[llvm] r277956 - [X86][AVX512] Add sext/zext to 512-bit vector tests

Simon Pilgrim via llvm-commits llvm-commits at lists.llvm.org
Sun Aug 7 05:10:48 PDT 2016


Author: rksimon
Date: Sun Aug  7 07:10:46 2016
New Revision: 277956

URL: http://llvm.org/viewvc/llvm-project?rev=277956&view=rev
Log:
[X86][AVX512] Add sext/zext to 512-bit vector tests

Modified:
    llvm/trunk/test/CodeGen/X86/vector-sext.ll
    llvm/trunk/test/CodeGen/X86/vector-zext.ll

Modified: llvm/trunk/test/CodeGen/X86/vector-sext.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/vector-sext.ll?rev=277956&r1=277955&r2=277956&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/X86/vector-sext.ll (original)
+++ llvm/trunk/test/CodeGen/X86/vector-sext.ll Sun Aug  7 07:10:46 2016
@@ -99,6 +99,93 @@ entry:
   ret <16 x i16> %B
 }
 
+define <32 x i16> @sext_32i8_to_32i16(<32 x i8> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: sext_32i8_to_32i16:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1],xmm4[2],xmm0[2],xmm4[3],xmm0[3],xmm4[4],xmm0[4],xmm4[5],xmm0[5],xmm4[6],xmm0[6],xmm4[7],xmm0[7]
+; SSE2-NEXT:    psraw $8, %xmm4
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm5 = xmm5[8],xmm0[8],xmm5[9],xmm0[9],xmm5[10],xmm0[10],xmm5[11],xmm0[11],xmm5[12],xmm0[12],xmm5[13],xmm0[13],xmm5[14],xmm0[14],xmm5[15],xmm0[15]
+; SSE2-NEXT:    psraw $8, %xmm5
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3],xmm2[4],xmm1[4],xmm2[5],xmm1[5],xmm2[6],xmm1[6],xmm2[7],xmm1[7]
+; SSE2-NEXT:    psraw $8, %xmm2
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm1[8],xmm3[9],xmm1[9],xmm3[10],xmm1[10],xmm3[11],xmm1[11],xmm3[12],xmm1[12],xmm3[13],xmm1[13],xmm3[14],xmm1[14],xmm3[15],xmm1[15]
+; SSE2-NEXT:    psraw $8, %xmm3
+; SSE2-NEXT:    movdqa %xmm4, %xmm0
+; SSE2-NEXT:    movdqa %xmm5, %xmm1
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: sext_32i8_to_32i16:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1],xmm4[2],xmm0[2],xmm4[3],xmm0[3],xmm4[4],xmm0[4],xmm4[5],xmm0[5],xmm4[6],xmm0[6],xmm4[7],xmm0[7]
+; SSSE3-NEXT:    psraw $8, %xmm4
+; SSSE3-NEXT:    punpckhbw {{.*#+}} xmm5 = xmm5[8],xmm0[8],xmm5[9],xmm0[9],xmm5[10],xmm0[10],xmm5[11],xmm0[11],xmm5[12],xmm0[12],xmm5[13],xmm0[13],xmm5[14],xmm0[14],xmm5[15],xmm0[15]
+; SSSE3-NEXT:    psraw $8, %xmm5
+; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3],xmm2[4],xmm1[4],xmm2[5],xmm1[5],xmm2[6],xmm1[6],xmm2[7],xmm1[7]
+; SSSE3-NEXT:    psraw $8, %xmm2
+; SSSE3-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm1[8],xmm3[9],xmm1[9],xmm3[10],xmm1[10],xmm3[11],xmm1[11],xmm3[12],xmm1[12],xmm3[13],xmm1[13],xmm3[14],xmm1[14],xmm3[15],xmm1[15]
+; SSSE3-NEXT:    psraw $8, %xmm3
+; SSSE3-NEXT:    movdqa %xmm4, %xmm0
+; SSSE3-NEXT:    movdqa %xmm5, %xmm1
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: sext_32i8_to_32i16:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovsxbw %xmm0, %xmm5
+; SSE41-NEXT:    pmovsxbw %xmm1, %xmm2
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; SSE41-NEXT:    pmovsxbw %xmm0, %xmm4
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,3,0,1]
+; SSE41-NEXT:    pmovsxbw %xmm0, %xmm3
+; SSE41-NEXT:    movdqa %xmm5, %xmm0
+; SSE41-NEXT:    movdqa %xmm4, %xmm1
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: sext_32i8_to_32i16:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovsxbw %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxbw %xmm2, %xmm2
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpmovsxbw %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxbw %xmm0, %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: sext_32i8_to_32i16:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovsxbw %xmm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpmovsxbw %xmm0, %ymm1
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: sext_32i8_to_32i16:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovsxbw %xmm0, %ymm2
+; AVX512-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX512-NEXT:    vpmovsxbw %xmm0, %ymm1
+; AVX512-NEXT:    vmovdqa64 %zmm2, %zmm0
+; AVX512-NEXT:    retq
+;
+; X32-SSE41-LABEL: sext_32i8_to_32i16:
+; X32-SSE41:       # BB#0: # %entry
+; X32-SSE41-NEXT:    pmovsxbw %xmm0, %xmm5
+; X32-SSE41-NEXT:    pmovsxbw %xmm1, %xmm2
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxbw %xmm0, %xmm4
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxbw %xmm0, %xmm3
+; X32-SSE41-NEXT:    movdqa %xmm5, %xmm0
+; X32-SSE41-NEXT:    movdqa %xmm4, %xmm1
+; X32-SSE41-NEXT:    retl
+entry:
+  %B = sext <32 x i8> %A to <32 x i16>
+  ret <32 x i16> %B
+}
+
 define <4 x i32> @sext_16i8_to_4i32(<16 x i8> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: sext_16i8_to_4i32:
 ; SSE2:       # BB#0: # %entry
@@ -196,6 +283,97 @@ entry:
   ret <8 x i32> %C
 }
 
