+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mcpu=x86-64 -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX1
; RUN: llc < %s -mcpu=x86-64 -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX2
; RUN: llc < %s -mcpu=knl -mattr=+avx512vl | FileCheck %s --check-prefix=ALL --check-prefix=AVX512VL
}
define <4 x double> @shuffle_v4f64_0023(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_0023:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,0,2,3]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_0023:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,0,2,3]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_0023:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $8, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_0023:
+; ALL: # BB#0:
+; ALL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[0,0,2,3]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 0, i32 0, i32 2, i32 3>
ret <4 x double> %shuffle
ret <4 x double> %shuffle
}
+define <4 x double> @shuffle_v4f64mem_0022(<4 x double>* %ptr, <4 x double> %b) {
+; ALL-LABEL: shuffle_v4f64mem_0022:
+; ALL: # BB#0:
+; ALL-NEXT: vmovddup {{.*#+}} ymm0 = mem[0,0,2,2]
+; ALL-NEXT: retq
+ %a = load <4 x double>, <4 x double>* %ptr
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 0, i32 0, i32 2, i32 2>
+ ret <4 x double> %shuffle
+}
+
define <4 x double> @shuffle_v4f64_1032(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_1032:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_1032:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_1032:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $5, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_1032:
+; ALL: # BB#0:
+; ALL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_1133(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_1133:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,1,3,3]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_1133:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,1,3,3]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_1133:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $15, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_1133:
+; ALL: # BB#0:
+; ALL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,1,3,3]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 1, i32 3, i32 3>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_1023(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_1023:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,3]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_1023:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,3]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_1023:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $9, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_1023:
+; ALL: # BB#0:
+; ALL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,3]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 0, i32 2, i32 3>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_1022(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_1022:
-; AVX1: # BB#0:
-; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,2]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_1022:
-; AVX2: # BB#0:
-; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,2]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_1022:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpermilpd $1, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_1022:
+; ALL: # BB#0:
+; ALL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,2]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 0, i32 2, i32 2>
ret <4 x double> %shuffle
}
}
define <4 x double> @shuffle_v4f64_0426(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_0426:
-; AVX1: # BB#0:
-; AVX1-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_0426:
-; AVX2: # BB#0:
-; AVX2-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_0426:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vunpcklpd %ymm1, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_0426:
+; ALL: # BB#0:
+; ALL-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 0, i32 4, i32 2, i32 6>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_1537(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_1537:
-; AVX1: # BB#0:
-; AVX1-NEXT: vunpckhpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[3],ymm1[3]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_1537:
-; AVX2: # BB#0:
