAAAA
+; AVX-NEXT:    andq %rax, %rdx
+; AVX-NEXT:    shrq %rdx
+; AVX-NEXT:    leaq (%rdx,%rcx,2), %rax
 ; AVX-NEXT:    retq
 ;
 ; XOP-LABEL: test_bitreverse_i64:
 ; XOP:       # BB#0:
 ; XOP-NEXT:    vmovq %rdi, %xmm0
 ; XOP-NEXT:    vpperm {{.*}}(%rip), %xmm0, %xmm0, %xmm0
 ; XOP-NEXT:    vmovq %xmm0, %rax
 ; XOP-NEXT:    retq
   %b = call i64 @llvm.bitreverse.i64(i64 %a)
   ret i64 %b
 }
 
 define <16 x i8> @test_bitreverse_v16i8(<16 x i8> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v16i8:
 ; SSE2:       # BB#0:
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    psrlw $7, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm1, %xmm1
-; SSE2-NEXT:    pand %xmm2, %xmm1
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    psllw $7, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm3, %xmm3
+; SSE2-NEXT:    pand %xmm1, %xmm2
+; SSE2-NEXT:    psllw $4, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
 ; SSE2-NEXT:    pand %xmm3, %xmm2
-; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psllw $5, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $3, %xmm4
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm4
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm4
-; SSE2-NEXT:    por %xmm3, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    paddb %xmm3, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm4, %xmm3
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psrlw $1, %xmm4
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm4
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm4
-; SSE2-NEXT:    por %xmm3, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psrlw $3, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm4, %xmm3
-; SSE2-NEXT:    psrlw $5, %xmm0
+; SSE2-NEXT:    pand %xmm3, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm1, %xmm0
+; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    psllw $2, %xmm1
+; SSE2-NEXT:    pand {{.*}}(%rip), %xmm1
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    por %xmm1, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
+; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    psrlw $1, %xmm0
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
 ; SSE2-NEXT:    por %xmm1, %xmm0
-; SSE2-NEXT:    por %xmm2, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v16i8:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2
 ; SSSE3-NEXT:    pand %xmm1, %xmm2
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm3 = [0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240]

 ; SSE2:       # BB#0:
 ; SSE2-NEXT:    pxor %xmm1, %xmm1
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
 ; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,5,4,7,6]
 ; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm0[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    packuswb %xmm2, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psllw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm2, %xmm2
-; SSE2-NEXT:    pand %xmm0, %xmm2
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psllw $5, %xmm0
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
+; SSE2-NEXT:    packuswb %xmm2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
+; SSE2-NEXT:    movdqa %xmm0, %xmm2
+; SSE2-NEXT:    pand %xmm1, %xmm2
+; SSE2-NEXT:    psllw $4, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm3, %xmm2
+; SSE2-NEXT:    pand %xmm3, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm1, %xmm0
+; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    psllw $2, %xmm1
+; SSE2-NEXT:    pand {{.*}}(%rip), %xmm1
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psllw $3, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    paddb %xmm0, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psrlw $1, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psrlw $3, %xmm0
+; SSE2-NEXT:    por %xmm1, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    psrlw $1, %xmm0
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psrlw $5, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    movdqa {{.*#+}} xmm0 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm0, %xmm0
-; SSE2-NEXT:    pand %xmm1, %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    por %xmm1, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v8i16:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[1,0,3,2,5,4,7,6,9,8,11,10,13,12,15,14]
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2
 ; SSSE3-NEXT:    pand %xmm1, %xmm2

 ; SSE2:       # BB#0:
 ; SSE2-NEXT:    pxor %xmm1, %xmm1
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
 ; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4]
 ; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm0[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm2, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psllw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm2, %xmm2
-; SSE2-NEXT:    pand %xmm0, %xmm2
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psllw $5, %xmm0
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    packuswb %xmm2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
+; SSE2-NEXT:    movdqa %xmm0, %xmm2
+; SSE2-NEXT:    pand %xmm1, %xmm2
+; SSE2-NEXT:    psllw $4, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm3, %xmm2
+; SSE2-NEXT:    pand %xmm3, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm1, %xmm0
+; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    psllw $2, %xmm1
+; SSE2-NEXT:    pand {{.*}}(%rip), %xmm1
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psllw $3, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    paddb %xmm0, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psrlw $1, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psrlw $3, %xmm0
+; SSE2-NEXT:    por %xmm1, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    psrlw $1, %xmm0
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psrlw $5, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    movdqa {{.*#+}} xmm0 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm0, %xmm0
-; SSE2-NEXT:    pand %xmm1, %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    por %xmm1, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v4i32:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[3,2,1,0,7,6,5,4,11,10,9,8,15,14,13,12]
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2
 ; SSSE3-NEXT:    pand %xmm1, %xmm2

 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
 ; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm2[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4]
 ; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm0[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm2, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psllw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm2, %xmm2
-; SSE2-NEXT:    pand %xmm0, %xmm2
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psllw $5, %xmm0
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    packuswb %xmm2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
+; SSE2-NEXT:    movdqa %xmm0, %xmm2
+; SSE2-NEXT:    pand %xmm1, %xmm2
+; SSE2-NEXT:    psllw $4, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm3, %xmm2
+; SSE2-NEXT:    pand %xmm3, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm1, %xmm0
+; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    psllw $2, %xmm1
+; SSE2-NEXT:    pand {{.*}}(%rip), %xmm1
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psllw $3, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    paddb %xmm0, %xmm0
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psrlw $1, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm0
-; SSE2-NEXT:    psrlw $3, %xmm0
+; SSE2-NEXT:    por %xmm1, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm1 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm0, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
+; SSE2-NEXT:    psrlw $1, %xmm0
 ; SSE2-NEXT:    pand {{.*}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psrlw $5, %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm3
-; SSE2-NEXT:    por %xmm0, %xmm3
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    movdqa {{.*#+}} xmm0 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm0, %xmm0
-; SSE2-NEXT:    pand %xmm1, %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    por %xmm1, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v2i64:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    pshufb {{.*#+}} xmm0 = xmm0[7,6,5,4,3,2,1,0,15,14,13,12,11,10,9,8]
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2
 ; SSSE3-NEXT:    pand %xmm1, %xmm2

 ; XOP-NEXT:    retq
   %b = call <2 x i64> @llvm.bitreverse.v2i64(<2 x i64> %a)
   ret <2 x i64> %b
 }
 
 define <32 x i8> @test_bitreverse_v32i8(<32 x i8> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v32i8:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    psllw $5, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm9
-; SSE2-NEXT:    pand %xmm9, %xmm2
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psllw $7, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm10, %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psllw $3, %xmm3
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm3
-; SSE2-NEXT:    por %xmm2, %xmm3
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    paddb %xmm2, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
-; SSE2-NEXT:    pand %xmm8, %xmm2
-; SSE2-NEXT:    por %xmm3, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psrlw $1, %xmm3
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm3
-; SSE2-NEXT:    por %xmm2, %xmm3
+; SSE2-NEXT:    pand %xmm2, %xmm3
+; SSE2-NEXT:    psllw $4, %xmm3
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm5, %xmm3
+; SSE2-NEXT:    pand %xmm5, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm2, %xmm0
+; SSE2-NEXT:    por %xmm3, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psrlw $3, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm6
-; SSE2-NEXT:    pand %xmm6, %xmm4
-; SSE2-NEXT:    por %xmm3, %xmm4
+; SSE2-NEXT:    pand %xmm3, %xmm4
+; SSE2-NEXT:    psllw $2, %xmm4
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm8, %xmm4
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm9, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    por %xmm4, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psrlw $5, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm2
-; SSE2-NEXT:    pand %xmm2, %xmm7
-; SSE2-NEXT:    por %xmm4, %xmm7
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm3, %xmm3
-; SSE2-NEXT:    pand %xmm3, %xmm0
-; SSE2-NEXT:    por %xmm7, %xmm0
-; SSE2-NEXT:    por %xmm5, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psllw $5, %xmm4
-; SSE2-NEXT:    pand %xmm9, %xmm4
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $7, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psllw $3, %xmm7
+; SSE2-NEXT:    pand %xmm4, %xmm7
+; SSE2-NEXT:    psrlw $1, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
 ; SSE2-NEXT:    pand %xmm11, %xmm7
-; SSE2-NEXT:    por %xmm4, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    paddb %xmm4, %xmm4
-; SSE2-NEXT:    pand %xmm8, %xmm4
-; SSE2-NEXT:    por %xmm7, %xmm4
+; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm6, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
+; SSE2-NEXT:    por %xmm7, %xmm0
 ; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psrlw $1, %xmm7
-; SSE2-NEXT:    pand %xmm12, %xmm7
-; SSE2-NEXT:    por %xmm4, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psrlw $3, %xmm4
-; SSE2-NEXT:    pand %xmm6, %xmm4
-; SSE2-NEXT:    por %xmm7, %xmm4
-; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm2, %xmm6
-; SSE2-NEXT:    por %xmm4, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm3, %xmm1
-; SSE2-NEXT:    por %xmm6, %xmm1
-; SSE2-NEXT:    por %xmm5, %xmm1
+; SSE2-NEXT:    pand %xmm2, %xmm7
+; SSE2-NEXT:    psllw $4, %xmm7
+; SSE2-NEXT:    pand %xmm5, %xmm7
+; SSE2-NEXT:    pand %xmm5, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm2, %xmm1
+; SSE2-NEXT:    por %xmm7, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm3
+; SSE2-NEXT:    psllw $2, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm3
+; SSE2-NEXT:    pand %xmm9, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm10, %xmm1
+; SSE2-NEXT:    por %xmm3, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm4
+; SSE2-NEXT:    psrlw $1, %xmm4
+; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    pand %xmm6, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
+; SSE2-NEXT:    por %xmm4, %xmm1
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v32i8:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm4 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2
 ; SSSE3-NEXT:    pand %xmm4, %xmm2
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm5 = [0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240]

