1 files changed, 19 insertions, 25 deletions
diff --git a/test/CodeGen/X86/vector-shift-lshr-256.ll b/test/CodeGen/X86/vector-shift-lshr-256.ll
index ecc68cf2e278..f9ff3092388b 100644
--- a/test/CodeGen/X86/vector-shift-lshr-256.ll
+++ b/test/CodeGen/X86/vector-shift-lshr-256.ll
@@ -155,11 +155,11 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
 ; AVX2:       # BB#0:
 ; AVX2-NEXT:    vpxor %ymm2, %ymm2, %ymm2
 ; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm3 = ymm1[4],ymm2[4],ymm1[5],ymm2[5],ymm1[6],ymm2[6],ymm1[7],ymm2[7],ymm1[12],ymm2[12],ymm1[13],ymm2[13],ymm1[14],ymm2[14],ymm1[15],ymm2[15]
-; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm4 = ymm0[4,4,5,5,6,6,7,7,12,12,13,13,14,14,15,15]
+; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm4 = ymm2[4],ymm0[4],ymm2[5],ymm0[5],ymm2[6],ymm0[6],ymm2[7],ymm0[7],ymm2[12],ymm0[12],ymm2[13],ymm0[13],ymm2[14],ymm0[14],ymm2[15],ymm0[15]
 ; AVX2-NEXT:    vpsrlvd %ymm3, %ymm4, %ymm3
 ; AVX2-NEXT:    vpsrld $16, %ymm3, %ymm3
 ; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm1 = ymm1[0],ymm2[0],ymm1[1],ymm2[1],ymm1[2],ymm2[2],ymm1[3],ymm2[3],ymm1[8],ymm2[8],ymm1[9],ymm2[9],ymm1[10],ymm2[10],ymm1[11],ymm2[11]
-; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm0 = ymm0[0,0,1,1,2,2,3,3,8,8,9,9,10,10,11,11]
+; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm0 = ymm2[0],ymm0[0],ymm2[1],ymm0[1],ymm2[2],ymm0[2],ymm2[3],ymm0[3],ymm2[8],ymm0[8],ymm2[9],ymm0[9],ymm2[10],ymm0[10],ymm2[11],ymm0[11]
 ; AVX2-NEXT:    vpsrlvd %ymm1, %ymm0, %ymm0
 ; AVX2-NEXT:    vpsrld $16, %ymm0, %ymm0
 ; AVX2-NEXT:    vpackusdw %ymm3, %ymm0, %ymm0
@@ -191,7 +191,10 @@ define <16 x i16> @var_shift_v16i16(<16 x i16> %a, <16 x i16> %b) nounwind {
 ;
 ; AVX512-LABEL: var_shift_v16i16:
 ; AVX512:       ## BB#0:
+; AVX512-NEXT:    ## kill: %YMM1<def> %YMM1<kill> %ZMM1<def>
+; AVX512-NEXT:    ## kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
 ; AVX512-NEXT:    vpsrlvw %zmm1, %zmm0, %zmm0
+; AVX512-NEXT:    ## kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
 ; AVX512-NEXT:    retq
   %shift = lshr <16 x i16> %a, %b
   ret <16 x i16> %shift
@@ -369,7 +372,7 @@ define <8 x i32> @splatvar_shift_v8i32(<8 x i32> %a, <8 x i32> %b) nounwind {
 ; AVX512-LABEL: splatvar_shift_v8i32:
 ; AVX512:       ## BB#0:
 ; AVX512-NEXT:    vxorps %xmm2, %xmm2, %xmm2
-; AVX512-NEXT:    vmovss %xmm1, %xmm2, %xmm1
+; AVX512-NEXT:    vmovss {{.*#+}} xmm1 = xmm1[0],xmm2[1,2,3]
 ; AVX512-NEXT:    vpsrld %xmm1, %ymm0, %ymm0
 ; AVX512-NEXT:    retq
   %splat = shufflevector <8 x i32> %b, <8 x i32> undef, <8 x i32> zeroinitializer
@@ -618,30 +621,19 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) nounwind {
 ; AVX1-LABEL: constant_shift_v16i16:
 ; AVX1:       # BB#0:
 ; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm1
-; AVX1-NEXT:    vpsrlw $8, %xmm1, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [32896,37008,41120,45232,49344,53456,57568,61680]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpsrlw $8, %xmm1, %xmm1
 ; AVX1-NEXT:    vpsrlw $4, %xmm1, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [256,8480,16704,24928,33152,41376,49600,57824]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0,1,2,3],xmm2[4,5,6,7]
 ; AVX1-NEXT:    vpsrlw $2, %xmm1, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [512,16960,33408,49856,768,17216,33664,50112]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm1, %xmm1
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0,1],xmm2[2,3],xmm1[4,5],xmm2[6,7]