+define <16 x i32> @sext_16i8_to_16i32(<16 x i8> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: sext_16i8_to_16i32:
+; SSE2:       # BB#0: # %entry
+; 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:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm1[0],xmm4[1],xmm1[1],xmm4[2],xmm1[2],xmm4[3],xmm1[3]
+; SSE2-NEXT:    psrad $24, %xmm4
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm1 = 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]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3]
+; SSE2-NEXT:    psrad $24, %xmm2
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3]
+; SSE2-NEXT:    psrad $24, %xmm1
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3,4,4,5,5,6,6,7,7]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm0[0],xmm3[1],xmm0[1],xmm3[2],xmm0[2],xmm3[3],xmm0[3]
+; SSE2-NEXT:    psrad $24, %xmm3
+; SSE2-NEXT:    movdqa %xmm4, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: sext_16i8_to_16i32:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm0, %xmm3
+; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1],xmm0[2],xmm3[2],xmm0[3],xmm3[3],xmm0[4],xmm3[4],xmm0[5],xmm3[5],xmm0[6],xmm3[6],xmm0[7],xmm3[7]
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0,0,1,1,2,2,3,3]
+; SSSE3-NEXT:    psrad $24, %xmm0
+; SSSE3-NEXT:    punpckhbw {{.*#+}} xmm1 = xmm1[8],xmm3[8],xmm1[9],xmm3[9],xmm1[10],xmm3[10],xmm1[11],xmm3[11],xmm1[12],xmm3[12],xmm1[13],xmm3[13],xmm1[14],xmm3[14],xmm1[15],xmm3[15]
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3]
+; SSSE3-NEXT:    psrad $24, %xmm2
+; SSSE3-NEXT:    movdqa %xmm3, %xmm1
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm1 = xmm1[u,u,u,4,u,u,u,5,u,u,u,6,u,u,u,7]
+; SSSE3-NEXT:    psrad $24, %xmm1
+; SSSE3-NEXT:    pshufb {{.*#+}} xmm3 = xmm3[u,u,u,12,u,u,u,13,u,u,u,14,u,u,u,15]
+; SSSE3-NEXT:    psrad $24, %xmm3
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: sext_16i8_to_16i32:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovsxbd %xmm0, %xmm4
+; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; SSE41-NEXT:    pmovsxbd %xmm1, %xmm1
+; SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; SSE41-NEXT:    pmovsxbd %xmm2, %xmm2
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; SSE41-NEXT:    pmovsxbd %xmm0, %xmm3
+; SSE41-NEXT:    movdqa %xmm4, %xmm0
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: sext_16i8_to_16i32:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovsxbd %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
+; AVX1-NEXT:    vpmovsxbd %xmm2, %xmm2
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxbd %xmm1, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; AVX1-NEXT:    vpmovsxbd %xmm0, %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: sext_16i8_to_16i32:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovsxbd %xmm0, %ymm2
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; AVX2-NEXT:    vpmovsxbd %xmm0, %ymm1
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: sext_16i8_to_16i32:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovsxbd %xmm0, %zmm0
+; AVX512-NEXT:    retq
+;
+; X32-SSE41-LABEL: sext_16i8_to_16i32:
+; X32-SSE41:       # BB#0: # %entry
+; X32-SSE41-NEXT:    pmovsxbd %xmm0, %xmm4
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; X32-SSE41-NEXT:    pmovsxbd %xmm1, %xmm1
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxbd %xmm2, %xmm2
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; X32-SSE41-NEXT:    pmovsxbd %xmm0, %xmm3
+; X32-SSE41-NEXT:    movdqa %xmm4, %xmm0
+; X32-SSE41-NEXT:    retl
+entry:
+  %B = sext <16 x i8> %A to <16 x i32>
+  ret <16 x i32> %B
+}
+
 define <2 x i64> @sext_16i8_to_2i64(<16 x i8> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: sext_16i8_to_2i64:
 ; SSE2:       # BB#0: # %entry
@@ -528,6 +706,90 @@ entry:
   ret <8 x i32> %B
 }
 