 ; XOPAVX2-NEXT:    retq
   %b = call <32 x i8> @llvm.bitreverse.v32i8(<32 x i8> %a)
   ret <32 x i8> %b
 }
 
 define <16 x i16> @test_bitreverse_v16i16(<16 x i16> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v16i16:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm9, %xmm9
+; SSE2-NEXT:    pxor %xmm4, %xmm4
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm9[8],xmm2[9],xmm9[9],xmm2[10],xmm9[10],xmm2[11],xmm9[11],xmm2[12],xmm9[12],xmm2[13],xmm9[13],xmm2[14],xmm9[14],xmm2[15],xmm9[15]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm4[8],xmm2[9],xmm4[9],xmm2[10],xmm4[10],xmm2[11],xmm4[11],xmm2[12],xmm4[12],xmm2[13],xmm4[13],xmm2[14],xmm4[14],xmm2[15],xmm4[15]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm9[0],xmm0[1],xmm9[1],xmm0[2],xmm9[2],xmm0[3],xmm9[3],xmm0[4],xmm9[4],xmm0[5],xmm9[5],xmm0[6],xmm9[6],xmm0[7],xmm9[7]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1],xmm0[2],xmm4[2],xmm0[3],xmm4[3],xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
 ; SSE2-NEXT:    packuswb %xmm2, %xmm0
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    psllw $5, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psllw $7, %xmm3
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm11, %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm3
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $3, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm4
-; SSE2-NEXT:    por %xmm2, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    paddb %xmm2, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
-; SSE2-NEXT:    pand %xmm8, %xmm2
-; SSE2-NEXT:    por %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psrlw $1, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm13
-; SSE2-NEXT:    pand %xmm13, %xmm4
-; SSE2-NEXT:    por %xmm2, %xmm4
+; SSE2-NEXT:    pand %xmm2, %xmm3
+; SSE2-NEXT:    psllw $4, %xmm3
+; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm6, %xmm3
+; SSE2-NEXT:    pand %xmm6, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm2, %xmm0
+; SSE2-NEXT:    por %xmm3, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm6
-; SSE2-NEXT:    pand %xmm6, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm3, %xmm5
+; SSE2-NEXT:    psllw $2, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm8, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm9, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    por %xmm5, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psrlw $5, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm2
-; SSE2-NEXT:    pand %xmm2, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm4, %xmm4
-; SSE2-NEXT:    pand %xmm4, %xmm0
+; SSE2-NEXT:    pand %xmm5, %xmm7
+; SSE2-NEXT:    psrlw $1, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; SSE2-NEXT:    pand %xmm11, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm12, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
 ; SSE2-NEXT:    por %xmm7, %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm9[8],xmm3[9],xmm9[9],xmm3[10],xmm9[10],xmm3[11],xmm9[11],xmm3[12],xmm9[12],xmm3[13],xmm9[13],xmm3[14],xmm9[14],xmm3[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm3 = xmm3[1,0,3,2,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm9[0],xmm1[1],xmm9[1],xmm1[2],xmm9[2],xmm1[3],xmm9[3],xmm1[4],xmm9[4],xmm1[5],xmm9[5],xmm1[6],xmm9[6],xmm1[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm1, %xmm7
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm7 = xmm7[8],xmm4[8],xmm7[9],xmm4[9],xmm7[10],xmm4[10],xmm7[11],xmm4[11],xmm7[12],xmm4[12],xmm7[13],xmm4[13],xmm7[14],xmm4[14],xmm7[15],xmm4[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm7 = xmm7[1,0,3,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,5,4,7,6]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3],xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    packuswb %xmm3, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psllw $7, %xmm3
-; SSE2-NEXT:    pand %xmm11, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psllw $3, %xmm7
-; SSE2-NEXT:    pand %xmm12, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psrlw $1, %xmm7
-; SSE2-NEXT:    pand %xmm13, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm6, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm2, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm4, %xmm1
-; SSE2-NEXT:    por %xmm6, %xmm1
+; SSE2-NEXT:    packuswb %xmm7, %xmm1
+; SSE2-NEXT:    movdqa %xmm1, %xmm4
+; SSE2-NEXT:    pand %xmm2, %xmm4
+; SSE2-NEXT:    psllw $4, %xmm4
+; SSE2-NEXT:    pand %xmm6, %xmm4
+; SSE2-NEXT:    pand %xmm6, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm2, %xmm1
+; SSE2-NEXT:    por %xmm4, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm3
+; SSE2-NEXT:    psllw $2, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm3
+; SSE2-NEXT:    pand %xmm9, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm10, %xmm1
 ; SSE2-NEXT:    por %xmm3, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm5
+; SSE2-NEXT:    psrlw $1, %xmm5
+; SSE2-NEXT:    pand %xmm11, %xmm5
+; SSE2-NEXT:    pand %xmm12, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
+; SSE2-NEXT:    por %xmm5, %xmm1
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v16i16:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm4 = [1,0,3,2,5,4,7,6,9,8,11,10,13,12,15,14]
 ; SSSE3-NEXT:    pshufb %xmm4, %xmm0
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm5 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2

 ; XOPAVX2-NEXT:    retq
   %b = call <16 x i16> @llvm.bitreverse.v16i16(<16 x i16> %a)
   ret <16 x i16> %b
 }
 