 ; AVX1-NEXT:    vpsrlw $1, %xmm1, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [1024,33920,1280,34176,1536,34432,1792,34688]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm1, %xmm1
-; AVX1-NEXT:    vpsrlw $8, %xmm0, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [0,4112,8224,12336,16448,20560,24672,28784]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm1 = xmm1[0],xmm2[1],xmm1[2],xmm2[3],xmm1[4],xmm2[5],xmm1[6],xmm2[7]
 ; AVX1-NEXT:    vpsrlw $4, %xmm0, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [0,8224,16448,24672,32896,41120,49344,57568]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1,2,3],xmm2[4,5,6,7]
 ; AVX1-NEXT:    vpsrlw $2, %xmm0, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [0,16448,32896,49344,256,16704,33152,49600]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0,1],xmm2[2,3],xmm0[4,5],xmm2[6,7]
 ; AVX1-NEXT:    vpsrlw $1, %xmm0, %xmm2
-; AVX1-NEXT:    vmovdqa {{.*#+}} xmm3 = [0,32896,256,33152,512,33408,768,33664]
-; AVX1-NEXT:    vpblendvb %xmm3, %xmm2, %xmm0, %xmm0
+; AVX1-NEXT:    vpblendw {{.*#+}} xmm0 = xmm0[0],xmm2[1],xmm0[2],xmm2[3],xmm0[4],xmm2[5],xmm0[6],xmm2[7]
 ; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
 ; AVX1-NEXT:    retq
 ;
@@ -650,12 +642,12 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) nounwind {
 ; AVX2-NEXT:    vpxor %ymm1, %ymm1, %ymm1
 ; AVX2-NEXT:    vmovdqa {{.*#+}} ymm2 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
 ; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm3 = ymm2[4],ymm1[4],ymm2[5],ymm1[5],ymm2[6],ymm1[6],ymm2[7],ymm1[7],ymm2[12],ymm1[12],ymm2[13],ymm1[13],ymm2[14],ymm1[14],ymm2[15],ymm1[15]
-; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm4 = ymm0[4,4,5,5,6,6,7,7,12,12,13,13,14,14,15,15]
+; AVX2-NEXT:    vpunpckhwd {{.*#+}} ymm4 = ymm1[4],ymm0[4],ymm1[5],ymm0[5],ymm1[6],ymm0[6],ymm1[7],ymm0[7],ymm1[12],ymm0[12],ymm1[13],ymm0[13],ymm1[14],ymm0[14],ymm1[15],ymm0[15]
 ; AVX2-NEXT:    vpsrlvd %ymm3, %ymm4, %ymm3
 ; AVX2-NEXT:    vpsrld $16, %ymm3, %ymm3
-; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm1 = ymm2[0],ymm1[0],ymm2[1],ymm1[1],ymm2[2],ymm1[2],ymm2[3],ymm1[3],ymm2[8],ymm1[8],ymm2[9],ymm1[9],ymm2[10],ymm1[10],ymm2[11],ymm1[11]
-; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm0 = ymm0[0,0,1,1,2,2,3,3,8,8,9,9,10,10,11,11]
-; AVX2-NEXT:    vpsrlvd %ymm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm2 = ymm2[0],ymm1[0],ymm2[1],ymm1[1],ymm2[2],ymm1[2],ymm2[3],ymm1[3],ymm2[8],ymm1[8],ymm2[9],ymm1[9],ymm2[10],ymm1[10],ymm2[11],ymm1[11]
+; AVX2-NEXT:    vpunpcklwd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[1],ymm0[1],ymm1[2],ymm0[2],ymm1[3],ymm0[3],ymm1[8],ymm0[8],ymm1[9],ymm0[9],ymm1[10],ymm0[10],ymm1[11],ymm0[11]
+; AVX2-NEXT:    vpsrlvd %ymm2, %ymm0, %ymm0
 ; AVX2-NEXT:    vpsrld $16, %ymm0, %ymm0
 ; AVX2-NEXT:    vpackusdw %ymm3, %ymm0, %ymm0
 ; AVX2-NEXT:    retq
@@ -684,8 +676,10 @@ define <16 x i16> @constant_shift_v16i16(<16 x i16> %a) nounwind {
 ;
 ; AVX512-LABEL: constant_shift_v16i16:
 ; AVX512:       ## BB#0:
+; AVX512-NEXT:    ## kill: %YMM0<def> %YMM0<kill> %ZMM0<def>
 ; AVX512-NEXT:    vmovdqa {{.*#+}} ymm1 = [0,1,2,3,4,5,6,7,8,9,10,11,12,13,14,15]
 ; AVX512-NEXT:    vpsrlvw %zmm1, %zmm0, %zmm0
+; AVX512-NEXT:    ## kill: %YMM0<def> %YMM0<kill> %ZMM0<kill>
 ; AVX512-NEXT:    retq
   %shift = lshr <16 x i16> %a, <i16 0, i16 1, i16 2, i16 3, i16 4, i16 5, i16 6, i16 7, i16 8, i16 9, i16 10, i16 11, i16 12, i16 13, i16 14, i16 15>
   ret <16 x i16> %shift