From 7e849fd97b6abd1de30ffc66da1ebf47afc7ab70 Mon Sep 17 00:00:00 2001
From: Simon Pilgrim <llvm-dev@redking.me.uk>
Date: Sat, 29 Jan 2022 13:58:25 +0000
Subject: [PATCH] [X86] LowerFunnelShift - allow non-constant vXi8 unpack(y,x)
 << zext(z) lowering pre-AVX512

Without AVX512 (which can efficiently extend/truncate to vXi16/vXi32), unpacking/packing to vXi16 is more efficient that relying on the (uops-heavy) PBLENDV shift expansion
---
 llvm/lib/Target/X86/X86ISelLowering.cpp  |   2 +-
 llvm/test/CodeGen/X86/vector-fshl-128.ll | 368 +++++++++++------------
 llvm/test/CodeGen/X86/vector-fshl-256.ll | 177 ++++++-----
 3 files changed, 259 insertions(+), 288 deletions(-)

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 4fc79bc49f52..378eb97a7469 100644
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -29915,7 +29915,7 @@ static SDValue LowerFunnelShift(SDValue Op, const X86Subtarget &Subtarget,
     }
 
     // Attempt to fold per-element (ExtVT) shift as unpack(y,x) << zext(z)
-    if ((IsCst && !IsFSHR && EltSizeInBits == 8) ||
+    if (((IsCst || !Subtarget.hasAVX512()) && !IsFSHR && EltSizeInBits == 8) ||
         supportedVectorVarShift(ExtVT, Subtarget, ShiftOpc)) {
       SDValue Z = DAG.getConstant(0, DL, VT);
       SDValue RLo = DAG.getBitcast(ExtVT, getUnpackl(DAG, DL, VT, Op1, Op0));
diff --git a/llvm/test/CodeGen/X86/vector-fshl-128.ll b/llvm/test/CodeGen/X86/vector-fshl-128.ll
index b5d8c4f21a21..dae44a55c617 100644
--- a/llvm/test/CodeGen/X86/vector-fshl-128.ll
+++ b/llvm/test/CodeGen/X86/vector-fshl-128.ll
@@ -726,149 +726,145 @@ define <8 x i16> @var_funnnel_v8i16(<8 x i16> %x, <8 x i16> %y, <8 x i16> %amt)
 define <16 x i8> @var_funnnel_v16i8(<16 x i8> %x, <16 x i8> %y, <16 x i8> %amt) nounwind {
 ; SSE2-LABEL: var_funnnel_v16i8:
 ; SSE2:       # %bb.0:
-; SSE2-NEXT:    psrlw $1, %xmm1
-; SSE2-NEXT:    movdqa {{.*#+}} xmm5 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; SSE2-NEXT:    pand %xmm5, %xmm1
-; SSE2-NEXT:    movdqa {{.*#+}} xmm8 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
-; SSE2-NEXT:    movdqa %xmm2, %xmm7
-; SSE2-NEXT:    pandn %xmm8, %xmm7
-; SSE2-NEXT:    psllw $5, %xmm7
-; SSE2-NEXT:    pxor %xmm3, %xmm3
-; SSE2-NEXT:    pxor %xmm6, %xmm6
-; SSE2-NEXT:    pcmpgtb %xmm7, %xmm6
-; SSE2-NEXT:    movdqa %xmm6, %xmm4
-; SSE2-NEXT:    pandn %xmm1, %xmm4
-; SSE2-NEXT:    psrlw $4, %xmm1
-; SSE2-NEXT:    pand %xmm1, %xmm6
-; SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6
-; SSE2-NEXT:    por %xmm4, %xmm6
-; SSE2-NEXT:    paddb %xmm7, %xmm7
-; SSE2-NEXT:    pxor %xmm1, %xmm1
-; SSE2-NEXT:    pcmpgtb %xmm7, %xmm1
+; SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
+; SSE2-NEXT:    pxor %xmm5, %xmm5
+; SSE2-NEXT:    movdqa %xmm2, %xmm4
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm5[8],xmm4[9],xmm5[9],xmm4[10],xmm5[10],xmm4[11],xmm5[11],xmm4[12],xmm5[12],xmm4[13],xmm5[13],xmm4[14],xmm5[14],xmm4[15],xmm5[15]
+; SSE2-NEXT:    movdqa %xmm4, %xmm6
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm6 = xmm6[4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pslld $23, %xmm6
+; SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [1065353216,1065353216,1065353216,1065353216]
+; SSE2-NEXT:    paddd %xmm3, %xmm6
+; SSE2-NEXT:    cvttps2dq %xmm6, %xmm6
+; SSE2-NEXT:    pslld $16, %xmm6
+; SSE2-NEXT:    psrad $16, %xmm6
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0,0,1,1,2,2,3,3]
+; SSE2-NEXT:    pslld $23, %xmm4
+; SSE2-NEXT:    paddd %xmm3, %xmm4
+; SSE2-NEXT:    cvttps2dq %xmm4, %xmm7
+; SSE2-NEXT:    pslld $16, %xmm7
+; SSE2-NEXT:    psrad $16, %xmm7
+; SSE2-NEXT:    packssdw %xmm6, %xmm7
 ; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    pandn %xmm6, %xmm4
-; SSE2-NEXT:    psrlw $2, %xmm6
-; SSE2-NEXT:    pand %xmm1, %xmm6
-; SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6
-; SSE2-NEXT:    por %xmm4, %xmm6
-; SSE2-NEXT:    paddb %xmm7, %xmm7
-; SSE2-NEXT:    pxor %xmm1, %xmm1
-; SSE2-NEXT:    pcmpgtb %xmm7, %xmm1
-; SSE2-NEXT:    movdqa %xmm1, %xmm4
-; SSE2-NEXT:    pandn %xmm6, %xmm4
-; SSE2-NEXT:    psrlw $1, %xmm6
-; SSE2-NEXT:    pand %xmm5, %xmm1
-; SSE2-NEXT:    pand %xmm6, %xmm1
-; SSE2-NEXT:    por %xmm4, %xmm1
-; SSE2-NEXT:    pand %xmm8, %xmm2
-; SSE2-NEXT:    psllw $5, %xmm2
-; SSE2-NEXT:    pxor %xmm4, %xmm4