 define <8 x i32> @test_bitreverse_v8i32(<8 x i32> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v8i32:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm9, %xmm9
+; SSE2-NEXT:    pxor %xmm4, %xmm4
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm9[8],xmm2[9],xmm9[9],xmm2[10],xmm9[10],xmm2[11],xmm9[11],xmm2[12],xmm9[12],xmm2[13],xmm9[13],xmm2[14],xmm9[14],xmm2[15],xmm9[15]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm4[8],xmm2[9],xmm4[9],xmm2[10],xmm4[10],xmm2[11],xmm4[11],xmm2[12],xmm4[12],xmm2[13],xmm4[13],xmm2[14],xmm4[14],xmm2[15],xmm4[15]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm9[0],xmm0[1],xmm9[1],xmm0[2],xmm9[2],xmm0[3],xmm9[3],xmm0[4],xmm9[4],xmm0[5],xmm9[5],xmm0[6],xmm9[6],xmm0[7],xmm9[7]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1],xmm0[2],xmm4[2],xmm0[3],xmm4[3],xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
 ; SSE2-NEXT:    packuswb %xmm2, %xmm0
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    psllw $5, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm2
+; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psllw $7, %xmm3
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm11, %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm3
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $3, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm4
-; SSE2-NEXT:    por %xmm2, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    paddb %xmm2, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
-; SSE2-NEXT:    pand %xmm8, %xmm2
-; SSE2-NEXT:    por %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psrlw $1, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm13
-; SSE2-NEXT:    pand %xmm13, %xmm4
-; SSE2-NEXT:    por %xmm2, %xmm4
+; SSE2-NEXT:    pand %xmm2, %xmm3
+; SSE2-NEXT:    psllw $4, %xmm3
+; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm6, %xmm3
+; SSE2-NEXT:    pand %xmm6, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm2, %xmm0
+; SSE2-NEXT:    por %xmm3, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm6
-; SSE2-NEXT:    pand %xmm6, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm3, %xmm5
+; SSE2-NEXT:    psllw $2, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm8, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm9, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    por %xmm5, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psrlw $5, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm2
-; SSE2-NEXT:    pand %xmm2, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm4, %xmm4
-; SSE2-NEXT:    pand %xmm4, %xmm0
+; SSE2-NEXT:    pand %xmm5, %xmm7
+; SSE2-NEXT:    psrlw $1, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; SSE2-NEXT:    pand %xmm11, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm12, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
 ; SSE2-NEXT:    por %xmm7, %xmm0
-; SSE2-NEXT:    por %xmm3, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm9[8],xmm3[9],xmm9[9],xmm3[10],xmm9[10],xmm3[11],xmm9[11],xmm3[12],xmm9[12],xmm3[13],xmm9[13],xmm3[14],xmm9[14],xmm3[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm3 = xmm3[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm9[0],xmm1[1],xmm9[1],xmm1[2],xmm9[2],xmm1[3],xmm9[3],xmm1[4],xmm9[4],xmm1[5],xmm9[5],xmm1[6],xmm9[6],xmm1[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm1, %xmm7
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm7 = xmm7[8],xmm4[8],xmm7[9],xmm4[9],xmm7[10],xmm4[10],xmm7[11],xmm4[11],xmm7[12],xmm4[12],xmm7[13],xmm4[13],xmm7[14],xmm4[14],xmm7[15],xmm4[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm7 = xmm7[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3],xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm3, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm3
-; SSE2-NEXT:    psllw $7, %xmm3
-; SSE2-NEXT:    pand %xmm11, %xmm3
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psllw $3, %xmm7
-; SSE2-NEXT:    pand %xmm12, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psrlw $1, %xmm7
-; SSE2-NEXT:    pand %xmm13, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm6, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm2, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm4, %xmm1
-; SSE2-NEXT:    por %xmm6, %xmm1
+; SSE2-NEXT:    packuswb %xmm7, %xmm1
+; SSE2-NEXT:    movdqa %xmm1, %xmm4
+; SSE2-NEXT:    pand %xmm2, %xmm4
+; SSE2-NEXT:    psllw $4, %xmm4
+; SSE2-NEXT:    pand %xmm6, %xmm4
+; SSE2-NEXT:    pand %xmm6, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm2, %xmm1
+; SSE2-NEXT:    por %xmm4, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm3
+; SSE2-NEXT:    psllw $2, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm3
+; SSE2-NEXT:    pand %xmm9, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm10, %xmm1
 ; SSE2-NEXT:    por %xmm3, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm5
+; SSE2-NEXT:    psrlw $1, %xmm5
+; SSE2-NEXT:    pand %xmm11, %xmm5
+; SSE2-NEXT:    pand %xmm12, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
+; SSE2-NEXT:    por %xmm5, %xmm1
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v8i32:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm4 = [3,2,1,0,7,6,5,4,11,10,9,8,15,14,13,12]
 ; SSSE3-NEXT:    pshufb %xmm4, %xmm0
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm5 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2

 ; XOPAVX2-NEXT:    retq
   %b = call <8 x i32> @llvm.bitreverse.v8i32(<8 x i32> %a)
   ret <8 x i32> %b
 }
 
 define <4 x i64> @test_bitreverse_v4i64(<4 x i64> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v4i64:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm9, %xmm9
+; SSE2-NEXT:    pxor %xmm4, %xmm4
 ; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm9[8],xmm2[9],xmm9[9],xmm2[10],xmm9[10],xmm2[11],xmm9[11],xmm2[12],xmm9[12],xmm2[13],xmm9[13],xmm2[14],xmm9[14],xmm2[15],xmm9[15]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm4[8],xmm2[9],xmm4[9],xmm2[10],xmm4[10],xmm2[11],xmm4[11],xmm2[12],xmm4[12],xmm2[13],xmm4[13],xmm2[14],xmm4[14],xmm2[15],xmm4[15]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm2[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm9[0],xmm0[1],xmm9[1],xmm0[2],xmm9[2],xmm0[3],xmm9[3],xmm0[4],xmm9[4],xmm0[5],xmm9[5],xmm0[6],xmm9[6],xmm0[7],xmm9[7]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1],xmm0[2],xmm4[2],xmm0[3],xmm4[3],xmm0[4],xmm4[4],xmm0[5],xmm4[5],xmm0[6],xmm4[6],xmm0[7],xmm4[7]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
 ; SSE2-NEXT:    packuswb %xmm2, %xmm0
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    psllw $5, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm2
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm11, %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psllw $3, %xmm3
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm3
-; SSE2-NEXT:    por %xmm2, %xmm3
-; SSE2-NEXT:    movdqa %xmm0, %xmm2
-; SSE2-NEXT:    paddb %xmm2, %xmm2
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
-; SSE2-NEXT:    pand %xmm8, %xmm2
-; SSE2-NEXT:    por %xmm3, %xmm2
-; SSE2-NEXT:    movdqa %xmm0, %xmm3
-; SSE2-NEXT:    psrlw $1, %xmm3
-; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm13
-; SSE2-NEXT:    pand %xmm13, %xmm3
-; SSE2-NEXT:    por %xmm2, %xmm3
+; SSE2-NEXT:    pand %xmm2, %xmm3
+; SSE2-NEXT:    psllw $4, %xmm3
+; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm6, %xmm3
+; SSE2-NEXT:    pand %xmm6, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm2, %xmm0
+; SSE2-NEXT:    por %xmm3, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm6
-; SSE2-NEXT:    pand %xmm6, %xmm5
-; SSE2-NEXT:    por %xmm3, %xmm5
+; SSE2-NEXT:    pand %xmm3, %xmm5
+; SSE2-NEXT:    psllw $2, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm8, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm9, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    por %xmm5, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psrlw $5, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm2 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm2
-; SSE2-NEXT:    pand %xmm2, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm3, %xmm3
-; SSE2-NEXT:    pand %xmm3, %xmm0
+; SSE2-NEXT:    pand %xmm5, %xmm7
+; SSE2-NEXT:    psrlw $1, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; SSE2-NEXT:    pand %xmm11, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm12, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
 ; SSE2-NEXT:    por %xmm7, %xmm0
-; SSE2-NEXT:    por %xmm4, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm4[2,3,0,1]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm9[0],xmm1[1],xmm9[1],xmm1[2],xmm9[2],xmm1[3],xmm9[3],xmm1[4],xmm9[4],xmm1[5],xmm9[5],xmm1[6],xmm9[6],xmm1[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm1, %xmm7
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm7 = xmm7[8],xmm4[8],xmm7[9],xmm4[9],xmm7[10],xmm4[10],xmm7[11],xmm4[11],xmm7[12],xmm4[12],xmm7[13],xmm4[13],xmm7[14],xmm4[14],xmm7[15],xmm4[15]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm7 = xmm7[2,3,0,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm7 = xmm7[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm7 = xmm7[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3],xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
+; SSE2-NEXT:    packuswb %xmm7, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psllw $3, %xmm7
-; SSE2-NEXT:    pand %xmm12, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psrlw $1, %xmm7
-; SSE2-NEXT:    pand %xmm13, %xmm7
-; SSE2-NEXT:    por %xmm5, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm6, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm2, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm3, %xmm1
-; SSE2-NEXT:    por %xmm6, %xmm1
+; SSE2-NEXT:    pand %xmm2, %xmm4
+; SSE2-NEXT:    psllw $4, %xmm4
+; SSE2-NEXT:    pand %xmm6, %xmm4
+; SSE2-NEXT:    pand %xmm6, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm2, %xmm1
 ; SSE2-NEXT:    por %xmm4, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm3
+; SSE2-NEXT:    psllw $2, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm3
+; SSE2-NEXT:    pand %xmm9, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm10, %xmm1
+; SSE2-NEXT:    por %xmm3, %xmm1
+; SSE2-NEXT:    pand %xmm1, %xmm5
+; SSE2-NEXT:    psrlw $1, %xmm5
+; SSE2-NEXT:    pand %xmm11, %xmm5
+; SSE2-NEXT:    pand %xmm12, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
+; SSE2-NEXT:    por %xmm5, %xmm1
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v4i64:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm4 = [7,6,5,4,3,2,1,0,15,14,13,12,11,10,9,8]
 ; SSSE3-NEXT:    pshufb %xmm4, %xmm0
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm5 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm2