+define <16 x i32> @sext_16i16_to_16i32(<16 x i16> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: sext_16i16_to_16i32:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1],xmm4[2],xmm0[2],xmm4[3],xmm0[3]
+; SSE2-NEXT:    psrad $16, %xmm4
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm5 = xmm5[4],xmm0[4],xmm5[5],xmm0[5],xmm5[6],xmm0[6],xmm5[7],xmm0[7]
+; SSE2-NEXT:    psrad $16, %xmm5
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3]
+; SSE2-NEXT:    psrad $16, %xmm2
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm1[4],xmm3[5],xmm1[5],xmm3[6],xmm1[6],xmm3[7],xmm1[7]
+; SSE2-NEXT:    psrad $16, %xmm3
+; SSE2-NEXT:    movdqa %xmm4, %xmm0
+; SSE2-NEXT:    movdqa %xmm5, %xmm1
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: sext_16i16_to_16i32:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1],xmm4[2],xmm0[2],xmm4[3],xmm0[3]
+; SSSE3-NEXT:    psrad $16, %xmm4
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm5 = xmm5[4],xmm0[4],xmm5[5],xmm0[5],xmm5[6],xmm0[6],xmm5[7],xmm0[7]
+; SSSE3-NEXT:    psrad $16, %xmm5
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1],xmm2[2],xmm1[2],xmm2[3],xmm1[3]
+; SSSE3-NEXT:    psrad $16, %xmm2
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm1[4],xmm3[5],xmm1[5],xmm3[6],xmm1[6],xmm3[7],xmm1[7]
+; SSSE3-NEXT:    psrad $16, %xmm3
+; SSSE3-NEXT:    movdqa %xmm4, %xmm0
+; SSSE3-NEXT:    movdqa %xmm5, %xmm1
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: sext_16i16_to_16i32:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovsxwd %xmm0, %xmm5
+; SSE41-NEXT:    pmovsxwd %xmm1, %xmm2
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; SSE41-NEXT:    pmovsxwd %xmm0, %xmm4
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,3,0,1]
+; SSE41-NEXT:    pmovsxwd %xmm0, %xmm3
+; SSE41-NEXT:    movdqa %xmm5, %xmm0
+; SSE41-NEXT:    movdqa %xmm4, %xmm1
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: sext_16i16_to_16i32:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovsxwd %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxwd %xmm2, %xmm2
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpmovsxwd %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxwd %xmm0, %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: sext_16i16_to_16i32:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovsxwd %xmm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpmovsxwd %xmm0, %ymm1
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: sext_16i16_to_16i32:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovsxwd %ymm0, %zmm0
+; AVX512-NEXT:    retq
+;
+; X32-SSE41-LABEL: sext_16i16_to_16i32:
+; X32-SSE41:       # BB#0: # %entry
+; X32-SSE41-NEXT:    pmovsxwd %xmm0, %xmm5
+; X32-SSE41-NEXT:    pmovsxwd %xmm1, %xmm2
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxwd %xmm0, %xmm4
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxwd %xmm0, %xmm3
+; X32-SSE41-NEXT:    movdqa %xmm5, %xmm0
+; X32-SSE41-NEXT:    movdqa %xmm4, %xmm1
+; X32-SSE41-NEXT:    retl
+entry:
+  %B = sext <16 x i16> %A to <16 x i32>
+  ret <16 x i32> %B
+}
+
 define <2 x i64> @sext_8i16_to_2i64(<8 x i16> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: sext_8i16_to_2i64:
 ; SSE2:       # BB#0: # %entry
@@ -639,6 +901,116 @@ entry:
   ret <4 x i64> %C
 }
 
+define <8 x i64> @sext_8i16_to_8i64(<8 x i16> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: sext_8i16_to_8i64:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1],xmm4[2],xmm0[2],xmm4[3],xmm0[3]
+; SSE2-NEXT:    movdqa %xmm4, %xmm1
+; SSE2-NEXT:    psrad $31, %xmm1
+; SSE2-NEXT:    psrad $16, %xmm4
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm4 = xmm4[0],xmm1[0],xmm4[1],xmm1[1]
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm2 = xmm2[4],xmm0[4],xmm2[5],xmm0[5],xmm2[6],xmm0[6],xmm2[7],xmm0[7]
+; SSE2-NEXT:    movdqa %xmm2, %xmm1
+; SSE2-NEXT:    psrad $31, %xmm1
+; SSE2-NEXT:    psrad $16, %xmm2
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3]
+; SSE2-NEXT:    movdqa %xmm1, %xmm3
+; SSE2-NEXT:    psrad $31, %xmm3
+; SSE2-NEXT:    psrad $16, %xmm1
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm0[0],xmm3[1],xmm0[1],xmm3[2],xmm0[2],xmm3[3],xmm0[3]
+; SSE2-NEXT:    movdqa %xmm3, %xmm0
+; SSE2-NEXT:    psrad $31, %xmm0
+; SSE2-NEXT:    psrad $16, %xmm3
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm3 = xmm3[0],xmm0[0],xmm3[1],xmm0[1]
+; SSE2-NEXT:    movdqa %xmm4, %xmm0
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: sext_8i16_to_8i64:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm4 = xmm4[0],xmm0[0],xmm4[1],xmm0[1],xmm4[2],xmm0[2],xmm4[3],xmm0[3]
+; SSSE3-NEXT:    movdqa %xmm4, %xmm1
+; SSSE3-NEXT:    psrad $31, %xmm1
+; SSSE3-NEXT:    psrad $16, %xmm4
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm4 = xmm4[0],xmm1[0],xmm4[1],xmm1[1]
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm2 = xmm2[4],xmm0[4],xmm2[5],xmm0[5],xmm2[6],xmm0[6],xmm2[7],xmm0[7]
+; SSSE3-NEXT:    movdqa %xmm2, %xmm1
+; SSSE3-NEXT:    psrad $31, %xmm1
+; SSSE3-NEXT:    psrad $16, %xmm2
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm1[0],xmm2[1],xmm1[1]
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm1 = xmm1[0,0,1,1,2,2,3,3]
+; SSSE3-NEXT:    movdqa %xmm1, %xmm3
+; SSSE3-NEXT:    psrad $31, %xmm3
+; SSSE3-NEXT:    psrad $16, %xmm1
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1]
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm0[0],xmm3[1],xmm0[1],xmm3[2],xmm0[2],xmm3[3],xmm0[3]
+; SSSE3-NEXT:    movdqa %xmm3, %xmm0
+; SSSE3-NEXT:    psrad $31, %xmm0
+; SSSE3-NEXT:    psrad $16, %xmm3
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm3 = xmm3[0],xmm0[0],xmm3[1],xmm0[1]
+; SSSE3-NEXT:    movdqa %xmm4, %xmm0
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: sext_8i16_to_8i64:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovsxwq %xmm0, %xmm4
+; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; SSE41-NEXT:    pmovsxwq %xmm1, %xmm1
+; SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; SSE41-NEXT:    pmovsxwq %xmm2, %xmm2
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; SSE41-NEXT:    pmovsxwq %xmm0, %xmm3
+; SSE41-NEXT:    movdqa %xmm4, %xmm0
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: sext_8i16_to_8i64:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovsxwq %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
+; AVX1-NEXT:    vpmovsxwq %xmm2, %xmm2
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxwq %xmm1, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; AVX1-NEXT:    vpmovsxwq %xmm0, %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: sext_8i16_to_8i64:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovsxwq %xmm0, %ymm2
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; AVX2-NEXT:    vpmovsxwq %xmm0, %ymm1
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: sext_8i16_to_8i64:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovsxwq %xmm0, %zmm0
+; AVX512-NEXT:    retq
+;
+; X32-SSE41-LABEL: sext_8i16_to_8i64:
+; X32-SSE41:       # BB#0: # %entry
+; X32-SSE41-NEXT:    pmovsxwq %xmm0, %xmm4
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; X32-SSE41-NEXT:    pmovsxwq %xmm1, %xmm1
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxwq %xmm2, %xmm2
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; X32-SSE41-NEXT:    pmovsxwq %xmm0, %xmm3
+; X32-SSE41-NEXT:    movdqa %xmm4, %xmm0
+; X32-SSE41-NEXT:    retl
+entry:
+  %B = sext <8 x i16> %A to <8 x i64>
+  ret <8 x i64> %B
+}
+
 define <2 x i64> @sext_4i32_to_2i64(<4 x i32> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: sext_4i32_to_2i64:
 ; SSE2:       # BB#0: # %entry
@@ -735,6 +1107,100 @@ entry:
   ret <4 x i64> %B
 }
 