-; AVX2-NEXT: vunpckhpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[3],ymm1[3]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_1537:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vunpckhpd %ymm1, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_1537:
+; ALL: # BB#0:
+; ALL-NEXT: vunpckhpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[3],ymm1[3]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 5, i32 3, i32 7>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_4062(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_4062:
-; AVX1: # BB#0:
-; AVX1-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_4062:
-; AVX2: # BB#0:
-; AVX2-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_4062:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vunpcklpd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_4062:
+; ALL: # BB#0:
+; ALL-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 4, i32 0, i32 6, i32 2>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_5173(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_5173:
-; AVX1: # BB#0:
-; AVX1-NEXT: vunpckhpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[3],ymm0[3]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_5173:
-; AVX2: # BB#0:
-; AVX2-NEXT: vunpckhpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[3],ymm0[3]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_5173:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vunpckhpd %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_5173:
+; ALL: # BB#0:
+; ALL-NEXT: vunpckhpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[3],ymm0[3]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 5, i32 1, i32 7, i32 3>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_5163(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_5163:
-; AVX1: # BB#0:
-; AVX1-NEXT: vshufpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[2],ymm0[3]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_5163:
-; AVX2: # BB#0:
-; AVX2-NEXT: vshufpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[2],ymm0[3]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_5163:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vshufpd $11, %ymm0, %ymm1, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_5163:
+; ALL: # BB#0:
+; ALL-NEXT: vshufpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[2],ymm0[3]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 5, i32 1, i32 6, i32 3>
ret <4 x double> %shuffle
}
; AVX512VL-LABEL: shuffle_v4f64_1054:
; AVX512VL: # BB#0:
; AVX512VL-NEXT: vinsertf32x4 $1, %xmm1, %ymm0, %ymm0
-; AVX512VL-NEXT: vpermilpd $5, %ymm0, %ymm0
+; AVX512VL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
; AVX512VL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 0, i32 5, i32 4>
ret <4 x double> %shuffle
; AVX512VL-LABEL: shuffle_v4f64_3254:
; AVX512VL: # BB#0:
; AVX512VL-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[0,1]
-; AVX512VL-NEXT: vpermilpd $5, %ymm0, %ymm0
+; AVX512VL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
; AVX512VL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 3, i32 2, i32 5, i32 4>
ret <4 x double> %shuffle
; AVX512VL-LABEL: shuffle_v4f64_3276:
; AVX512VL: # BB#0:
; AVX512VL-NEXT: vperm2i128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; AVX512VL-NEXT: vpermilpd $5, %ymm0, %ymm0
+; AVX512VL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
; AVX512VL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 3, i32 2, i32 7, i32 6>
ret <4 x double> %shuffle
}
define <4 x double> @shuffle_v4f64_1076(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_1076:
-; AVX1: # BB#0:
-; AVX1-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
-; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX1-NEXT: retq
-;
-; AVX2-LABEL: shuffle_v4f64_1076:
-; AVX2: # BB#0:
-; AVX2-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
-; AVX2-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX2-NEXT: retq
-;
-; AVX512VL-LABEL: shuffle_v4f64_1076:
-; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
-; AVX512VL-NEXT: vpermilpd $5, %ymm0, %ymm0
-; AVX512VL-NEXT: retq
+; ALL-LABEL: shuffle_v4f64_1076:
+; ALL: # BB#0:
+; ALL-NEXT: vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
+; ALL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; ALL-NEXT: retq
%shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 0, i32 7, i32 6>
ret <4 x double> %shuffle
}
}
define <4 x double> @shuffle_v4f64_u062(<4 x double> %a, <4 x double> %b) {
-; AVX1-LABEL: shuffle_v4f64_u062:
+; ALL-LABEL: shuffle_v4f64_u062:
+; ALL: # BB#0:
+; ALL-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
+; ALL-NEXT: retq