 ; XOPAVX2-NEXT:    retq
   %b = call <4 x i64> @llvm.bitreverse.v4i64(<4 x i64> %a)
   ret <4 x i64> %b
 }
 
 define <64 x i8> @test_bitreverse_v64i8(<64 x i8> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v64i8:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $5, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm9
-; SSE2-NEXT:    pand %xmm9, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psllw $7, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm10, %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psllw $3, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    paddb %xmm4, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
-; SSE2-NEXT:    pand %xmm8, %xmm4
-; SSE2-NEXT:    por %xmm5, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psrlw $1, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm13, %xmm5
+; SSE2-NEXT:    psllw $4, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm7 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm7, %xmm5
+; SSE2-NEXT:    pand %xmm7, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm13, %xmm0
+; SSE2-NEXT:    por %xmm5, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm6
-; SSE2-NEXT:    psrlw $3, %xmm6
-; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm13
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
+; SSE2-NEXT:    pand %xmm5, %xmm6
+; SSE2-NEXT:    psllw $2, %xmm6
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm9, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    por %xmm6, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psrlw $5, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm14 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm14
-; SSE2-NEXT:    pand %xmm14, %xmm4
-; SSE2-NEXT:    por %xmm6, %xmm4
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm6 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm6, %xmm6
-; SSE2-NEXT:    pand %xmm6, %xmm0
+; SSE2-NEXT:    pand %xmm6, %xmm4
+; SSE2-NEXT:    psrlw $1, %xmm4
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm12, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
 ; SSE2-NEXT:    por %xmm4, %xmm0
-; SSE2-NEXT:    por %xmm7, %xmm0
 ; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psllw $5, %xmm4
-; SSE2-NEXT:    pand %xmm9, %xmm4
-; SSE2-NEXT:    movdqa %xmm1, %xmm7
-; SSE2-NEXT:    psllw $7, %xmm7
-; SSE2-NEXT:    pand %xmm10, %xmm7
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $3, %xmm5
-; SSE2-NEXT:    pand %xmm11, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm13, %xmm4
+; SSE2-NEXT:    psllw $4, %xmm4
+; SSE2-NEXT:    pand %xmm7, %xmm4
+; SSE2-NEXT:    pand %xmm7, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm13, %xmm1
+; SSE2-NEXT:    por %xmm4, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    paddb %xmm4, %xmm4
+; SSE2-NEXT:    pand %xmm5, %xmm4
+; SSE2-NEXT:    psllw $2, %xmm4
 ; SSE2-NEXT:    pand %xmm8, %xmm4
-; SSE2-NEXT:    por %xmm5, %xmm4
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $1, %xmm5
-; SSE2-NEXT:    pand %xmm12, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm9, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm10, %xmm1
+; SSE2-NEXT:    por %xmm4, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psrlw $3, %xmm4
-; SSE2-NEXT:    pand %xmm13, %xmm4
-; SSE2-NEXT:    por %xmm5, %xmm4
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $5, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm6, %xmm1
-; SSE2-NEXT:    por %xmm5, %xmm1
-; SSE2-NEXT:    por %xmm7, %xmm1
+; SSE2-NEXT:    pand %xmm6, %xmm4
+; SSE2-NEXT:    psrlw $1, %xmm4
+; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    pand %xmm12, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
+; SSE2-NEXT:    por %xmm4, %xmm1
 ; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    psllw $5, %xmm4
-; SSE2-NEXT:    pand %xmm9, %xmm4
-; SSE2-NEXT:    movdqa %xmm2, %xmm7
-; SSE2-NEXT:    psllw $7, %xmm7
-; SSE2-NEXT:    pand %xmm10, %xmm7
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psllw $3, %xmm5
-; SSE2-NEXT:    pand %xmm11, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm13, %xmm4
+; SSE2-NEXT:    psllw $4, %xmm4
+; SSE2-NEXT:    pand %xmm7, %xmm4
+; SSE2-NEXT:    pand %xmm7, %xmm2
+; SSE2-NEXT:    psrlw $4, %xmm2
+; SSE2-NEXT:    pand %xmm13, %xmm2
+; SSE2-NEXT:    por %xmm4, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    paddb %xmm4, %xmm4
+; SSE2-NEXT:    pand %xmm5, %xmm4
+; SSE2-NEXT:    psllw $2, %xmm4
 ; SSE2-NEXT:    pand %xmm8, %xmm4
-; SSE2-NEXT:    por %xmm5, %xmm4
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psrlw $1, %xmm5
-; SSE2-NEXT:    pand %xmm12, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm9, %xmm2
+; SSE2-NEXT:    psrlw $2, %xmm2
+; SSE2-NEXT:    pand %xmm10, %xmm2
+; SSE2-NEXT:    por %xmm4, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    psrlw $3, %xmm4
-; SSE2-NEXT:    pand %xmm13, %xmm4
-; SSE2-NEXT:    por %xmm5, %xmm4
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psrlw $5, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
-; SSE2-NEXT:    psrlw $7, %xmm2
-; SSE2-NEXT:    pand %xmm6, %xmm2
-; SSE2-NEXT:    por %xmm5, %xmm2
-; SSE2-NEXT:    por %xmm7, %xmm2
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    psllw $5, %xmm4
-; SSE2-NEXT:    pand %xmm9, %xmm4
-; SSE2-NEXT:    movdqa %xmm3, %xmm7
-; SSE2-NEXT:    psllw $7, %xmm7
-; SSE2-NEXT:    pand %xmm10, %xmm7
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psllw $3, %xmm5
-; SSE2-NEXT:    pand %xmm11, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    paddb %xmm4, %xmm4
-; SSE2-NEXT:    pand %xmm8, %xmm4
-; SSE2-NEXT:    por %xmm5, %xmm4
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psrlw $1, %xmm5
-; SSE2-NEXT:    pand %xmm12, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm6, %xmm4
+; SSE2-NEXT:    psrlw $1, %xmm4
+; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    pand %xmm12, %xmm2
+; SSE2-NEXT:    paddb %xmm2, %xmm2
+; SSE2-NEXT:    por %xmm4, %xmm2
 ; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    psrlw $3, %xmm4
 ; SSE2-NEXT:    pand %xmm13, %xmm4
-; SSE2-NEXT:    por %xmm5, %xmm4
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psrlw $5, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm4, %xmm5
-; SSE2-NEXT:    psrlw $7, %xmm3
-; SSE2-NEXT:    pand %xmm6, %xmm3
+; SSE2-NEXT:    psllw $4, %xmm4
+; SSE2-NEXT:    pand %xmm7, %xmm4
+; SSE2-NEXT:    pand %xmm7, %xmm3
+; SSE2-NEXT:    psrlw $4, %xmm3
+; SSE2-NEXT:    pand %xmm13, %xmm3
+; SSE2-NEXT:    por %xmm4, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm5
+; SSE2-NEXT:    psllw $2, %xmm5
+; SSE2-NEXT:    pand %xmm8, %xmm5
+; SSE2-NEXT:    pand %xmm9, %xmm3
+; SSE2-NEXT:    psrlw $2, %xmm3
+; SSE2-NEXT:    pand %xmm10, %xmm3
 ; SSE2-NEXT:    por %xmm5, %xmm3
-; SSE2-NEXT:    por %xmm7, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm6
+; SSE2-NEXT:    psrlw $1, %xmm6
+; SSE2-NEXT:    pand %xmm11, %xmm6
+; SSE2-NEXT:    pand %xmm12, %xmm3
+; SSE2-NEXT:    paddb %xmm3, %xmm3
+; SSE2-NEXT:    por %xmm6, %xmm3
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v64i8:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm5
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm8 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSSE3-NEXT:    pand %xmm8, %xmm0
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm9 = [0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240]

 ; XOPAVX2-NEXT:    retq
   %b = call <64 x i8> @llvm.bitreverse.v64i8(<64 x i8> %a)
   ret <64 x i8> %b
 }
 