+define <8 x i64> @sext_8i32_to_8i64(<8 x i32> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: sext_8i32_to_8i64:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movdqa %xmm1, %xmm2
+; SSE2-NEXT:    movdqa %xmm0, %xmm3
+; SSE2-NEXT:    psrad $31, %xmm3
+; SSE2-NEXT:    movdqa %xmm2, %xmm4
+; SSE2-NEXT:    psrad $31, %xmm4
+; SSE2-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1]
+; SSE2-NEXT:    movdqa %xmm1, %xmm3
+; SSE2-NEXT:    psrad $31, %xmm3
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1]
+; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm2[2,3,0,1]
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1]
+; SSE2-NEXT:    movdqa %xmm3, %xmm4
+; SSE2-NEXT:    psrad $31, %xmm4
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: sext_8i32_to_8i64:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm1, %xmm2
+; SSSE3-NEXT:    movdqa %xmm0, %xmm3
+; SSSE3-NEXT:    psrad $31, %xmm3
+; SSSE3-NEXT:    movdqa %xmm2, %xmm4
+; SSSE3-NEXT:    psrad $31, %xmm4
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm3[0],xmm0[1],xmm3[1]
+; SSSE3-NEXT:    movdqa %xmm1, %xmm3
+; SSSE3-NEXT:    psrad $31, %xmm3
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm1 = xmm1[0],xmm3[0],xmm1[1],xmm3[1]
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm3 = xmm2[2,3,0,1]
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1]
+; SSSE3-NEXT:    movdqa %xmm3, %xmm4
+; SSSE3-NEXT:    psrad $31, %xmm4
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: sext_8i32_to_8i64:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovsxdq %xmm0, %xmm5
+; SSE41-NEXT:    pmovsxdq %xmm1, %xmm2
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; SSE41-NEXT:    pmovsxdq %xmm0, %xmm4
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,3,0,1]
+; SSE41-NEXT:    pmovsxdq %xmm0, %xmm3
+; SSE41-NEXT:    movdqa %xmm5, %xmm0
+; SSE41-NEXT:    movdqa %xmm4, %xmm1
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: sext_8i32_to_8i64:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovsxdq %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxdq %xmm2, %xmm2
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpmovsxdq %xmm0, %xmm1
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovsxdq %xmm0, %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: sext_8i32_to_8i64:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovsxdq %xmm0, %ymm2
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpmovsxdq %xmm0, %ymm1
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: sext_8i32_to_8i64:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovsxdq %ymm0, %zmm0
+; AVX512-NEXT:    retq
+;
+; X32-SSE41-LABEL: sext_8i32_to_8i64:
+; X32-SSE41:       # BB#0: # %entry
+; X32-SSE41-NEXT:    pmovsxdq %xmm0, %xmm5
+; X32-SSE41-NEXT:    pmovsxdq %xmm1, %xmm2
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxdq %xmm0, %xmm4
+; X32-SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm1[2,3,0,1]
+; X32-SSE41-NEXT:    pmovsxdq %xmm0, %xmm3
+; X32-SSE41-NEXT:    movdqa %xmm5, %xmm0
+; X32-SSE41-NEXT:    movdqa %xmm4, %xmm1
+; X32-SSE41-NEXT:    retl
+entry:
+  %B = sext <8 x i32> %A to <8 x i64>
+  ret <8 x i64> %B
+}
+
 define <2 x i64> @load_sext_2i1_to_2i64(<2 x i1> *%ptr) {
 ; SSE-LABEL: load_sext_2i1_to_2i64:
 ; SSE:       # BB#0: # %entry

Modified: llvm/trunk/test/CodeGen/X86/vector-zext.ll
URL: http://llvm.org/viewvc/llvm-project/llvm/trunk/test/CodeGen/X86/vector-zext.ll?rev=277956&r1=277955&r2=277956&view=diff
==============================================================================
--- llvm/trunk/test/CodeGen/X86/vector-zext.ll (original)
+++ llvm/trunk/test/CodeGen/X86/vector-zext.ll Sun Aug  7 07:10:46 2016
@@ -82,6 +82,75 @@ entry:
   ret <16 x i16> %B
 }
 