-; SSE2-NEXT:    pcmpgtb %xmm2, %xmm4
-; SSE2-NEXT:    movdqa %xmm4, %xmm5
-; SSE2-NEXT:    pandn %xmm0, %xmm5
-; SSE2-NEXT:    psllw $4, %xmm0
-; SSE2-NEXT:    pand %xmm4, %xmm0
-; SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm5, %xmm0
-; SSE2-NEXT:    paddb %xmm2, %xmm2
-; SSE2-NEXT:    pxor %xmm4, %xmm4
-; SSE2-NEXT:    pcmpgtb %xmm2, %xmm4
-; SSE2-NEXT:    movdqa %xmm4, %xmm5
-; SSE2-NEXT:    pandn %xmm0, %xmm5
-; SSE2-NEXT:    psllw $2, %xmm0
-; SSE2-NEXT:    pand %xmm4, %xmm0
-; SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0
-; SSE2-NEXT:    por %xmm5, %xmm0
-; SSE2-NEXT:    paddb %xmm2, %xmm2
-; SSE2-NEXT:    pcmpgtb %xmm2, %xmm3
-; SSE2-NEXT:    movdqa %xmm3, %xmm2
-; SSE2-NEXT:    pandn %xmm0, %xmm2
-; SSE2-NEXT:    por %xmm1, %xmm2
-; SSE2-NEXT:    paddb %xmm0, %xmm0
-; SSE2-NEXT:    pand %xmm3, %xmm0
-; SSE2-NEXT:    por %xmm2, %xmm0
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm0[8],xmm4[9],xmm0[9],xmm4[10],xmm0[10],xmm4[11],xmm0[11],xmm4[12],xmm0[12],xmm4[13],xmm0[13],xmm4[14],xmm0[14],xmm4[15],xmm0[15]
+; SSE2-NEXT:    pmullw %xmm7, %xmm4
+; SSE2-NEXT:    psrlw $8, %xmm4
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm5[0],xmm2[1],xmm5[1],xmm2[2],xmm5[2],xmm2[3],xmm5[3],xmm2[4],xmm5[4],xmm2[5],xmm5[5],xmm2[6],xmm5[6],xmm2[7],xmm5[7]
+; SSE2-NEXT:    movdqa %xmm2, %xmm5
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm5 = xmm5[4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    pslld $23, %xmm5
+; SSE2-NEXT:    paddd %xmm3, %xmm5
+; SSE2-NEXT:    cvttps2dq %xmm5, %xmm5
+; SSE2-NEXT:    pslld $16, %xmm5
+; SSE2-NEXT:    psrad $16, %xmm5
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0,0,1,1,2,2,3,3]
+; SSE2-NEXT:    pslld $23, %xmm2
+; SSE2-NEXT:    paddd %xmm3, %xmm2
+; SSE2-NEXT:    cvttps2dq %xmm2, %xmm2
+; SSE2-NEXT:    pslld $16, %xmm2
+; SSE2-NEXT:    psrad $16, %xmm2
+; SSE2-NEXT:    packssdw %xmm5, %xmm2
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3],xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
+; SSE2-NEXT:    pmullw %xmm1, %xmm2
+; SSE2-NEXT:    psrlw $8, %xmm2
+; SSE2-NEXT:    packuswb %xmm4, %xmm2
+; SSE2-NEXT:    movdqa %xmm2, %xmm0
 ; SSE2-NEXT:    retq
 ;
 ; SSE41-LABEL: var_funnnel_v16i8:
 ; SSE41:       # %bb.0:
-; SSE41-NEXT:    movdqa %xmm0, %xmm4
-; SSE41-NEXT:    movdqa {{.*#+}} xmm3 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
+; SSE41-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2
+; SSE41-NEXT:    pxor %xmm3, %xmm3
+; SSE41-NEXT:    pmovzxbd {{.*#+}} xmm4 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero
+; SSE41-NEXT:    pmovzxbw {{.*#+}} xmm5 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero
+; SSE41-NEXT:    punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm3[8],xmm2[9],xmm3[9],xmm2[10],xmm3[10],xmm2[11],xmm3[11],xmm2[12],xmm3[12],xmm2[13],xmm3[13],xmm2[14],xmm3[14],xmm2[15],xmm3[15]
+; SSE41-NEXT:    pmovzxwd {{.*#+}} xmm3 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero
+; SSE41-NEXT:    punpckhwd {{.*#+}} xmm2 = xmm2[4,4,5,5,6,6,7,7]
+; SSE41-NEXT:    pslld $23, %xmm2
+; SSE41-NEXT:    movdqa {{.*#+}} xmm6 = [1065353216,1065353216,1065353216,1065353216]
+; SSE41-NEXT:    paddd %xmm6, %xmm2
+; SSE41-NEXT:    cvttps2dq %xmm2, %xmm2
+; SSE41-NEXT:    pslld $23, %xmm3
+; SSE41-NEXT:    paddd %xmm6, %xmm3
+; SSE41-NEXT:    cvttps2dq %xmm3, %xmm3
+; SSE41-NEXT:    packusdw %xmm2, %xmm3
+; SSE41-NEXT:    movdqa %xmm1, %xmm7
+; SSE41-NEXT:    punpckhbw {{.*#+}} xmm7 = xmm7[8],xmm0[8],xmm7[9],xmm0[9],xmm7[10],xmm0[10],xmm7[11],xmm0[11],xmm7[12],xmm0[12],xmm7[13],xmm0[13],xmm7[14],xmm0[14],xmm7[15],xmm0[15]
+; SSE41-NEXT:    pmullw %xmm3, %xmm7
+; SSE41-NEXT:    psrlw $8, %xmm7
+; SSE41-NEXT:    pslld $23, %xmm4
+; SSE41-NEXT:    paddd %xmm6, %xmm4
+; SSE41-NEXT:    cvttps2dq %xmm4, %xmm2
+; SSE41-NEXT:    punpckhwd {{.*#+}} xmm5 = xmm5[4,4,5,5,6,6,7,7]
+; SSE41-NEXT:    pslld $23, %xmm5
+; SSE41-NEXT:    paddd %xmm6, %xmm5
+; SSE41-NEXT:    cvttps2dq %xmm5, %xmm3
+; SSE41-NEXT:    packusdw %xmm3, %xmm2
+; SSE41-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3],xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
+; SSE41-NEXT:    pmullw %xmm1, %xmm2
+; SSE41-NEXT:    psrlw $8, %xmm2
+; SSE41-NEXT:    packuswb %xmm7, %xmm2
 ; SSE41-NEXT:    movdqa %xmm2, %xmm0
-; SSE41-NEXT:    pandn %xmm3, %xmm0