 define <32 x i16> @test_bitreverse_v32i16(<32 x i16> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v32i16:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm9, %xmm9
+; SSE2-NEXT:    pxor %xmm14, %xmm14
 ; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm14[8],xmm4[9],xmm14[9],xmm4[10],xmm14[10],xmm4[11],xmm14[11],xmm4[12],xmm14[12],xmm4[13],xmm14[13],xmm4[14],xmm14[14],xmm4[15],xmm14[15]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm9[0],xmm0[1],xmm9[1],xmm0[2],xmm9[2],xmm0[3],xmm9[3],xmm0[4],xmm9[4],xmm0[5],xmm9[5],xmm0[6],xmm9[6],xmm0[7],xmm9[7]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm14[0],xmm0[1],xmm14[1],xmm0[2],xmm14[2],xmm0[3],xmm14[3],xmm0[4],xmm14[4],xmm0[5],xmm14[5],xmm0[6],xmm14[6],xmm0[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,5,4,7,6]
 ; SSE2-NEXT:    packuswb %xmm4, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm11, %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    psllw $4, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm4, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm8, %xmm0
+; SSE2-NEXT:    por %xmm5, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
+; SSE2-NEXT:    movdqa %xmm0, %xmm7
+; SSE2-NEXT:    pand %xmm5, %xmm7
+; SSE2-NEXT:    psllw $2, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm9, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm11, %xmm0
+; SSE2-NEXT:    por %xmm7, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm7 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm6
+; SSE2-NEXT:    pand %xmm7, %xmm6
 ; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm13
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psrlw $3, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm14 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm14
-; SSE2-NEXT:    pand %xmm14, %xmm7
-; SSE2-NEXT:    por %xmm6, %xmm7
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psrlw $5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm15 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm15
-; SSE2-NEXT:    pand %xmm15, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm7 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm7, %xmm7
-; SSE2-NEXT:    pand %xmm7, %xmm0
-; SSE2-NEXT:    por %xmm5, %xmm0
-; SSE2-NEXT:    por %xmm4, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[1,0,3,2,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm9[0],xmm1[1],xmm9[1],xmm1[2],xmm9[2],xmm1[3],xmm9[3],xmm1[4],xmm9[4],xmm1[5],xmm9[5],xmm1[6],xmm9[6],xmm1[7],xmm9[7]
+; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm13, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
+; SSE2-NEXT:    por %xmm6, %xmm0
+; SSE2-NEXT:    movdqa %xmm1, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[1,0,3,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,5,4,7,6]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm14[0],xmm1[1],xmm14[1],xmm1[2],xmm14[2],xmm1[3],xmm14[3],xmm1[4],xmm14[4],xmm1[5],xmm14[5],xmm1[6],xmm14[6],xmm1[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    packuswb %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm8, %xmm1
+; SSE2-NEXT:    por %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm5, %xmm6
+; SSE2-NEXT:    psllw $2, %xmm6
+; SSE2-NEXT:    pand %xmm9, %xmm6
+; SSE2-NEXT:    pand %xmm10, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm11, %xmm1
+; SSE2-NEXT:    por %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm7, %xmm1
+; SSE2-NEXT:    pand %xmm7, %xmm6
+; SSE2-NEXT:    psrlw $1, %xmm6
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    pand %xmm13, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
 ; SSE2-NEXT:    por %xmm6, %xmm1
-; SSE2-NEXT:    por %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[1,0,3,2,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm9[0],xmm2[1],xmm9[1],xmm2[2],xmm9[2],xmm2[3],xmm9[3],xmm2[4],xmm9[4],xmm2[5],xmm9[5],xmm2[6],xmm9[6],xmm2[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm2, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[1,0,3,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,5,4,7,6]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm14[0],xmm2[1],xmm14[1],xmm2[2],xmm14[2],xmm2[3],xmm14[3],xmm2[4],xmm14[4],xmm2[5],xmm14[5],xmm2[6],xmm14[6],xmm2[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    packuswb %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm2
+; SSE2-NEXT:    psrlw $4, %xmm2
+; SSE2-NEXT:    pand %xmm8, %xmm2
+; SSE2-NEXT:    por %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm5, %xmm6
+; SSE2-NEXT:    psllw $2, %xmm6
+; SSE2-NEXT:    pand %xmm9, %xmm6
+; SSE2-NEXT:    pand %xmm10, %xmm2
+; SSE2-NEXT:    psrlw $2, %xmm2
+; SSE2-NEXT:    pand %xmm11, %xmm2
+; SSE2-NEXT:    por %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm2
-; SSE2-NEXT:    pand %xmm7, %xmm2
+; SSE2-NEXT:    pand %xmm7, %xmm6
+; SSE2-NEXT:    psrlw $1, %xmm6
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    pand %xmm13, %xmm2
+; SSE2-NEXT:    paddb %xmm2, %xmm2
 ; SSE2-NEXT:    por %xmm6, %xmm2
-; SSE2-NEXT:    por %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[1,0,3,2,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm9[0],xmm3[1],xmm9[1],xmm3[2],xmm9[2],xmm3[3],xmm9[3],xmm3[4],xmm9[4],xmm3[5],xmm9[5],xmm3[6],xmm9[6],xmm3[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm3, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[1,0,3,2,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,5,4,7,6]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm14[0],xmm3[1],xmm14[1],xmm3[2],xmm14[2],xmm3[3],xmm14[3],xmm3[4],xmm14[4],xmm3[5],xmm14[5],xmm3[6],xmm14[6],xmm3[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm3 = xmm3[1,0,3,2,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,5,4,7,6]
-; SSE2-NEXT:    packuswb %xmm4, %xmm3
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm3
 ; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm3
-; SSE2-NEXT:    pand %xmm7, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm3
+; SSE2-NEXT:    psrlw $4, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm3
 ; SSE2-NEXT:    por %xmm6, %xmm3
-; SSE2-NEXT:    por %xmm4, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm5
+; SSE2-NEXT:    psllw $2, %xmm5
+; SSE2-NEXT:    pand %xmm9, %xmm5
+; SSE2-NEXT:    pand %xmm10, %xmm3
+; SSE2-NEXT:    psrlw $2, %xmm3
+; SSE2-NEXT:    pand %xmm11, %xmm3
+; SSE2-NEXT:    por %xmm5, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm7
+; SSE2-NEXT:    psrlw $1, %xmm7
+; SSE2-NEXT:    pand %xmm12, %xmm7
+; SSE2-NEXT:    pand %xmm13, %xmm3
+; SSE2-NEXT:    paddb %xmm3, %xmm3
+; SSE2-NEXT:    por %xmm7, %xmm3
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v32i16:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa %xmm1, %xmm5
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm1
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm8 = [1,0,3,2,5,4,7,6,9,8,11,10,13,12,15,14]
 ; SSSE3-NEXT:    pshufb %xmm8, %xmm1

 ; XOPAVX2-NEXT:    retq
   %b = call <32 x i16> @llvm.bitreverse.v32i16(<32 x i16> %a)
   ret <32 x i16> %b
 }
 