+define <32 x i16> @zext_32i8_to_32i16(<32 x i8> %A) {
+; SSE2-LABEL: zext_32i8_to_32i16:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movdqa %xmm1, %xmm3
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    pxor %xmm4, %xmm4
+; 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:    punpckhbw {{.*#+}} xmm1 = xmm1[8],xmm4[8],xmm1[9],xmm4[9],xmm1[10],xmm4[10],xmm1[11],xmm4[11],xmm1[12],xmm4[12],xmm1[13],xmm4[13],xmm1[14],xmm4[14],xmm1[15],xmm4[15]
+; SSE2-NEXT:    movdqa %xmm3, %xmm2
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1],xmm2[2],xmm4[2],xmm2[3],xmm4[3],xmm2[4],xmm4[4],xmm2[5],xmm4[5],xmm2[6],xmm4[6],xmm2[7],xmm4[7]
+; SSE2-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm4[8],xmm3[9],xmm4[9],xmm3[10],xmm4[10],xmm3[11],xmm4[11],xmm3[12],xmm4[12],xmm3[13],xmm4[13],xmm3[14],xmm4[14],xmm3[15],xmm4[15]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: zext_32i8_to_32i16:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm1, %xmm3
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    pxor %xmm4, %xmm4
+; SSSE3-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]
+; SSSE3-NEXT:    punpckhbw {{.*#+}} xmm1 = xmm1[8],xmm4[8],xmm1[9],xmm4[9],xmm1[10],xmm4[10],xmm1[11],xmm4[11],xmm1[12],xmm4[12],xmm1[13],xmm4[13],xmm1[14],xmm4[14],xmm1[15],xmm4[15]
+; SSSE3-NEXT:    movdqa %xmm3, %xmm2
+; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1],xmm2[2],xmm4[2],xmm2[3],xmm4[3],xmm2[4],xmm4[4],xmm2[5],xmm4[5],xmm2[6],xmm4[6],xmm2[7],xmm4[7]
+; SSSE3-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm4[8],xmm3[9],xmm4[9],xmm3[10],xmm4[10],xmm3[11],xmm4[11],xmm3[12],xmm4[12],xmm3[13],xmm4[13],xmm3[14],xmm4[14],xmm3[15],xmm4[15]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: zext_32i8_to_32i16:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    movdqa %xmm1, %xmm3
+; SSE41-NEXT:    movdqa %xmm0, %xmm1
+; SSE41-NEXT:    pxor %xmm4, %xmm4
+; SSE41-NEXT:    pmovzxbw {{.*#+}} xmm0 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero,xmm1[4],zero,xmm1[5],zero,xmm1[6],zero,xmm1[7],zero
+; SSE41-NEXT:    punpckhbw {{.*#+}} xmm1 = xmm1[8],xmm4[8],xmm1[9],xmm4[9],xmm1[10],xmm4[10],xmm1[11],xmm4[11],xmm1[12],xmm4[12],xmm1[13],xmm4[13],xmm1[14],xmm4[14],xmm1[15],xmm4[15]
+; SSE41-NEXT:    pmovzxbw {{.*#+}} xmm2 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero,xmm3[4],zero,xmm3[5],zero,xmm3[6],zero,xmm3[7],zero
+; SSE41-NEXT:    punpckhbw {{.*#+}} xmm3 = xmm3[8],xmm4[8],xmm3[9],xmm4[9],xmm3[10],xmm4[10],xmm3[11],xmm4[11],xmm3[12],xmm4[12],xmm3[13],xmm4[13],xmm3[14],xmm4[14],xmm3[15],xmm4[15]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: zext_32i8_to_32i16:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm2 = xmm0[8],xmm1[8],xmm0[9],xmm1[9],xmm0[10],xmm1[10],xmm0[11],xmm1[11],xmm0[12],xmm1[12],xmm0[13],xmm1[13],xmm0[14],xmm1[14],xmm0[15],xmm1[15]
+; AVX1-NEXT:    vpmovzxbw {{.*#+}} xmm3 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm3, %ymm2
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpunpckhbw {{.*#+}} xmm1 = xmm0[8],xmm1[8],xmm0[9],xmm1[9],xmm0[10],xmm1[10],xmm0[11],xmm1[11],xmm0[12],xmm1[12],xmm0[13],xmm1[13],xmm0[14],xmm1[14],xmm0[15],xmm1[15]
+; AVX1-NEXT:    vpmovzxbw {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: zext_32i8_to_32i16:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: zext_32i8_to_32i16:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovzxbw {{.*#+}} ymm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
+; AVX512-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX512-NEXT:    vpmovzxbw {{.*#+}} ymm1 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero,xmm0[4],zero,xmm0[5],zero,xmm0[6],zero,xmm0[7],zero,xmm0[8],zero,xmm0[9],zero,xmm0[10],zero,xmm0[11],zero,xmm0[12],zero,xmm0[13],zero,xmm0[14],zero,xmm0[15],zero
+; AVX512-NEXT:    vmovdqa64 %zmm2, %zmm0
+; AVX512-NEXT:    retq
+entry:
+  %B = zext <32 x i8> %A to <32 x i16>
+  ret <32 x i16> %B
+}
+
 define <4 x i32> @zext_16i8_to_4i32(<16 x i8> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: zext_16i8_to_4i32:
 ; SSE2:       # BB#0: # %entry
@@ -164,6 +233,80 @@ entry:
   ret <8 x i32> %C
 }
 