-; SSE41-NEXT:    psllw $5, %xmm0
-; SSE41-NEXT:    psrlw $1, %xmm1
-; SSE41-NEXT:    movdqa {{.*#+}} xmm5 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; SSE41-NEXT:    pand %xmm5, %xmm1
-; SSE41-NEXT:    movdqa %xmm1, %xmm6
-; SSE41-NEXT:    psrlw $4, %xmm6
-; SSE41-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6
-; SSE41-NEXT:    pblendvb %xmm0, %xmm6, %xmm1
-; SSE41-NEXT:    movdqa %xmm1, %xmm6
-; SSE41-NEXT:    psrlw $2, %xmm6
-; SSE41-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6
-; SSE41-NEXT:    paddb %xmm0, %xmm0
-; SSE41-NEXT:    pblendvb %xmm0, %xmm6, %xmm1
-; SSE41-NEXT:    movdqa %xmm1, %xmm6
-; SSE41-NEXT:    psrlw $1, %xmm6
-; SSE41-NEXT:    pand %xmm5, %xmm6
-; SSE41-NEXT:    paddb %xmm0, %xmm0
-; SSE41-NEXT:    pblendvb %xmm0, %xmm6, %xmm1
-; SSE41-NEXT:    pand %xmm2, %xmm3
-; SSE41-NEXT:    psllw $5, %xmm3
-; SSE41-NEXT:    movdqa %xmm3, %xmm2
-; SSE41-NEXT:    paddb %xmm3, %xmm2
-; SSE41-NEXT:    movdqa %xmm4, %xmm5
-; SSE41-NEXT:    psllw $4, %xmm5
-; SSE41-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm5
-; SSE41-NEXT:    movdqa %xmm3, %xmm0
-; SSE41-NEXT:    pblendvb %xmm0, %xmm5, %xmm4
-; SSE41-NEXT:    movdqa %xmm4, %xmm3
-; SSE41-NEXT:    psllw $2, %xmm3
-; SSE41-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm3
-; SSE41-NEXT:    movdqa %xmm2, %xmm0
-; SSE41-NEXT:    pblendvb %xmm0, %xmm3, %xmm4
-; SSE41-NEXT:    movdqa %xmm4, %xmm3
-; SSE41-NEXT:    paddb %xmm4, %xmm3
-; SSE41-NEXT:    paddb %xmm2, %xmm2
-; SSE41-NEXT:    movdqa %xmm2, %xmm0
-; SSE41-NEXT:    pblendvb %xmm0, %xmm3, %xmm4
-; SSE41-NEXT:    por %xmm1, %xmm4
-; SSE41-NEXT:    movdqa %xmm4, %xmm0
 ; SSE41-NEXT:    retq
 ;
-; AVX-LABEL: var_funnnel_v16i8:
-; AVX:       # %bb.0:
-; AVX-NEXT:    vmovdqa {{.*#+}} xmm3 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
-; AVX-NEXT:    vpandn %xmm3, %xmm2, %xmm4
-; AVX-NEXT:    vpsllw $5, %xmm4, %xmm4
-; AVX-NEXT:    vpsrlw $1, %xmm1, %xmm1
-; AVX-NEXT:    vmovdqa {{.*#+}} xmm5 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX-NEXT:    vpand %xmm5, %xmm1, %xmm1
-; AVX-NEXT:    vpsrlw $4, %xmm1, %xmm6
-; AVX-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6, %xmm6
-; AVX-NEXT:    vpblendvb %xmm4, %xmm6, %xmm1, %xmm1
-; AVX-NEXT:    vpsrlw $2, %xmm1, %xmm6
-; AVX-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm6, %xmm6
-; AVX-NEXT:    vpaddb %xmm4, %xmm4, %xmm4
-; AVX-NEXT:    vpblendvb %xmm4, %xmm6, %xmm1, %xmm1
-; AVX-NEXT:    vpsrlw $1, %xmm1, %xmm6
-; AVX-NEXT:    vpand %xmm5, %xmm6, %xmm5
-; AVX-NEXT:    vpaddb %xmm4, %xmm4, %xmm4
-; AVX-NEXT:    vpblendvb %xmm4, %xmm5, %xmm1, %xmm1
-; AVX-NEXT:    vpand %xmm3, %xmm2, %xmm2
-; AVX-NEXT:    vpsllw $5, %xmm2, %xmm2
-; AVX-NEXT:    vpaddb %xmm2, %xmm2, %xmm3
-; AVX-NEXT:    vpsllw $4, %xmm0, %xmm4
-; AVX-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm4, %xmm4
-; AVX-NEXT:    vpblendvb %xmm2, %xmm4, %xmm0, %xmm0
-; AVX-NEXT:    vpsllw $2, %xmm0, %xmm2
-; AVX-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2, %xmm2
-; AVX-NEXT:    vpblendvb %xmm3, %xmm2, %xmm0, %xmm0
-; AVX-NEXT:    vpaddb %xmm0, %xmm0, %xmm2
-; AVX-NEXT:    vpaddb %xmm3, %xmm3, %xmm3
-; AVX-NEXT:    vpblendvb %xmm3, %xmm2, %xmm0, %xmm0
-; AVX-NEXT:    vpor %xmm1, %xmm0, %xmm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: var_funnnel_v16i8:
+; AVX1:       # %bb.0:
+; AVX1-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2, %xmm2
+; AVX1-NEXT:    vpxor %xmm3, %xmm3, %xmm3
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm3 = xmm2[8],xmm3[8],xmm2[9],xmm3[9],xmm2[10],xmm3[10],xmm2[11],xmm3[11],xmm2[12],xmm3[12],xmm2[13],xmm3[13],xmm2[14],xmm3[14],xmm2[15],xmm3[15]
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm4 = xmm3[4,4,5,5,6,6,7,7]
+; AVX1-NEXT:    vpslld $23, %xmm4, %xmm4
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm5 = [1065353216,1065353216,1065353216,1065353216]
+; AVX1-NEXT:    vpaddd %xmm5, %xmm4, %xmm4
+; AVX1-NEXT:    vcvttps2dq %xmm4, %xmm4
+; AVX1-NEXT:    vpmovzxwd {{.*#+}} xmm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero
+; AVX1-NEXT:    vpslld $23, %xmm3, %xmm3
+; AVX1-NEXT:    vpaddd %xmm5, %xmm3, %xmm3
+; AVX1-NEXT:    vcvttps2dq %xmm3, %xmm3
+; AVX1-NEXT:    vpackusdw %xmm4, %xmm3, %xmm3