 define <16 x i32> @test_bitreverse_v16i32(<16 x i32> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v16i32:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm9, %xmm9
+; SSE2-NEXT:    pxor %xmm14, %xmm14
 ; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm14[8],xmm4[9],xmm14[9],xmm4[10],xmm14[10],xmm4[11],xmm14[11],xmm4[12],xmm14[12],xmm4[13],xmm14[13],xmm4[14],xmm14[14],xmm4[15],xmm14[15]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm9[0],xmm0[1],xmm9[1],xmm0[2],xmm9[2],xmm0[3],xmm9[3],xmm0[4],xmm9[4],xmm0[5],xmm9[5],xmm0[6],xmm9[6],xmm0[7],xmm9[7]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm14[0],xmm0[1],xmm14[1],xmm0[2],xmm14[2],xmm0[3],xmm14[3],xmm0[4],xmm14[4],xmm0[5],xmm14[5],xmm0[6],xmm14[6],xmm0[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
 ; SSE2-NEXT:    packuswb %xmm4, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm11, %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    psllw $4, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm4, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm8, %xmm0
+; SSE2-NEXT:    por %xmm5, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
+; SSE2-NEXT:    movdqa %xmm0, %xmm7
+; SSE2-NEXT:    pand %xmm5, %xmm7
+; SSE2-NEXT:    psllw $2, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm9, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm11, %xmm0
+; SSE2-NEXT:    por %xmm7, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm7 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm6
+; SSE2-NEXT:    pand %xmm7, %xmm6
 ; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm13
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psrlw $3, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm14 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm14
-; SSE2-NEXT:    pand %xmm14, %xmm7
-; SSE2-NEXT:    por %xmm6, %xmm7
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psrlw $5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm15 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm15
-; SSE2-NEXT:    pand %xmm15, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm7 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm7, %xmm7
-; SSE2-NEXT:    pand %xmm7, %xmm0
-; SSE2-NEXT:    por %xmm5, %xmm0
-; SSE2-NEXT:    por %xmm4, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm9[0],xmm1[1],xmm9[1],xmm1[2],xmm9[2],xmm1[3],xmm9[3],xmm1[4],xmm9[4],xmm1[5],xmm9[5],xmm1[6],xmm9[6],xmm1[7],xmm9[7]
+; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm13, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
+; SSE2-NEXT:    por %xmm6, %xmm0
+; SSE2-NEXT:    movdqa %xmm1, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm14[0],xmm1[1],xmm14[1],xmm1[2],xmm14[2],xmm1[3],xmm14[3],xmm1[4],xmm14[4],xmm1[5],xmm14[5],xmm1[6],xmm14[6],xmm1[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm8, %xmm1
+; SSE2-NEXT:    por %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm5, %xmm6
+; SSE2-NEXT:    psllw $2, %xmm6
+; SSE2-NEXT:    pand %xmm9, %xmm6
+; SSE2-NEXT:    pand %xmm10, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm11, %xmm1
+; SSE2-NEXT:    por %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm7, %xmm1
+; SSE2-NEXT:    pand %xmm7, %xmm6
+; SSE2-NEXT:    psrlw $1, %xmm6
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    pand %xmm13, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
 ; SSE2-NEXT:    por %xmm6, %xmm1
-; SSE2-NEXT:    por %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm9[0],xmm2[1],xmm9[1],xmm2[2],xmm9[2],xmm2[3],xmm9[3],xmm2[4],xmm9[4],xmm2[5],xmm9[5],xmm2[6],xmm9[6],xmm2[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm2, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm14[0],xmm2[1],xmm14[1],xmm2[2],xmm14[2],xmm2[3],xmm14[3],xmm2[4],xmm14[4],xmm2[5],xmm14[5],xmm2[6],xmm14[6],xmm2[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm2
+; SSE2-NEXT:    psrlw $4, %xmm2
+; SSE2-NEXT:    pand %xmm8, %xmm2
+; SSE2-NEXT:    por %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm5, %xmm6
+; SSE2-NEXT:    psllw $2, %xmm6
+; SSE2-NEXT:    pand %xmm9, %xmm6
+; SSE2-NEXT:    pand %xmm10, %xmm2
+; SSE2-NEXT:    psrlw $2, %xmm2
+; SSE2-NEXT:    pand %xmm11, %xmm2
+; SSE2-NEXT:    por %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm2
-; SSE2-NEXT:    pand %xmm7, %xmm2
+; SSE2-NEXT:    pand %xmm7, %xmm6
+; SSE2-NEXT:    psrlw $1, %xmm6
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    pand %xmm13, %xmm2
+; SSE2-NEXT:    paddb %xmm2, %xmm2
 ; SSE2-NEXT:    por %xmm6, %xmm2
-; SSE2-NEXT:    por %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm9[0],xmm3[1],xmm9[1],xmm3[2],xmm9[2],xmm3[3],xmm9[3],xmm3[4],xmm9[4],xmm3[5],xmm9[5],xmm3[6],xmm9[6],xmm3[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm3, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm14[0],xmm3[1],xmm14[1],xmm3[2],xmm14[2],xmm3[3],xmm14[3],xmm3[4],xmm14[4],xmm3[5],xmm14[5],xmm3[6],xmm14[6],xmm3[7],xmm14[7]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm3 = xmm3[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm4, %xmm3
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
-; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    packuswb %xmm6, %xmm3
 ; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm3
-; SSE2-NEXT:    pand %xmm7, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm3
+; SSE2-NEXT:    psrlw $4, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm3
 ; SSE2-NEXT:    por %xmm6, %xmm3
-; SSE2-NEXT:    por %xmm4, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm5
+; SSE2-NEXT:    psllw $2, %xmm5
+; SSE2-NEXT:    pand %xmm9, %xmm5
+; SSE2-NEXT:    pand %xmm10, %xmm3
+; SSE2-NEXT:    psrlw $2, %xmm3
+; SSE2-NEXT:    pand %xmm11, %xmm3
+; SSE2-NEXT:    por %xmm5, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm7
+; SSE2-NEXT:    psrlw $1, %xmm7
+; SSE2-NEXT:    pand %xmm12, %xmm7
+; SSE2-NEXT:    pand %xmm13, %xmm3
+; SSE2-NEXT:    paddb %xmm3, %xmm3
+; SSE2-NEXT:    por %xmm7, %xmm3
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v16i32:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa %xmm1, %xmm5
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm1
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm8 = [3,2,1,0,7,6,5,4,11,10,9,8,15,14,13,12]
 ; SSSE3-NEXT:    pshufb %xmm8, %xmm1

 ; AVX2-NEXT:    vpsrlw $4, %ymm1, %ymm1
 ; AVX2-NEXT:    vpand %ymm3, %ymm1, %ymm1
 ; AVX2-NEXT:    vpshufb %ymm1, %ymm6, %ymm1
 ; AVX2-NEXT:    vpor %ymm1, %ymm2, %ymm1
 ; AVX2-NEXT:    retq
 ;
 ; AVX512F-LABEL: test_bitreverse_v16i32:
 ; AVX512F:       # BB#0:
-; AVX512F-NEXT:    vpslld $29, %zmm0, %zmm1
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $31, %zmm0, %zmm2
+; AVX512F-NEXT:    vpsrld $24, %zmm0, %zmm1
+; AVX512F-NEXT:    vpsrld $8, %zmm0, %zmm2
 ; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
 ; AVX512F-NEXT:    vpord %zmm1, %zmm2, %zmm1
-; AVX512F-NEXT:    vpslld $27, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $25, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $23, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $21, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $19, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $17, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $15, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $13, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $11, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $9, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $7, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $5, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $3, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpslld $1, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $1, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $3, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $5, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $7, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $9, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $11, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $13, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $15, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $17, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $19, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $21, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $23, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $25, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $27, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $29, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm2, %zmm2
-; AVX512F-NEXT:    vpord %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrld $31, %zmm0, %zmm0
+; AVX512F-NEXT:    vpslld $24, %zmm0, %zmm2
+; AVX512F-NEXT:    vpslld $8, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm0, %zmm0
+; AVX512F-NEXT:    vpord %zmm0, %zmm2, %zmm0
+; AVX512F-NEXT:    vpord %zmm1, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm0, %zmm1
+; AVX512F-NEXT:    vpslld $4, %zmm1, %zmm1
+; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm0, %zmm0
+; AVX512F-NEXT:    vpsrld $4, %zmm0, %zmm0
+; AVX512F-NEXT:    vpord %zmm1, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm0, %zmm1
+; AVX512F-NEXT:    vpslld $2, %zmm1, %zmm1
+; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm0, %zmm0
+; AVX512F-NEXT:    vpsrld $2, %zmm0, %zmm0
+; AVX512F-NEXT:    vpord %zmm1, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm0, %zmm1
+; AVX512F-NEXT:    vpslld $1, %zmm1, %zmm1
 ; AVX512F-NEXT:    vpandd {{.*}}(%rip){1to16}, %zmm0, %zmm0
-; AVX512F-NEXT:    vpord %zmm0, %zmm1, %zmm0
+; AVX512F-NEXT:    vpsrld $1, %zmm0, %zmm0
+; AVX512F-NEXT:    vpord %zmm1, %zmm0, %zmm0
 ; AVX512F-NEXT:    retq
 ;
 ; AVX512BW-LABEL: test_bitreverse_v16i32:
 ; AVX512BW:       # BB#0:
 ; AVX512BW-NEXT:    vpshufb {{.*#+}} zmm0 = zmm0[3,2,1,0,7,6,5,4,11,10,9,8,15,14,13,12,19,18,17,16,23,22,21,20,27,26,25,24,31,30,29,28,35,34,33,32,39,38,37,36,43,42,41,40,47,46,45,44,51,50,49,48,55,54,53,52,59,58,57,56,63,62,61,60]
 ; AVX512BW-NEXT:    vmovdqa64 {{.*#+}} zmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; AVX512BW-NEXT:    vpandq %zmm1, %zmm0, %zmm2
 ; AVX512BW-NEXT:    vmovdqu8 {{.*#+}} zmm3 = [0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240,0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240,0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240,0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240]

