+define <16 x i8> @trunc2x8i16(<8 x i16> %a, <8 x i16> %b) {
+; SSE2-LABEL: trunc2x8i16:
+; SSE2: # BB#0: # %entry
+; SSE2-NEXT: movdqa {{.*#+}} xmm2 = [255,255,255,255,255,255,255,255]
+; SSE2-NEXT: pand %xmm2, %xmm1
+; SSE2-NEXT: pand %xmm2, %xmm0
+; SSE2-NEXT: packuswb %xmm1, %xmm0
+; SSE2-NEXT: retq
+
+; SSSE3-LABEL: trunc2x8i16:
+; SSSE3: # BB#0: # %entry
+; SSSE3-NEXT: pshufb {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0,2,4,6,8,10,12,14]
+; SSSE3-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,2,4,6,8,10,12,14],zero,zero,zero,zero,zero,zero,zero,zero
+; SSSE3-NEXT: por %xmm1, %xmm0
+; SSSE3-NEXT: retq
+
+; SSE41-LABEL: trunc2x8i16:
+; SSE41: # BB#0: # %entry
+; SSE41-NEXT: pshufb {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0,2,4,6,8,10,12,14]
+; SSE41-NEXT: pshufb {{.*#+}} xmm0 = xmm0[0,2,4,6,8,10,12,14],zero,zero,zero,zero,zero,zero,zero,zero
+; SSE41-NEXT: por %xmm1, %xmm0
+; SSE41-NEXT: retq
+
+; AVX-LABEL: trunc2x8i16:
+; AVX: # BB#0: # %entry
+; AVX-NEXT: vpshufb {{.*#+}} xmm1 = zero,zero,zero,zero,zero,zero,zero,zero,xmm1[0,2,4,6,8,10,12,14]
+; AVX-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[0,2,4,6,8,10,12,14],zero,zero,zero,zero,zero,zero,zero,zero
+; AVX-NEXT: vpor %xmm1, %xmm0, %xmm0
+; AVX-NEXT: retq
+
+entry:
+ %0 = trunc <8 x i16> %a to <8 x i8>
+ %1 = trunc <8 x i16> %b to <8 x i8>
+ %2 = shufflevector <8 x i8> %0, <8 x i8> %1, <16 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7, i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
+ ret <16 x i8> %2
+}
+