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm4 = xmm1[8],xmm0[8],xmm1[9],xmm0[9],xmm1[10],xmm0[10],xmm1[11],xmm0[11],xmm1[12],xmm0[12],xmm1[13],xmm0[13],xmm1[14],xmm0[14],xmm1[15],xmm0[15]
+; AVX1-NEXT:    vpmullw %xmm3, %xmm4, %xmm3
+; AVX1-NEXT:    vpsrlw $8, %xmm3, %xmm3
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm4 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero
+; AVX1-NEXT:    vpslld $23, %xmm4, %xmm4
+; AVX1-NEXT:    vpaddd %xmm5, %xmm4, %xmm4
+; AVX1-NEXT:    vcvttps2dq %xmm4, %xmm4
+; AVX1-NEXT:    vpmovzxbw {{.*#+}} xmm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm2[4,4,5,5,6,6,7,7]
+; AVX1-NEXT:    vpslld $23, %xmm2, %xmm2
+; AVX1-NEXT:    vpaddd %xmm5, %xmm2, %xmm2
+; AVX1-NEXT:    vcvttps2dq %xmm2, %xmm2
+; AVX1-NEXT:    vpackusdw %xmm2, %xmm4, %xmm2
+; AVX1-NEXT:    vpunpcklbw {{.*#+}} xmm0 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3],xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
+; AVX1-NEXT:    vpmullw %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpsrlw $8, %xmm0, %xmm0
+; AVX1-NEXT:    vpackuswb %xmm3, %xmm0, %xmm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: var_funnnel_v16i8:
+; AVX2:       # %bb.0:
+; AVX2-NEXT:    vpunpckhbw {{.*#+}} xmm3 = xmm1[8],xmm0[8],xmm1[9],xmm0[9],xmm1[10],xmm0[10],xmm1[11],xmm0[11],xmm1[12],xmm0[12],xmm1[13],xmm0[13],xmm1[14],xmm0[14],xmm1[15],xmm0[15]
+; AVX2-NEXT:    vpmovzxwd {{.*#+}} ymm3 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero
+; AVX2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm2, %xmm2
+; AVX2-NEXT:    vpxor %xmm4, %xmm4, %xmm4
+; AVX2-NEXT:    vpunpckhbw {{.*#+}} xmm4 = 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]
+; AVX2-NEXT:    vpmovzxwd {{.*#+}} ymm4 = xmm4[0],zero,xmm4[1],zero,xmm4[2],zero,xmm4[3],zero,xmm4[4],zero,xmm4[5],zero,xmm4[6],zero,xmm4[7],zero
+; AVX2-NEXT:    vpsllvd %ymm4, %ymm3, %ymm3
+; AVX2-NEXT:    vmovdqa {{.*#+}} ymm4 = [0,1,4,5,8,9,12,13,8,9,12,13,12,13,14,15,16,17,20,21,24,25,28,29,24,25,28,29,28,29,30,31]
+; AVX2-NEXT:    vpshufb %ymm4, %ymm3, %ymm3
+; AVX2-NEXT:    vpermq {{.*#+}} ymm3 = ymm3[0,2,2,3]
+; AVX2-NEXT:    vpsrlw $8, %xmm3, %xmm3
+; AVX2-NEXT:    vpunpcklbw {{.*#+}} xmm0 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3],xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
+; AVX2-NEXT:    vpmovzxwd {{.*#+}} ymm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} xmm1 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero
+; AVX2-NEXT:    vpmovzxwd {{.*#+}} ymm1 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero
+; AVX2-NEXT:    vpsllvd %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpshufb %ymm4, %ymm0, %ymm0
+; AVX2-NEXT:    vpermq {{.*#+}} ymm0 = ymm0[0,2,2,3]
+; AVX2-NEXT:    vpsrlw $8, %xmm0, %xmm0
+; AVX2-NEXT:    vpackuswb %xmm3, %xmm0, %xmm0
+; AVX2-NEXT:    vzeroupper
+; AVX2-NEXT:    retq
 ;
 ; AVX512F-LABEL: var_funnnel_v16i8:
 ; AVX512F:       # %bb.0:
@@ -980,67 +976,49 @@ define <16 x i8> @var_funnnel_v16i8(<16 x i8> %x, <16 x i8> %y, <16 x i8> %amt)
 ;
 ; X86-SSE2-LABEL: var_funnnel_v16i8:
 ; X86-SSE2:       # %bb.0:
-; X86-SSE2-NEXT:    psrlw $1, %xmm1
-; X86-SSE2-NEXT:    movdqa {{.*#+}} xmm4 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; X86-SSE2-NEXT:    pand %xmm4, %xmm1
-; X86-SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
-; X86-SSE2-NEXT:    movdqa %xmm2, %xmm6
-; X86-SSE2-NEXT:    pandn %xmm3, %xmm6
-; X86-SSE2-NEXT:    psllw $5, %xmm6
+; X86-SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}, %xmm2
 ; X86-SSE2-NEXT:    pxor %xmm5, %xmm5
-; X86-SSE2-NEXT:    pcmpgtb %xmm6, %xmm5
-; X86-SSE2-NEXT:    movdqa %xmm5, %xmm7
-; X86-SSE2-NEXT:    pandn %xmm1, %xmm7
-; X86-SSE2-NEXT:    psrlw $4, %xmm1
-; X86-SSE2-NEXT:    pand %xmm1, %xmm5
-; X86-SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}, %xmm5
-; X86-SSE2-NEXT:    por %xmm7, %xmm5
-; X86-SSE2-NEXT:    paddb %xmm6, %xmm6
-; X86-SSE2-NEXT:    pxor %xmm1, %xmm1
-; X86-SSE2-NEXT:    pcmpgtb %xmm6, %xmm1
-; X86-SSE2-NEXT:    movdqa %xmm1, %xmm7
-; X86-SSE2-NEXT:    pandn %xmm5, %xmm7
-; X86-SSE2-NEXT:    psrlw $2, %xmm5
-; X86-SSE2-NEXT:    pand %xmm1, %xmm5
-; X86-SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}, %xmm5
-; X86-SSE2-NEXT:    por %xmm7, %xmm5
-; X86-SSE2-NEXT:    paddb %xmm6, %xmm6
-; X86-SSE2-NEXT:    pxor %xmm1, %xmm1