 ; XOPAVX2-NEXT:    retq
   %b = call <16 x i32> @llvm.bitreverse.v16i32(<16 x i32> %a)
   ret <16 x i32> %b
 }
 
 define <8 x i64> @test_bitreverse_v8i64(<8 x i64> %a) nounwind {
 ; SSE2-LABEL: test_bitreverse_v8i64:
 ; SSE2:       # BB#0:
-; SSE2-NEXT:    pxor %xmm9, %xmm9
+; SSE2-NEXT:    pxor %xmm14, %xmm14
 ; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm14[8],xmm4[9],xmm14[9],xmm4[10],xmm14[10],xmm4[11],xmm14[11],xmm4[12],xmm14[12],xmm4[13],xmm14[13],xmm4[14],xmm14[14],xmm4[15],xmm14[15]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm4[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm9[0],xmm0[1],xmm9[1],xmm0[2],xmm9[2],xmm0[3],xmm9[3],xmm0[4],xmm9[4],xmm0[5],xmm9[5],xmm0[6],xmm9[6],xmm0[7],xmm9[7]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm14[0],xmm0[1],xmm14[1],xmm0[2],xmm14[2],xmm0[3],xmm14[3],xmm0[4],xmm14[4],xmm0[5],xmm14[5],xmm0[6],xmm14[6],xmm0[7],xmm14[7]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
 ; SSE2-NEXT:    packuswb %xmm4, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [64,64,64,64,64,64,64,64,64,64,64,64,64,64,64,64]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm10
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm0, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [128,128,128,128,128,128,128,128,128,128,128,128,128,128,128,128]
-; SSE2-NEXT:    pand %xmm11, %xmm11
-; SSE2-NEXT:    pand %xmm11, %xmm4
-; SSE2-NEXT:    movdqa %xmm0, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [32,32,32,32,32,32,32,32,32,32,32,32,32,32,32,32]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm12
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [16,16,16,16,16,16,16,16,16,16,16,16,16,16,16,16]
 ; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    psllw $4, %xmm5
+; SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
+; SSE2-NEXT:    pand %xmm4, %xmm5
+; SSE2-NEXT:    pand %xmm4, %xmm0
+; SSE2-NEXT:    psrlw $4, %xmm0
+; SSE2-NEXT:    pand %xmm8, %xmm0
+; SSE2-NEXT:    por %xmm5, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
+; SSE2-NEXT:    movdqa %xmm0, %xmm7
+; SSE2-NEXT:    pand %xmm5, %xmm7
+; SSE2-NEXT:    psllw $2, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm9 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
+; SSE2-NEXT:    pand %xmm9, %xmm7
+; SSE2-NEXT:    movdqa {{.*#+}} xmm10 = [204,204,204,204,204,204,204,204,204,204,204,204,204,204,204,204]
+; SSE2-NEXT:    pand %xmm10, %xmm0
+; SSE2-NEXT:    psrlw $2, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm11 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
+; SSE2-NEXT:    pand %xmm11, %xmm0
+; SSE2-NEXT:    por %xmm7, %xmm0
+; SSE2-NEXT:    movdqa {{.*#+}} xmm7 = [170,170,170,170,170,170,170,170,170,170,170,170,170,170,170,170]
 ; SSE2-NEXT:    movdqa %xmm0, %xmm6
+; SSE2-NEXT:    pand %xmm7, %xmm6
 ; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [8,8,8,8,8,8,8,8,8,8,8,8,8,8,8,8]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm13
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm0, %xmm7
-; SSE2-NEXT:    psrlw $3, %xmm7
-; SSE2-NEXT:    movdqa {{.*#+}} xmm14 = [4,4,4,4,4,4,4,4,4,4,4,4,4,4,4,4]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm14
-; SSE2-NEXT:    pand %xmm14, %xmm7
-; SSE2-NEXT:    por %xmm6, %xmm7
-; SSE2-NEXT:    movdqa %xmm0, %xmm5
-; SSE2-NEXT:    psrlw $5, %xmm5
-; SSE2-NEXT:    movdqa {{.*#+}} xmm15 = [2,2,2,2,2,2,2,2,2,2,2,2,2,2,2,2]
-; SSE2-NEXT:    pand {{.*}}(%rip), %xmm15
-; SSE2-NEXT:    pand %xmm15, %xmm5
-; SSE2-NEXT:    por %xmm7, %xmm5
-; SSE2-NEXT:    psrlw $7, %xmm0
-; SSE2-NEXT:    movdqa {{.*#+}} xmm7 = [1,1,1,1,1,1,1,1,1,1,1,1,1,1,1,1]
-; SSE2-NEXT:    pand %xmm7, %xmm7
-; SSE2-NEXT:    pand %xmm7, %xmm0
-; SSE2-NEXT:    por %xmm5, %xmm0
-; SSE2-NEXT:    por %xmm4, %xmm0
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm4[2,3,0,1]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm9[0],xmm1[1],xmm9[1],xmm1[2],xmm9[2],xmm1[3],xmm9[3],xmm1[4],xmm9[4],xmm1[5],xmm9[5],xmm1[6],xmm9[6],xmm1[7],xmm9[7]
+; SSE2-NEXT:    movdqa {{.*#+}} xmm12 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    movdqa {{.*#+}} xmm13 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
+; SSE2-NEXT:    pand %xmm13, %xmm0
+; SSE2-NEXT:    paddb %xmm0, %xmm0
+; SSE2-NEXT:    por %xmm6, %xmm0
+; SSE2-NEXT:    movdqa %xmm1, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm6[2,3,0,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm14[0],xmm1[1],xmm14[1],xmm1[2],xmm14[2],xmm1[3],xmm14[3],xmm1[4],xmm14[4],xmm1[5],xmm14[5],xmm1[6],xmm14[6],xmm1[7],xmm14[7]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm1[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm1 = xmm1[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm1 = xmm1[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm1
+; SSE2-NEXT:    psrlw $4, %xmm1
+; SSE2-NEXT:    pand %xmm8, %xmm1
+; SSE2-NEXT:    por %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm1, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm5, %xmm6
+; SSE2-NEXT:    psllw $2, %xmm6
+; SSE2-NEXT:    pand %xmm9, %xmm6
+; SSE2-NEXT:    pand %xmm10, %xmm1
+; SSE2-NEXT:    psrlw $2, %xmm1
+; SSE2-NEXT:    pand %xmm11, %xmm1
+; SSE2-NEXT:    por %xmm6, %xmm1
 ; SSE2-NEXT:    movdqa %xmm1, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm1
-; SSE2-NEXT:    pand %xmm7, %xmm1
+; SSE2-NEXT:    pand %xmm7, %xmm6
+; SSE2-NEXT:    psrlw $1, %xmm6
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    pand %xmm13, %xmm1
+; SSE2-NEXT:    paddb %xmm1, %xmm1
 ; SSE2-NEXT:    por %xmm6, %xmm1
-; SSE2-NEXT:    por %xmm4, %xmm1
-; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm4[2,3,0,1]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm9[0],xmm2[1],xmm9[1],xmm2[2],xmm9[2],xmm2[3],xmm9[3],xmm2[4],xmm9[4],xmm2[5],xmm9[5],xmm2[6],xmm9[6],xmm2[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm2, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm6[2,3,0,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm14[0],xmm2[1],xmm14[1],xmm2[2],xmm14[2],xmm2[3],xmm14[3],xmm2[4],xmm14[4],xmm2[5],xmm14[5],xmm2[6],xmm14[6],xmm2[7],xmm14[7]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm2 = xmm2[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm2, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm2
+; SSE2-NEXT:    psrlw $4, %xmm2
+; SSE2-NEXT:    pand %xmm8, %xmm2
+; SSE2-NEXT:    por %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm2, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
+; SSE2-NEXT:    pand %xmm5, %xmm6
+; SSE2-NEXT:    psllw $2, %xmm6
+; SSE2-NEXT:    pand %xmm9, %xmm6
+; SSE2-NEXT:    pand %xmm10, %xmm2
+; SSE2-NEXT:    psrlw $2, %xmm2
+; SSE2-NEXT:    pand %xmm11, %xmm2
+; SSE2-NEXT:    por %xmm6, %xmm2
 ; SSE2-NEXT:    movdqa %xmm2, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm2
-; SSE2-NEXT:    pand %xmm7, %xmm2
+; SSE2-NEXT:    pand %xmm7, %xmm6
+; SSE2-NEXT:    psrlw $1, %xmm6
+; SSE2-NEXT:    pand %xmm12, %xmm6
+; SSE2-NEXT:    pand %xmm13, %xmm2
+; SSE2-NEXT:    paddb %xmm2, %xmm2
 ; SSE2-NEXT:    por %xmm6, %xmm2
-; SSE2-NEXT:    por %xmm4, %xmm2
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm9[8],xmm4[9],xmm9[9],xmm4[10],xmm9[10],xmm4[11],xmm9[11],xmm4[12],xmm9[12],xmm4[13],xmm9[13],xmm4[14],xmm9[14],xmm4[15],xmm9[15]
-; SSE2-NEXT:    pshufd {{.*#+}} xmm4 = xmm4[2,3,0,1]
-; SSE2-NEXT:    pshuflw {{.*#+}} xmm4 = xmm4[3,2,1,0,4,5,6,7]
-; SSE2-NEXT:    pshufhw {{.*#+}} xmm4 = xmm4[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm9[0],xmm3[1],xmm9[1],xmm3[2],xmm9[2],xmm3[3],xmm9[3],xmm3[4],xmm9[4],xmm3[5],xmm9[5],xmm3[6],xmm9[6],xmm3[7],xmm9[7]
+; SSE2-NEXT:    movdqa %xmm3, %xmm6
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm6 = xmm6[8],xmm14[8],xmm6[9],xmm14[9],xmm6[10],xmm14[10],xmm6[11],xmm14[11],xmm6[12],xmm14[12],xmm6[13],xmm14[13],xmm6[14],xmm14[14],xmm6[15],xmm14[15]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm6 = xmm6[2,3,0,1]
+; SSE2-NEXT:    pshuflw {{.*#+}} xmm6 = xmm6[3,2,1,0,4,5,6,7]
+; SSE2-NEXT:    pshufhw {{.*#+}} xmm6 = xmm6[0,1,2,3,7,6,5,4]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm14[0],xmm3[1],xmm14[1],xmm3[2],xmm14[2],xmm3[3],xmm14[3],xmm3[4],xmm14[4],xmm3[5],xmm14[5],xmm3[6],xmm14[6],xmm3[7],xmm14[7]
 ; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm3[2,3,0,1]
 ; SSE2-NEXT:    pshuflw {{.*#+}} xmm3 = xmm3[3,2,1,0,4,5,6,7]
 ; SSE2-NEXT:    pshufhw {{.*#+}} xmm3 = xmm3[0,1,2,3,7,6,5,4]
-; SSE2-NEXT:    packuswb %xmm4, %xmm3
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psllw $5, %xmm5
-; SSE2-NEXT:    pand %xmm10, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm4
-; SSE2-NEXT:    psllw $7, %xmm4
-; SSE2-NEXT:    pand %xmm11, %xmm4
+; SSE2-NEXT:    packuswb %xmm6, %xmm3
 ; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psllw $3, %xmm6
-; SSE2-NEXT:    pand %xmm12, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    paddb %xmm5, %xmm5
-; SSE2-NEXT:    pand %xmm8, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm13, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    movdqa %xmm3, %xmm5
-; SSE2-NEXT:    psrlw $3, %xmm5
-; SSE2-NEXT:    pand %xmm14, %xmm5
-; SSE2-NEXT:    por %xmm6, %xmm5
-; SSE2-NEXT:    movdqa %xmm3, %xmm6
-; SSE2-NEXT:    psrlw $5, %xmm6
-; SSE2-NEXT:    pand %xmm15, %xmm6
-; SSE2-NEXT:    por %xmm5, %xmm6
-; SSE2-NEXT:    psrlw $7, %xmm3
-; SSE2-NEXT:    pand %xmm7, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm6
+; SSE2-NEXT:    psllw $4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm6
+; SSE2-NEXT:    pand %xmm4, %xmm3
+; SSE2-NEXT:    psrlw $4, %xmm3
+; SSE2-NEXT:    pand %xmm8, %xmm3
 ; SSE2-NEXT:    por %xmm6, %xmm3
-; SSE2-NEXT:    por %xmm4, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm5
+; SSE2-NEXT:    psllw $2, %xmm5
+; SSE2-NEXT:    pand %xmm9, %xmm5
+; SSE2-NEXT:    pand %xmm10, %xmm3
+; SSE2-NEXT:    psrlw $2, %xmm3
+; SSE2-NEXT:    pand %xmm11, %xmm3
+; SSE2-NEXT:    por %xmm5, %xmm3
+; SSE2-NEXT:    pand %xmm3, %xmm7
+; SSE2-NEXT:    psrlw $1, %xmm7
+; SSE2-NEXT:    pand %xmm12, %xmm7
+; SSE2-NEXT:    pand %xmm13, %xmm3
+; SSE2-NEXT:    paddb %xmm3, %xmm3
+; SSE2-NEXT:    por %xmm7, %xmm3
 ; SSE2-NEXT:    retq
 ;
 ; SSSE3-LABEL: test_bitreverse_v8i64:
 ; SSSE3:       # BB#0:
 ; SSSE3-NEXT:    movdqa %xmm1, %xmm5
 ; SSSE3-NEXT:    movdqa %xmm0, %xmm1
 ; SSSE3-NEXT:    movdqa {{.*#+}} xmm8 = [7,6,5,4,3,2,1,0,15,14,13,12,11,10,9,8]
 ; SSSE3-NEXT:    pshufb %xmm8, %xmm1