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 undef, i32 0, i32 6, i32 2>
+ ret <4 x double> %shuffle
+}
+
+define <4 x double> @shuffle_v4f64_11uu(<4 x double> %a, <4 x double> %b) {
+; ALL-LABEL: shuffle_v4f64_11uu:
+; ALL: # BB#0:
+; ALL-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,1,2,2]
+; ALL-NEXT: retq
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 1, i32 1, i32 undef, i32 undef>
+ ret <4 x double> %shuffle
+}
+
+define <4 x double> @shuffle_v4f64_22uu(<4 x double> %a, <4 x double> %b) {
+; AVX1-LABEL: shuffle_v4f64_22uu:
; AVX1: # BB#0:
-; AVX1-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
; AVX1-NEXT: retq
;
-; AVX2-LABEL: shuffle_v4f64_u062:
+; AVX2-LABEL: shuffle_v4f64_22uu:
; AVX2: # BB#0:
-; AVX2-NEXT: vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
+; AVX2-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX2-NEXT: retq
;
-; AVX512VL-LABEL: shuffle_v4f64_u062:
+; AVX512VL-LABEL: shuffle_v4f64_22uu:
; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vunpcklpd %ymm0, %ymm1, %ymm0
+; AVX512VL-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[2,2,2,3]
; AVX512VL-NEXT: retq
- %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 undef, i32 0, i32 6, i32 2>
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 2, i32 2, i32 undef, i32 undef>
+ ret <4 x double> %shuffle
+}
+
+define <4 x double> @shuffle_v4f64_3333(<4 x double> %a, <4 x double> %b) {
+; AVX1-LABEL: shuffle_v4f64_3333:
+; AVX1: # BB#0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vmovhlps {{.*#+}} xmm0 = xmm0[1,1]
+; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4f64_3333:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,3,3,3]
+; AVX2-NEXT: retq
+;
+; AVX512VL-LABEL: shuffle_v4f64_3333:
+; AVX512VL: # BB#0:
+; AVX512VL-NEXT: vpermpd {{.*#+}} ymm0 = ymm0[3,3,3,3]
+; AVX512VL-NEXT: retq
+ %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
ret <4 x double> %shuffle
}
;
; AVX512VL-LABEL: shuffle_v4i64_40u2:
; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpunpcklqdq %ymm0, %ymm1, %ymm0
+; AVX512VL-NEXT: vpunpcklqdq {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
; AVX512VL-NEXT: retq
%shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 4, i32 0, i32 undef, i32 2>
ret <4 x i64> %shuffle
}
+define <4 x i64> @shuffle_v4i64_11uu(<4 x i64> %a, <4 x i64> %b) {
+; AVX1-LABEL: shuffle_v4i64_11uu:
+; AVX1: # BB#0:
+; AVX1-NEXT: vpermilpd {{.*#+}} ymm0 = ymm0[1,1,2,2]
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4i64_11uu:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[2,3,2,3,6,7,6,7]
+; AVX2-NEXT: retq
+;
+; AVX512VL-LABEL: shuffle_v4i64_11uu:
+; AVX512VL: # BB#0:
+; AVX512VL-NEXT: vpshufd {{.*#+}} ymm0 = ymm0[2,3,2,3,6,7,6,7]
+; AVX512VL-NEXT: retq
+ %shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 1, i32 1, i32 undef, i32 undef>
+ ret <4 x i64> %shuffle
+}
+
+define <4 x i64> @shuffle_v4i64_22uu(<4 x i64> %a, <4 x i64> %b) {
+; AVX1-LABEL: shuffle_v4i64_22uu:
+; AVX1: # BB#0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4i64_22uu:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
+; AVX2-NEXT: retq
+;
+; AVX512VL-LABEL: shuffle_v4i64_22uu:
+; AVX512VL: # BB#0:
+; AVX512VL-NEXT: vpermq {{.*#+}} ymm0 = ymm0[2,2,2,3]
+; AVX512VL-NEXT: retq
+ %shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 2, i32 2, i32 undef, i32 undef>
+ ret <4 x i64> %shuffle
+}
+
+define <4 x i64> @shuffle_v4i64_3333(<4 x i64> %a, <4 x i64> %b) {
+; AVX1-LABEL: shuffle_v4i64_3333:
+; AVX1: # BB#0:
+; AVX1-NEXT: vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT: vpermilpd {{.*#+}} xmm0 = xmm0[1,1]
+; AVX1-NEXT: vinsertf128 $1, %xmm0, %ymm0, %ymm0
+; AVX1-NEXT: retq
+;
+; AVX2-LABEL: shuffle_v4i64_3333:
+; AVX2: # BB#0:
+; AVX2-NEXT: vpermq {{.*#+}} ymm0 = ymm0[3,3,3,3]
+; AVX2-NEXT: retq
+;
+; AVX512VL-LABEL: shuffle_v4i64_3333:
+; AVX512VL: # BB#0:
+; AVX512VL-NEXT: vpermq {{.*#+}} ymm0 = ymm0[3,3,3,3]
+; AVX512VL-NEXT: retq
+ %shuffle = shufflevector <4 x i64> %a, <4 x i64> %b, <4 x i32> <i32 3, i32 3, i32 3, i32 3>
+ ret <4 x i64> %shuffle
+}
+
define <4 x i64> @stress_test1(<4 x i64> %a, <4 x i64> %b) {
; ALL-LABEL: stress_test1:
; ALL: retq
;
; AVX512VL-LABEL: bitcast_v4f64_0426:
; AVX512VL: # BB#0:
-; AVX512VL-NEXT: vpunpcklqdq %ymm1, %ymm0, %ymm0
+; AVX512VL-NEXT: vpunpcklqdq {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
; AVX512VL-NEXT: retq
%shuffle64 = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 4, i32 0, i32 6, i32 2>
%bitcast32 = bitcast <4 x double> %shuffle64 to <8 x float>