-; X86-SSE2-NEXT:    pcmpgtb %xmm6, %xmm1
-; X86-SSE2-NEXT:    movdqa %xmm1, %xmm6
-; X86-SSE2-NEXT:    pand %xmm4, %xmm1
-; X86-SSE2-NEXT:    pandn %xmm5, %xmm6
-; X86-SSE2-NEXT:    psrlw $1, %xmm5
-; X86-SSE2-NEXT:    pand %xmm5, %xmm1
-; X86-SSE2-NEXT:    por %xmm6, %xmm1
-; X86-SSE2-NEXT:    pand %xmm3, %xmm2
-; X86-SSE2-NEXT:    psllw $5, %xmm2
-; X86-SSE2-NEXT:    pxor %xmm3, %xmm3
-; X86-SSE2-NEXT:    pcmpgtb %xmm2, %xmm3
-; X86-SSE2-NEXT:    movdqa %xmm3, %xmm4
-; X86-SSE2-NEXT:    pandn %xmm0, %xmm4
-; X86-SSE2-NEXT:    psllw $4, %xmm0
-; X86-SSE2-NEXT:    pand %xmm3, %xmm0
-; X86-SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0
-; X86-SSE2-NEXT:    por %xmm4, %xmm0
-; X86-SSE2-NEXT:    paddb %xmm2, %xmm2
-; X86-SSE2-NEXT:    pxor %xmm3, %xmm3
-; X86-SSE2-NEXT:    pcmpgtb %xmm2, %xmm3
-; X86-SSE2-NEXT:    movdqa %xmm3, %xmm4
-; X86-SSE2-NEXT:    pandn %xmm0, %xmm4
-; X86-SSE2-NEXT:    psllw $2, %xmm0
-; X86-SSE2-NEXT:    pand %xmm3, %xmm0
-; X86-SSE2-NEXT:    pand {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0
-; X86-SSE2-NEXT:    por %xmm4, %xmm0
-; X86-SSE2-NEXT:    pxor %xmm3, %xmm3
-; X86-SSE2-NEXT:    paddb %xmm2, %xmm2
-; X86-SSE2-NEXT:    pcmpgtb %xmm2, %xmm3
-; X86-SSE2-NEXT:    movdqa %xmm3, %xmm2
-; X86-SSE2-NEXT:    pandn %xmm0, %xmm2
-; X86-SSE2-NEXT:    por %xmm1, %xmm2
-; X86-SSE2-NEXT:    paddb %xmm0, %xmm0
-; X86-SSE2-NEXT:    pand %xmm3, %xmm0
-; X86-SSE2-NEXT:    por %xmm2, %xmm0
+; X86-SSE2-NEXT:    movdqa %xmm2, %xmm4
+; X86-SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm5[8],xmm4[9],xmm5[9],xmm4[10],xmm5[10],xmm4[11],xmm5[11],xmm4[12],xmm5[12],xmm4[13],xmm5[13],xmm4[14],xmm5[14],xmm4[15],xmm5[15]
+; X86-SSE2-NEXT:    movdqa %xmm4, %xmm6
+; X86-SSE2-NEXT:    punpckhwd {{.*#+}} xmm6 = xmm6[4,4,5,5,6,6,7,7]
+; X86-SSE2-NEXT:    pslld $23, %xmm6
+; X86-SSE2-NEXT:    movdqa {{.*#+}} xmm3 = [1065353216,1065353216,1065353216,1065353216]
+; X86-SSE2-NEXT:    paddd %xmm3, %xmm6
+; X86-SSE2-NEXT:    cvttps2dq %xmm6, %xmm6
+; X86-SSE2-NEXT:    pslld $16, %xmm6
+; X86-SSE2-NEXT:    psrad $16, %xmm6
+; X86-SSE2-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0,0,1,1,2,2,3,3]
+; X86-SSE2-NEXT:    pslld $23, %xmm4
+; X86-SSE2-NEXT:    paddd %xmm3, %xmm4
+; X86-SSE2-NEXT:    cvttps2dq %xmm4, %xmm7
+; X86-SSE2-NEXT:    pslld $16, %xmm7
+; X86-SSE2-NEXT:    psrad $16, %xmm7
+; X86-SSE2-NEXT:    packssdw %xmm6, %xmm7
+; X86-SSE2-NEXT:    movdqa %xmm1, %xmm4
+; X86-SSE2-NEXT:    punpckhbw {{.*#+}} xmm4 = xmm4[8],xmm0[8],xmm4[9],xmm0[9],xmm4[10],xmm0[10],xmm4[11],xmm0[11],xmm4[12],xmm0[12],xmm4[13],xmm0[13],xmm4[14],xmm0[14],xmm4[15],xmm0[15]
+; X86-SSE2-NEXT:    pmullw %xmm7, %xmm4
+; X86-SSE2-NEXT:    psrlw $8, %xmm4
+; X86-SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm5[0],xmm2[1],xmm5[1],xmm2[2],xmm5[2],xmm2[3],xmm5[3],xmm2[4],xmm5[4],xmm2[5],xmm5[5],xmm2[6],xmm5[6],xmm2[7],xmm5[7]
+; X86-SSE2-NEXT:    movdqa %xmm2, %xmm5
+; X86-SSE2-NEXT:    punpckhwd {{.*#+}} xmm5 = xmm5[4,4,5,5,6,6,7,7]
+; X86-SSE2-NEXT:    pslld $23, %xmm5
+; X86-SSE2-NEXT:    paddd %xmm3, %xmm5
+; X86-SSE2-NEXT:    cvttps2dq %xmm5, %xmm5
+; X86-SSE2-NEXT:    pslld $16, %xmm5
+; X86-SSE2-NEXT:    psrad $16, %xmm5
+; X86-SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0,0,1,1,2,2,3,3]
+; X86-SSE2-NEXT:    pslld $23, %xmm2
+; X86-SSE2-NEXT:    paddd %xmm3, %xmm2
+; X86-SSE2-NEXT:    cvttps2dq %xmm2, %xmm2
+; X86-SSE2-NEXT:    pslld $16, %xmm2
+; X86-SSE2-NEXT:    psrad $16, %xmm2
+; X86-SSE2-NEXT:    packssdw %xmm5, %xmm2
+; X86-SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3],xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
+; X86-SSE2-NEXT:    pmullw %xmm1, %xmm2
+; X86-SSE2-NEXT:    psrlw $8, %xmm2
+; X86-SSE2-NEXT:    packuswb %xmm4, %xmm2
+; X86-SSE2-NEXT:    movdqa %xmm2, %xmm0
 ; X86-SSE2-NEXT:    retl
   %res = call <16 x i8> @llvm.fshl.v16i8(<16 x i8> %x, <16 x i8> %y, <16 x i8> %amt)
   ret <16 x i8> %res
diff --git a/llvm/test/CodeGen/X86/vector-fshl-256.ll b/llvm/test/CodeGen/X86/vector-fshl-256.ll
index 41f980886d00..f29a9742e981 100644
--- a/llvm/test/CodeGen/X86/vector-fshl-256.ll
+++ b/llvm/test/CodeGen/X86/vector-fshl-256.ll
@@ -524,105 +524,98 @@ define <16 x i16> @var_funnnel_v16i16(<16 x i16> %x, <16 x i16> %y, <16 x i16> %
 define <32 x i8> @var_funnnel_v32i8(<32 x i8> %x, <32 x i8> %y, <32 x i8> %amt) nounwind {