+define <16 x i32> @zext_16i8_to_16i32(<16 x i8> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: zext_16i8_to_16i32:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    pxor %xmm4, %xmm4
+; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm1[2,3,0,1]
+; 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:    movdqa %xmm1, %xmm0
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1],xmm0[2],xmm4[2],xmm0[3],xmm4[3]
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3],xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSE2-NEXT:    movdqa %xmm3, %xmm2
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1],xmm2[2],xmm4[2],xmm2[3],xmm4[3]
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: zext_16i8_to_16i32:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    pxor %xmm4, %xmm4
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm3 = xmm1[2,3,0,1]
+; SSSE3-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]
+; SSSE3-NEXT:    movdqa %xmm1, %xmm0
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1],xmm0[2],xmm4[2],xmm0[3],xmm4[3]
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
+; SSSE3-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3],xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSSE3-NEXT:    movdqa %xmm3, %xmm2
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1],xmm2[2],xmm4[2],xmm2[3],xmm4[3]
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: zext_16i8_to_16i32:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovzxbd {{.*#+}} xmm4 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; SSE41-NEXT:    pmovzxbd {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero
+; SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; SSE41-NEXT:    pmovzxbd {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; SSE41-NEXT:    pmovzxbd {{.*#+}} xmm3 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; SSE41-NEXT:    movdqa %xmm4, %xmm0
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: zext_16i8_to_16i32:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero,xmm1[2],zero,zero,zero,xmm1[3],zero,zero,zero
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; AVX1-NEXT:    vpmovzxbd {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: zext_16i8_to_16i32:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm2 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; AVX2-NEXT:    vpmovzxbd {{.*#+}} ymm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: zext_16i8_to_16i32:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovzxbd {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero,xmm0[8],zero,zero,zero,xmm0[9],zero,zero,zero,xmm0[10],zero,zero,zero,xmm0[11],zero,zero,zero,xmm0[12],zero,zero,zero,xmm0[13],zero,zero,zero,xmm0[14],zero,zero,zero,xmm0[15],zero,zero,zero
+; AVX512-NEXT:    retq
+entry:
+  %B = zext <16 x i8> %A to <16 x i32>
+  ret <16 x i32> %B
+}
+
 define <2 x i64> @zext_16i8_to_2i64(<16 x i8> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: zext_16i8_to_2i64:
 ; SSE2:       # BB#0: # %entry
@@ -243,6 +386,83 @@ entry:
   ret <4 x i64> %C
 }
 
+define <8 x i64> @zext_16i8_to_8i64(<16 x i8> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: zext_16i8_to_8i64:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    pxor %xmm4, %xmm4
+; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm1[1,1,2,3]
+; 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:    punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSE2-NEXT:    movdqa %xmm1, %xmm0
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1]
+; SSE2-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSE2-NEXT:    punpcklbw {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3],xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSE2-NEXT:    movdqa %xmm3, %xmm2
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1]
+; SSE2-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: zext_16i8_to_8i64:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    movdqa {{.*#+}} xmm4 = [0,128,128,128,128,128,128,128,1,128,128,128,128,128,128,128]
+; SSSE3-NEXT:    pshufb %xmm4, %xmm0
+; SSSE3-NEXT:    movdqa {{.*#+}} xmm5 = [2,128,128,128,128,128,128,128,3,128,128,128,128,128,128,128]
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm3 = xmm1[1,1,2,3]
+; SSSE3-NEXT:    pshufb %xmm5, %xmm1
+; SSSE3-NEXT:    movdqa %xmm3, %xmm2
+; SSSE3-NEXT:    pshufb %xmm4, %xmm2
+; SSSE3-NEXT:    pshufb %xmm5, %xmm3
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: zext_16i8_to_8i64:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovzxbq {{.*#+}} xmm4 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero
+; SSE41-NEXT:    movdqa %xmm0, %xmm1
+; SSE41-NEXT:    psrld $16, %xmm1
+; SSE41-NEXT:    pmovzxbq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,zero,zero,zero,zero,xmm1[1],zero,zero,zero,zero,zero,zero,zero
+; SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
+; SSE41-NEXT:    pmovzxbq {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,zero,zero,zero,zero,xmm2[1],zero,zero,zero,zero,zero,zero,zero
+; SSE41-NEXT:    psrlq $48, %xmm0
+; SSE41-NEXT:    pmovzxbq {{.*#+}} xmm3 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero
+; SSE41-NEXT:    movdqa %xmm4, %xmm0
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: zext_16i8_to_8i64:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovzxbq {{.*#+}} xmm1 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero
+; AVX1-NEXT:    vpsrld $16, %xmm0, %xmm2
+; AVX1-NEXT:    vpmovzxbq {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,zero,zero,zero,zero,xmm2[1],zero,zero,zero,zero,zero,zero,zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; AVX1-NEXT:    vpmovzxbq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,zero,zero,zero,zero,xmm1[1],zero,zero,zero,zero,zero,zero,zero
+; AVX1-NEXT:    vpsrlq $48, %xmm0, %xmm0
+; AVX1-NEXT:    vpmovzxbq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: zext_16i8_to_8i64:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovzxbq {{.*#+}} ymm2 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero,xmm0[2],zero,zero,zero,zero,zero,zero,zero,xmm0[3],zero,zero,zero,zero,zero,zero,zero
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[1,1,2,3]
+; AVX2-NEXT:    vpmovzxbq {{.*#+}} ymm1 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero,xmm0[2],zero,zero,zero,zero,zero,zero,zero,xmm0[3],zero,zero,zero,zero,zero,zero,zero
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: zext_16i8_to_8i64:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovzxbq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,zero,zero,zero,zero,xmm0[1],zero,zero,zero,zero,zero,zero,zero,xmm0[2],zero,zero,zero,zero,zero,zero,zero,xmm0[3],zero,zero,zero,zero,zero,zero,zero,xmm0[4],zero,zero,zero,zero,zero,zero,zero,xmm0[5],zero,zero,zero,zero,zero,zero,zero,xmm0[6],zero,zero,zero,zero,zero,zero,zero,xmm0[7],zero,zero,zero,zero,zero,zero,zero
+; AVX512-NEXT:    vpandq {{.*}}(%rip){1to8}, %zmm0, %zmm0
+; AVX512-NEXT:    retq
+entry:
+  %B = shufflevector <16 x i8> %A, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %C = zext <8 x i8> %B to <8 x i64>
+  ret <8 x i64> %C
+}
+
 define <4 x i32> @zext_8i16_to_4i32(<8 x i16> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: zext_8i16_to_4i32:
 ; SSE2:       # BB#0: # %entry
@@ -318,6 +538,72 @@ entry:
   ret <8 x i32>%B
 }
 