 ; AVX2-NEXT:    vpsrlw $4, %ymm1, %ymm1
 ; AVX2-NEXT:    vpand %ymm3, %ymm1, %ymm1
 ; AVX2-NEXT:    vpshufb %ymm1, %ymm6, %ymm1
 ; AVX2-NEXT:    vpor %ymm1, %ymm2, %ymm1
 ; AVX2-NEXT:    retq
 ;
 ; AVX512F-LABEL: test_bitreverse_v8i64:
 ; AVX512F:       # BB#0:
-; AVX512F-NEXT:    vpsllq $61, %zmm0, %zmm1
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $63, %zmm0, %zmm2
+; AVX512F-NEXT:    vpsrlq $56, %zmm0, %zmm1
+; AVX512F-NEXT:    vpsrlq $40, %zmm0, %zmm2
 ; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
 ; AVX512F-NEXT:    vporq %zmm1, %zmm2, %zmm1
-; AVX512F-NEXT:    vpsllq $59, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $57, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $55, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $53, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $51, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $49, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $47, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $45, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $43, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $41, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $39, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $37, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $35, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $33, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $31, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $29, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $27, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $25, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $23, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $21, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $19, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $17, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $15, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $13, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $11, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $9, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $7, %zmm0, %zmm2
+; AVX512F-NEXT:    vpsrlq $24, %zmm0, %zmm2
 ; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $5, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $3, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsllq $1, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $1, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $3, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $5, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $7, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $9, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $11, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $13, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $15, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $17, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $19, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $21, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $23, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $25, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $27, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $29, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $31, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $33, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $35, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $37, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $39, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $41, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $43, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $45, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $47, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $49, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $51, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $53, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $55, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $57, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $59, %zmm0, %zmm2
-; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $61, %zmm0, %zmm2
+; AVX512F-NEXT:    vpsrlq $8, %zmm0, %zmm3
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm3, %zmm3
+; AVX512F-NEXT:    vporq %zmm2, %zmm3, %zmm2
+; AVX512F-NEXT:    vporq %zmm1, %zmm2, %zmm1
+; AVX512F-NEXT:    vpsllq $8, %zmm0, %zmm2
 ; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm2, %zmm2
-; AVX512F-NEXT:    vporq %zmm2, %zmm1, %zmm1
-; AVX512F-NEXT:    vpsrlq $63, %zmm0, %zmm0
+; AVX512F-NEXT:    vpsllq $24, %zmm0, %zmm3
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm3, %zmm3
+; AVX512F-NEXT:    vporq %zmm2, %zmm3, %zmm2
+; AVX512F-NEXT:    vpsllq $56, %zmm0, %zmm3
+; AVX512F-NEXT:    vpsllq $40, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm0
+; AVX512F-NEXT:    vporq %zmm0, %zmm3, %zmm0
+; AVX512F-NEXT:    vporq %zmm2, %zmm0, %zmm0
+; AVX512F-NEXT:    vporq %zmm1, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm1
+; AVX512F-NEXT:    vpsllq $4, %zmm1, %zmm1
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm0
+; AVX512F-NEXT:    vpsrlq $4, %zmm0, %zmm0
+; AVX512F-NEXT:    vporq %zmm1, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm1
+; AVX512F-NEXT:    vpsllq $2, %zmm1, %zmm1
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm0
+; AVX512F-NEXT:    vpsrlq $2, %zmm0, %zmm0
+; AVX512F-NEXT:    vporq %zmm1, %zmm0, %zmm0
+; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm1
+; AVX512F-NEXT:    vpsllq $1, %zmm1, %zmm1
 ; AVX512F-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm0
-; AVX512F-NEXT:    vporq %zmm0, %zmm1, %zmm0
+; AVX512F-NEXT:    vpsrlq $1, %zmm0, %zmm0
+; AVX512F-NEXT:    vporq %zmm1, %zmm0, %zmm0
 ; AVX512F-NEXT:    retq
 ;
 ; AVX512BW-LABEL: test_bitreverse_v8i64:
 ; AVX512BW:       # BB#0:
 ; AVX512BW-NEXT:    vpshufb {{.*#+}} zmm0 = zmm0[7,6,5,4,3,2,1,0,15,14,13,12,11,10,9,8,23,22,21,20,19,18,17,16,31,30,29,28,27,26,25,24,39,38,37,36,35,34,33,32,47,46,45,44,43,42,41,40,55,54,53,52,51,50,49,48,63,62,61,60,59,58,57,56]
 ; AVX512BW-NEXT:    vmovdqa64 {{.*#+}} zmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
 ; AVX512BW-NEXT:    vpandq %zmm1, %zmm0, %zmm2
 ; AVX512BW-NEXT:    vmovdqu8 {{.*#+}} zmm3 = [0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240,0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240,0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240,0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240]