 ; AVX1-LABEL: var_funnnel_v32i8:
 ; AVX1:       # %bb.0:
-; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm3
-; AVX1-NEXT:    vpsrlw $1, %xmm3, %xmm4
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX1-NEXT:    vpand %xmm3, %xmm4, %xmm6
-; AVX1-NEXT:    vpsrlw $4, %xmm6, %xmm5
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm8 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
-; AVX1-NEXT:    vpand %xmm5, %xmm8, %xmm7
 ; AVX1-NEXT:    vandps {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm2, %ymm2
 ; AVX1-NEXT:    vextractf128 $1, %ymm2, %xmm4
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm9 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
-; AVX1-NEXT:    vpxor %xmm4, %xmm9, %xmm5
-; AVX1-NEXT:    vpsllw $5, %xmm5, %xmm5
-; AVX1-NEXT:    vpblendvb %xmm5, %xmm7, %xmm6, %xmm7
-; AVX1-NEXT:    vpsrlw $2, %xmm7, %xmm6
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm10 = [63,63,63,63,63,63,63,63,63,63,63,63,63,63,63,63]
-; AVX1-NEXT:    vpand %xmm6, %xmm10, %xmm6
-; AVX1-NEXT:    vpaddb %xmm5, %xmm5, %xmm5
-; AVX1-NEXT:    vpblendvb %xmm5, %xmm6, %xmm7, %xmm6
-; AVX1-NEXT:    vpsrlw $1, %xmm6, %xmm7
-; AVX1-NEXT:    vpand %xmm3, %xmm7, %xmm7
-; AVX1-NEXT:    vpaddb %xmm5, %xmm5, %xmm5
-; AVX1-NEXT:    vpblendvb %xmm5, %xmm7, %xmm6, %xmm5
-; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm6
-; AVX1-NEXT:    vpsllw $4, %xmm6, %xmm7
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm11 = [240,240,240,240,240,240,240,240,240,240,240,240,240,240,240,240]
-; AVX1-NEXT:    vpand %xmm7, %xmm11, %xmm7
-; AVX1-NEXT:    vpsllw $5, %xmm4, %xmm4
-; AVX1-NEXT:    vpblendvb %xmm4, %xmm7, %xmm6, %xmm6
-; AVX1-NEXT:    vpsllw $2, %xmm6, %xmm7
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm12 = [252,252,252,252,252,252,252,252,252,252,252,252,252,252,252,252]
-; AVX1-NEXT:    vpand %xmm7, %xmm12, %xmm7
-; AVX1-NEXT:    vpaddb %xmm4, %xmm4, %xmm4
-; AVX1-NEXT:    vpblendvb %xmm4, %xmm7, %xmm6, %xmm6
-; AVX1-NEXT:    vpaddb %xmm6, %xmm6, %xmm7
-; AVX1-NEXT:    vpaddb %xmm4, %xmm4, %xmm4
-; AVX1-NEXT:    vpblendvb %xmm4, %xmm7, %xmm6, %xmm4
-; AVX1-NEXT:    vpor %xmm5, %xmm4, %xmm4
-; AVX1-NEXT:    vpsrlw $1, %xmm1, %xmm1
-; AVX1-NEXT:    vpand %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vpsrlw $4, %xmm1, %xmm5
-; AVX1-NEXT:    vpand %xmm5, %xmm8, %xmm5
-; AVX1-NEXT:    vpxor %xmm2, %xmm9, %xmm6
-; AVX1-NEXT:    vpsllw $5, %xmm6, %xmm6
-; AVX1-NEXT:    vpblendvb %xmm6, %xmm5, %xmm1, %xmm1
-; AVX1-NEXT:    vpsrlw $2, %xmm1, %xmm5
-; AVX1-NEXT:    vpand %xmm5, %xmm10, %xmm5
-; AVX1-NEXT:    vpaddb %xmm6, %xmm6, %xmm6
-; AVX1-NEXT:    vpblendvb %xmm6, %xmm5, %xmm1, %xmm1
-; AVX1-NEXT:    vpsrlw $1, %xmm1, %xmm5
-; AVX1-NEXT:    vpand %xmm3, %xmm5, %xmm3
-; AVX1-NEXT:    vpaddb %xmm6, %xmm6, %xmm5
-; AVX1-NEXT:    vpblendvb %xmm5, %xmm3, %xmm1, %xmm1
-; AVX1-NEXT:    vpsllw $4, %xmm0, %xmm3
-; AVX1-NEXT:    vpand %xmm3, %xmm11, %xmm3
-; AVX1-NEXT:    vpsllw $5, %xmm2, %xmm2
-; AVX1-NEXT:    vpblendvb %xmm2, %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpsllw $2, %xmm0, %xmm3
-; AVX1-NEXT:    vpand %xmm3, %xmm12, %xmm3
-; AVX1-NEXT:    vpaddb %xmm2, %xmm2, %xmm2
-; AVX1-NEXT:    vpblendvb %xmm2, %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpaddb %xmm0, %xmm0, %xmm3
-; AVX1-NEXT:    vpaddb %xmm2, %xmm2, %xmm2
-; AVX1-NEXT:    vpblendvb %xmm2, %xmm3, %xmm0, %xmm0
-; AVX1-NEXT:    vpor %xmm1, %xmm0, %xmm0
-; AVX1-NEXT:    vinsertf128 $1, %xmm4, %ymm0, %ymm0
+; AVX1-NEXT:    vpxor %xmm8, %xmm8, %xmm8
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm6 = xmm4[8],xmm8[8],xmm4[9],xmm8[9],xmm4[10],xmm8[10],xmm4[11],xmm8[11],xmm4[12],xmm8[12],xmm4[13],xmm8[13],xmm4[14],xmm8[14],xmm4[15],xmm8[15]
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm3 = xmm6[4,4,5,5,6,6,7,7]
+; AVX1-NEXT:    vpslld $23, %xmm3, %xmm7
+; AVX1-NEXT:    vmovdqa {{.*#+}} xmm9 = [1065353216,1065353216,1065353216,1065353216]
+; AVX1-NEXT:    vpaddd %xmm7, %xmm9, %xmm7
+; AVX1-NEXT:    vcvttps2dq %xmm7, %xmm7
+; AVX1-NEXT:    vpmovzxwd {{.*#+}} xmm6 = xmm6[0],zero,xmm6[1],zero,xmm6[2],zero,xmm6[3],zero
+; AVX1-NEXT:    vpslld $23, %xmm6, %xmm6
+; AVX1-NEXT:    vpaddd %xmm6, %xmm9, %xmm6
+; AVX1-NEXT:    vcvttps2dq %xmm6, %xmm6
+; AVX1-NEXT:    vpackusdw %xmm7, %xmm6, %xmm6