+define <16 x i32> @zext_16i16_to_16i32(<16 x i16> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: zext_16i16_to_16i32:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movdqa %xmm1, %xmm3
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    pxor %xmm4, %xmm4
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1],xmm0[2],xmm4[2],xmm0[3],xmm4[3]
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
+; SSE2-NEXT:    movdqa %xmm3, %xmm2
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1],xmm2[2],xmm4[2],xmm2[3],xmm4[3]
+; SSE2-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: zext_16i16_to_16i32:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm1, %xmm3
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    pxor %xmm4, %xmm4
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1],xmm0[2],xmm4[2],xmm0[3],xmm4[3]
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
+; SSSE3-NEXT:    movdqa %xmm3, %xmm2
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1],xmm2[2],xmm4[2],xmm2[3],xmm4[3]
+; SSSE3-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: zext_16i16_to_16i32:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    movdqa %xmm1, %xmm3
+; SSE41-NEXT:    movdqa %xmm0, %xmm1
+; SSE41-NEXT:    pxor %xmm4, %xmm4
+; SSE41-NEXT:    pmovzxwd {{.*#+}} xmm0 = xmm1[0],zero,xmm1[1],zero,xmm1[2],zero,xmm1[3],zero
+; SSE41-NEXT:    punpckhwd {{.*#+}} xmm1 = xmm1[4],xmm4[4],xmm1[5],xmm4[5],xmm1[6],xmm4[6],xmm1[7],xmm4[7]
+; SSE41-NEXT:    pmovzxwd {{.*#+}} xmm2 = xmm3[0],zero,xmm3[1],zero,xmm3[2],zero,xmm3[3],zero
+; SSE41-NEXT:    punpckhwd {{.*#+}} xmm3 = xmm3[4],xmm4[4],xmm3[5],xmm4[5],xmm3[6],xmm4[6],xmm3[7],xmm4[7]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: zext_16i16_to_16i32:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm2 = xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; AVX1-NEXT:    vpmovzxwd {{.*#+}} xmm3 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm3, %ymm2
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpunpckhwd {{.*#+}} xmm1 = xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
+; AVX1-NEXT:    vpmovzxwd {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: zext_16i16_to_16i32:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovzxwd {{.*#+}} ymm2 = 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:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpmovzxwd {{.*#+}} ymm1 = 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:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: zext_16i16_to_16i32:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovzxwd {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero,ymm0[8],zero,ymm0[9],zero,ymm0[10],zero,ymm0[11],zero,ymm0[12],zero,ymm0[13],zero,ymm0[14],zero,ymm0[15],zero
+; AVX512-NEXT:    retq
+entry:
+  %B = zext <16 x i16> %A to <16 x i32>
+  ret <16 x i32> %B
+}
+
 define <2 x i64> @zext_8i16_to_2i64(<8 x i16> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: zext_8i16_to_2i64:
 ; SSE2:       # BB#0: # %entry
@@ -400,6 +686,80 @@ entry:
   ret <4 x i64> %C
 }
 
+define <8 x i64> @zext_8i16_to_8i64(<8 x i16> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: zext_8i16_to_8i64:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    pxor %xmm4, %xmm4
+; SSE2-NEXT:    pshufd {{.*#+}} xmm3 = xmm1[2,3,0,1]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSE2-NEXT:    movdqa %xmm1, %xmm0
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1]
+; SSE2-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSE2-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSE2-NEXT:    movdqa %xmm3, %xmm2
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1]
+; SSE2-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: zext_8i16_to_8i64:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    pxor %xmm4, %xmm4
+; SSSE3-NEXT:    pshufd {{.*#+}} xmm3 = xmm1[2,3,0,1]
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm1 = xmm1[0],xmm4[0],xmm1[1],xmm4[1],xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSSE3-NEXT:    movdqa %xmm1, %xmm0
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1]
+; SSSE3-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSSE3-NEXT:    punpcklwd {{.*#+}} xmm3 = xmm3[0],xmm4[0],xmm3[1],xmm4[1],xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSSE3-NEXT:    movdqa %xmm3, %xmm2
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1]
+; SSSE3-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: zext_8i16_to_8i64:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    pmovzxwq {{.*#+}} xmm4 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
+; SSE41-NEXT:    pshufd {{.*#+}} xmm1 = xmm0[1,1,2,3]
+; SSE41-NEXT:    pmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
+; SSE41-NEXT:    pshufd {{.*#+}} xmm2 = xmm0[2,3,0,1]
+; SSE41-NEXT:    pmovzxwq {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero
+; SSE41-NEXT:    pshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; SSE41-NEXT:    pmovzxwq {{.*#+}} xmm3 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
+; SSE41-NEXT:    movdqa %xmm4, %xmm0
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: zext_8i16_to_8i64:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpmovzxwq {{.*#+}} xmm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm2 = xmm0[1,1,2,3]
+; AVX1-NEXT:    vpmovzxwq {{.*#+}} xmm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm1, %ymm2
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm1 = xmm0[2,3,0,1]
+; AVX1-NEXT:    vpmovzxwq {{.*#+}} xmm1 = xmm1[0],zero,zero,zero,xmm1[1],zero,zero,zero
+; AVX1-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[3,1,2,3]
+; AVX1-NEXT:    vpmovzxwq {{.*#+}} xmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: zext_8i16_to_8i64:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovzxwq {{.*#+}} ymm2 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; AVX2-NEXT:    vpshufd {{.*#+}} xmm0 = xmm0[2,3,0,1]
+; AVX2-NEXT:    vpmovzxwq {{.*#+}} ymm1 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: zext_8i16_to_8i64:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovzxwq {{.*#+}} zmm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero
+; AVX512-NEXT:    retq
+entry:
+  %B = zext <8 x i16> %A to <8 x i64>
+  ret <8 x i64> %B
+}
+
 define <2 x i64> @zext_4i32_to_2i64(<4 x i32> %A) nounwind uwtable readnone ssp {
 ; SSE2-LABEL: zext_4i32_to_2i64:
 ; SSE2:       # BB#0: # %entry
@@ -475,6 +835,72 @@ entry:
   ret <4 x i64>%B
 }
 