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm7
+; AVX1-NEXT:    vextractf128 $1, %ymm1, %xmm5
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm3 = xmm5[8],xmm7[8],xmm5[9],xmm7[9],xmm5[10],xmm7[10],xmm5[11],xmm7[11],xmm5[12],xmm7[12],xmm5[13],xmm7[13],xmm5[14],xmm7[14],xmm5[15],xmm7[15]
+; AVX1-NEXT:    vpmullw %xmm6, %xmm3, %xmm3
+; AVX1-NEXT:    vpsrlw $8, %xmm3, %xmm3
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm6 = xmm4[0],zero,zero,zero,xmm4[1],zero,zero,zero,xmm4[2],zero,zero,zero,xmm4[3],zero,zero,zero
+; AVX1-NEXT:    vpslld $23, %xmm6, %xmm6
+; AVX1-NEXT:    vpaddd %xmm6, %xmm9, %xmm6
+; AVX1-NEXT:    vcvttps2dq %xmm6, %xmm6
+; AVX1-NEXT:    vpmovzxbw {{.*#+}} xmm4 = xmm4[0],zero,xmm4[1],zero,xmm4[2],zero,xmm4[3],zero,xmm4[4],zero,xmm4[5],zero,xmm4[6],zero,xmm4[7],zero
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm4 = xmm4[4,4,5,5,6,6,7,7]
+; AVX1-NEXT:    vpslld $23, %xmm4, %xmm4
+; AVX1-NEXT:    vpaddd %xmm4, %xmm9, %xmm4
+; AVX1-NEXT:    vcvttps2dq %xmm4, %xmm4
+; AVX1-NEXT:    vpackusdw %xmm4, %xmm6, %xmm4
+; AVX1-NEXT:    vpunpcklbw {{.*#+}} xmm5 = xmm5[0],xmm7[0],xmm5[1],xmm7[1],xmm5[2],xmm7[2],xmm5[3],xmm7[3],xmm5[4],xmm7[4],xmm5[5],xmm7[5],xmm5[6],xmm7[6],xmm5[7],xmm7[7]
+; AVX1-NEXT:    vpmullw %xmm4, %xmm5, %xmm4
+; AVX1-NEXT:    vpsrlw $8, %xmm4, %xmm4
+; AVX1-NEXT:    vpackuswb %xmm3, %xmm4, %xmm3
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm4 = xmm2[8],xmm8[8],xmm2[9],xmm8[9],xmm2[10],xmm8[10],xmm2[11],xmm8[11],xmm2[12],xmm8[12],xmm2[13],xmm8[13],xmm2[14],xmm8[14],xmm2[15],xmm8[15]
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm5 = xmm4[4,4,5,5,6,6,7,7]
+; AVX1-NEXT:    vpslld $23, %xmm5, %xmm5
+; AVX1-NEXT:    vpaddd %xmm5, %xmm9, %xmm5
+; AVX1-NEXT:    vcvttps2dq %xmm5, %xmm5
+; AVX1-NEXT:    vpmovzxwd {{.*#+}} xmm4 = xmm4[0],zero,xmm4[1],zero,xmm4[2],zero,xmm4[3],zero
+; AVX1-NEXT:    vpslld $23, %xmm4, %xmm4
+; AVX1-NEXT:    vpaddd %xmm4, %xmm9, %xmm4
+; AVX1-NEXT:    vcvttps2dq %xmm4, %xmm4
+; AVX1-NEXT:    vpackusdw %xmm5, %xmm4, %xmm4
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm5 = xmm1[8],xmm0[8],xmm1[9],xmm0[9],xmm1[10],xmm0[10],xmm1[11],xmm0[11],xmm1[12],xmm0[12],xmm1[13],xmm0[13],xmm1[14],xmm0[14],xmm1[15],xmm0[15]
+; AVX1-NEXT:    vpmullw %xmm4, %xmm5, %xmm4
+; AVX1-NEXT:    vpsrlw $8, %xmm4, %xmm4
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm5 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero
+; AVX1-NEXT:    vpslld $23, %xmm5, %xmm5
+; AVX1-NEXT:    vpaddd %xmm5, %xmm9, %xmm5
+; AVX1-NEXT:    vcvttps2dq %xmm5, %xmm5
+; AVX1-NEXT:    vpmovzxbw {{.*#+}} xmm2 = xmm2[0],zero,xmm2[1],zero,xmm2[2],zero,xmm2[3],zero,xmm2[4],zero,xmm2[5],zero,xmm2[6],zero,xmm2[7],zero
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm2[4,4,5,5,6,6,7,7]
+; AVX1-NEXT:    vpslld $23, %xmm2, %xmm2
+; AVX1-NEXT:    vpaddd %xmm2, %xmm9, %xmm2
+; AVX1-NEXT:    vcvttps2dq %xmm2, %xmm2
+; AVX1-NEXT:    vpackusdw %xmm2, %xmm5, %xmm2
+; AVX1-NEXT:    vpunpcklbw {{.*#+}} xmm0 = xmm1[0],xmm0[0],xmm1[1],xmm0[1],xmm1[2],xmm0[2],xmm1[3],xmm0[3],xmm1[4],xmm0[4],xmm1[5],xmm0[5],xmm1[6],xmm0[6],xmm1[7],xmm0[7]
+; AVX1-NEXT:    vpmullw %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpsrlw $8, %xmm0, %xmm0
+; AVX1-NEXT:    vpackuswb %xmm4, %xmm0, %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm3, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
 ; AVX2-LABEL: var_funnnel_v32i8:
 ; AVX2:       # %bb.0:
-; AVX2-NEXT:    vmovdqa {{.*#+}} ymm3 = [7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7,7]
-; AVX2-NEXT:    vpandn %ymm3, %ymm2, %ymm4
-; AVX2-NEXT:    vpsllw $5, %ymm4, %ymm4
-; AVX2-NEXT:    vpsrlw $1, %ymm1, %ymm1
-; AVX2-NEXT:    vmovdqa {{.*#+}} ymm5 = [127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127,127]
-; AVX2-NEXT:    vpand %ymm5, %ymm1, %ymm1
-; AVX2-NEXT:    vpsrlw $4, %ymm1, %ymm6
-; AVX2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm6, %ymm6
-; AVX2-NEXT:    vpblendvb %ymm4, %ymm6, %ymm1, %ymm1
-; AVX2-NEXT:    vpsrlw $2, %ymm1, %ymm6
-; AVX2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm6, %ymm6
-; AVX2-NEXT:    vpaddb %ymm4, %ymm4, %ymm4
-; AVX2-NEXT:    vpblendvb %ymm4, %ymm6, %ymm1, %ymm1
-; AVX2-NEXT:    vpsrlw $1, %ymm1, %ymm6
-; AVX2-NEXT:    vpand %ymm5, %ymm6, %ymm5