+define <8 x i64> @zext_8i32_to_8i64(<8 x i32> %A) nounwind uwtable readnone ssp {
+; SSE2-LABEL: zext_8i32_to_8i64:
+; SSE2:       # BB#0: # %entry
+; SSE2-NEXT:    movdqa %xmm1, %xmm3
+; SSE2-NEXT:    movdqa %xmm0, %xmm1
+; SSE2-NEXT:    pxor %xmm4, %xmm4
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1]
+; SSE2-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSE2-NEXT:    movdqa %xmm3, %xmm2
+; SSE2-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1]
+; SSE2-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSE2-NEXT:    retq
+;
+; SSSE3-LABEL: zext_8i32_to_8i64:
+; SSSE3:       # BB#0: # %entry
+; SSSE3-NEXT:    movdqa %xmm1, %xmm3
+; SSSE3-NEXT:    movdqa %xmm0, %xmm1
+; SSSE3-NEXT:    pxor %xmm4, %xmm4
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm0 = xmm0[0],xmm4[0],xmm0[1],xmm4[1]
+; SSSE3-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSSE3-NEXT:    movdqa %xmm3, %xmm2
+; SSSE3-NEXT:    punpckldq {{.*#+}} xmm2 = xmm2[0],xmm4[0],xmm2[1],xmm4[1]
+; SSSE3-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSSE3-NEXT:    retq
+;
+; SSE41-LABEL: zext_8i32_to_8i64:
+; SSE41:       # BB#0: # %entry
+; SSE41-NEXT:    movdqa %xmm1, %xmm3
+; SSE41-NEXT:    movdqa %xmm0, %xmm1
+; SSE41-NEXT:    pxor %xmm4, %xmm4
+; SSE41-NEXT:    pmovzxdq {{.*#+}} xmm0 = xmm1[0],zero,xmm1[1],zero
+; SSE41-NEXT:    punpckhdq {{.*#+}} xmm1 = xmm1[2],xmm4[2],xmm1[3],xmm4[3]
+; SSE41-NEXT:    pmovzxdq {{.*#+}} xmm2 = xmm3[0],zero,xmm3[1],zero
+; SSE41-NEXT:    punpckhdq {{.*#+}} xmm3 = xmm3[2],xmm4[2],xmm3[3],xmm4[3]
+; SSE41-NEXT:    retq
+;
+; AVX1-LABEL: zext_8i32_to_8i64:
+; AVX1:       # BB#0: # %entry
+; AVX1-NEXT:    vpxor %xmm1, %xmm1, %xmm1
+; AVX1-NEXT:    vpunpckhdq {{.*#+}} xmm2 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
+; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm3 = xmm0[0],zero,xmm0[1],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm2, %ymm3, %ymm2
+; AVX1-NEXT:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vpunpckhdq {{.*#+}} xmm1 = xmm0[2],xmm1[2],xmm0[3],xmm1[3]
+; AVX1-NEXT:    vpmovzxdq {{.*#+}} xmm0 = xmm0[0],zero,xmm0[1],zero
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm1
+; AVX1-NEXT:    vmovaps %ymm2, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: zext_8i32_to_8i64:
+; AVX2:       # BB#0: # %entry
+; AVX2-NEXT:    vpmovzxdq {{.*#+}} ymm2 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; AVX2-NEXT:    vextracti128 $1, %ymm0, %xmm0
+; AVX2-NEXT:    vpmovzxdq {{.*#+}} ymm1 = xmm0[0],zero,xmm0[1],zero,xmm0[2],zero,xmm0[3],zero
+; AVX2-NEXT:    vmovdqa %ymm2, %ymm0
+; AVX2-NEXT:    retq
+;
+; AVX512-LABEL: zext_8i32_to_8i64:
+; AVX512:       # BB#0: # %entry
+; AVX512-NEXT:    vpmovzxdq {{.*#+}} zmm0 = ymm0[0],zero,ymm0[1],zero,ymm0[2],zero,ymm0[3],zero,ymm0[4],zero,ymm0[5],zero,ymm0[6],zero,ymm0[7],zero
+; AVX512-NEXT:    retq
+entry:
+  %B = zext <8 x i32> %A to <8 x i64>
+  ret <8 x i64>%B
+}
+
 define <2 x i64> @load_zext_2i8_to_2i64(<2 x i8> *%ptr) {
 ; SSE2-LABEL: load_zext_2i8_to_2i64:
 ; SSE2:       # BB#0: # %entry




More information about the llvm-commits mailing list