-; AVX2-NEXT:    vpaddb %ymm4, %ymm4, %ymm4
-; AVX2-NEXT:    vpblendvb %ymm4, %ymm5, %ymm1, %ymm1
-; AVX2-NEXT:    vpand %ymm3, %ymm2, %ymm2
-; AVX2-NEXT:    vpsllw $5, %ymm2, %ymm2
-; AVX2-NEXT:    vpaddb %ymm2, %ymm2, %ymm3
-; AVX2-NEXT:    vpsllw $4, %ymm0, %ymm4
-; AVX2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm4, %ymm4
-; AVX2-NEXT:    vpblendvb %ymm2, %ymm4, %ymm0, %ymm0
-; AVX2-NEXT:    vpsllw $2, %ymm0, %ymm2
+; AVX2-NEXT:    vpunpckhbw {{.*#+}} ymm3 = ymm1[8],ymm0[8],ymm1[9],ymm0[9],ymm1[10],ymm0[10],ymm1[11],ymm0[11],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15],ymm1[24],ymm0[24],ymm1[25],ymm0[25],ymm1[26],ymm0[26],ymm1[27],ymm0[27],ymm1[28],ymm0[28],ymm1[29],ymm0[29],ymm1[30],ymm0[30],ymm1[31],ymm0[31]
+; AVX2-NEXT:    vpxor %xmm4, %xmm4, %xmm4
+; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm5 = ymm4[4],ymm3[4],ymm4[5],ymm3[5],ymm4[6],ymm3[6],ymm4[7],ymm3[7],ymm4[12],ymm3[12],ymm4[13],ymm3[13],ymm4[14],ymm3[14],ymm4[15],ymm3[15]
 ; AVX2-NEXT:    vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %ymm2, %ymm2
-; AVX2-NEXT:    vpblendvb %ymm3, %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vpaddb %ymm0, %ymm0, %ymm2
-; AVX2-NEXT:    vpaddb %ymm3, %ymm3, %ymm3
-; AVX2-NEXT:    vpblendvb %ymm3, %ymm2, %ymm0, %ymm0
-; AVX2-NEXT:    vpor %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpunpckhbw {{.*#+}} ymm6 = ymm2[8],ymm4[8],ymm2[9],ymm4[9],ymm2[10],ymm4[10],ymm2[11],ymm4[11],ymm2[12],ymm4[12],ymm2[13],ymm4[13],ymm2[14],ymm4[14],ymm2[15],ymm4[15],ymm2[24],ymm4[24],ymm2[25],ymm4[25],ymm2[26],ymm4[26],ymm2[27],ymm4[27],ymm2[28],ymm4[28],ymm2[29],ymm4[29],ymm2[30],ymm4[30],ymm2[31],ymm4[31]
+; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm7 = ymm6[4],ymm4[4],ymm6[5],ymm4[5],ymm6[6],ymm4[6],ymm6[7],ymm4[7],ymm6[12],ymm4[12],ymm6[13],ymm4[13],ymm6[14],ymm4[14],ymm6[15],ymm4[15]
+; AVX2-NEXT:    vpsllvd %ymm7, %ymm5, %ymm5
+; AVX2-NEXT:    vpsrld $16, %ymm5, %ymm5
+; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm3 = ymm4[0],ymm3[0],ymm4[1],ymm3[1],ymm4[2],ymm3[2],ymm4[3],ymm3[3],ymm4[8],ymm3[8],ymm4[9],ymm3[9],ymm4[10],ymm3[10],ymm4[11],ymm3[11]
+; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm6 = ymm6[0],ymm4[0],ymm6[1],ymm4[1],ymm6[2],ymm4[2],ymm6[3],ymm4[3],ymm6[8],ymm4[8],ymm6[9],ymm4[9],ymm6[10],ymm4[10],ymm6[11],ymm4[11]
+; AVX2-NEXT:    vpsllvd %ymm6, %ymm3, %ymm3
+; AVX2-NEXT:    vpsrld $16, %ymm3, %ymm3
+; AVX2-NEXT:    vpackusdw %ymm5, %ymm3, %ymm3
+; AVX2-NEXT:    vpsrlw $8, %ymm3, %ymm3
+; AVX2-NEXT:    vpunpcklbw {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[1],ymm0[1],ymm1[2],ymm0[2],ymm1[3],ymm0[3],ymm1[4],ymm0[4],ymm1[5],ymm0[5],ymm1[6],ymm0[6],ymm1[7],ymm0[7],ymm1[16],ymm0[16],ymm1[17],ymm0[17],ymm1[18],ymm0[18],ymm1[19],ymm0[19],ymm1[20],ymm0[20],ymm1[21],ymm0[21],ymm1[22],ymm0[22],ymm1[23],ymm0[23]
+; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm1 = ymm4[4],ymm0[4],ymm4[5],ymm0[5],ymm4[6],ymm0[6],ymm4[7],ymm0[7],ymm4[12],ymm0[12],ymm4[13],ymm0[13],ymm4[14],ymm0[14],ymm4[15],ymm0[15]
+; AVX2-NEXT:    vpunpcklbw {{.*#+}} ymm2 = ymm2[0],ymm4[0],ymm2[1],ymm4[1],ymm2[2],ymm4[2],ymm2[3],ymm4[3],ymm2[4],ymm4[4],ymm2[5],ymm4[5],ymm2[6],ymm4[6],ymm2[7],ymm4[7],ymm2[16],ymm4[16],ymm2[17],ymm4[17],ymm2[18],ymm4[18],ymm2[19],ymm4[19],ymm2[20],ymm4[20],ymm2[21],ymm4[21],ymm2[22],ymm4[22],ymm2[23],ymm4[23]
+; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm5 = ymm2[4],ymm4[4],ymm2[5],ymm4[5],ymm2[6],ymm4[6],ymm2[7],ymm4[7],ymm2[12],ymm4[12],ymm2[13],ymm4[13],ymm2[14],ymm4[14],ymm2[15],ymm4[15]
+; AVX2-NEXT:    vpsllvd %ymm5, %ymm1, %ymm1
+; AVX2-NEXT:    vpsrld $16, %ymm1, %ymm1
+; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm0 = ymm4[0],ymm0[0],ymm4[1],ymm0[1],ymm4[2],ymm0[2],ymm4[3],ymm0[3],ymm4[8],ymm0[8],ymm4[9],ymm0[9],ymm4[10],ymm0[10],ymm4[11],ymm0[11]
+; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm2 = ymm2[0],ymm4[0],ymm2[1],ymm4[1],ymm2[2],ymm4[2],ymm2[3],ymm4[3],ymm2[8],ymm4[8],ymm2[9],ymm4[9],ymm2[10],ymm4[10],ymm2[11],ymm4[11]
+; AVX2-NEXT:    vpsllvd %ymm2, %ymm0, %ymm0
+; AVX2-NEXT:    vpsrld $16, %ymm0, %ymm0
+; AVX2-NEXT:    vpackusdw %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpsrlw $8, %ymm0, %ymm0
+; AVX2-NEXT:    vpackuswb %ymm3, %ymm0, %ymm0
 ; AVX2-NEXT:    retq
 ;
 ; AVX512F-LABEL: var_funnnel_v32i8: