Make utils/update_llc_test_checks.py note that the assertions are

[oota-llvm.git] / test / CodeGen / X86 / vector-shuffle-256-v4.ll

diff --git a/test/CodeGen/X86/vector-shuffle-256-v4.ll b/test/CodeGen/X86/vector-shuffle-256-v4.ll
index f6dd6c773dac10a51e135c253a9e8259cdab7a64..8c6d5105d8203775c529252cfec86c04e251a4ad 100644 (file)
--- a/test/CodeGen/X86/vector-shuffle-256-v4.ll
+++ b/test/CodeGen/X86/vector-shuffle-256-v4.ll
@@ -1,5 +1,6 @@
-; RUN: llc < %s -mcpu=x86-64 -mattr=+avx | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX1
-; RUN: llc < %s -mcpu=x86-64 -mattr=+avx2 | FileCheck %s --check-prefix=ALL --check-prefix=AVX --check-prefix=AVX2
+; 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
 
 target triple = "x86_64-unknown-unknown"
@@ -170,15 +171,10 @@ define <4 x double> @shuffle_v4f64_3210(<4 x double> %a, <4 x double> %b) {
 }
 
 define <4 x double> @shuffle_v4f64_0023(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_0023:
-; AVX:       # BB#0:
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[0,0,2,3]
-; AVX-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
@@ -193,58 +189,48 @@ define <4 x double> @shuffle_v4f64_0022(<4 x double> %a, <4 x double> %b) {
   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) {
-; AVX-LABEL: shuffle_v4f64_1032:
-; AVX:       # BB#0:
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX-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) {
-; AVX-LABEL: shuffle_v4f64_1133:
-; AVX:       # BB#0:
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,1,3,3]
-; AVX-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) {
-; AVX-LABEL: shuffle_v4f64_1023:
-; AVX:       # BB#0:
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,3]
-; AVX-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) {
-; AVX-LABEL: shuffle_v4f64_1022:
-; AVX:       # BB#0:
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,2,2]
-; AVX-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
 }
@@ -283,71 +269,46 @@ define <4 x double> @shuffle_v4f64_0462(<4 x double> %a, <4 x double> %b) {
 }
 
 define <4 x double> @shuffle_v4f64_0426(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_0426:
-; AVX:       # BB#0:
-; AVX-NEXT:    vunpcklpd {{.*#+}} ymm0 = ymm0[0],ymm1[0],ymm0[2],ymm1[2]
-; AVX-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) {
-; AVX-LABEL: shuffle_v4f64_1537:
-; AVX:       # BB#0:
-; AVX-NEXT:    vunpckhpd {{.*#+}} ymm0 = ymm0[1],ymm1[1],ymm0[3],ymm1[3]
-; AVX-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) {
-; AVX-LABEL: shuffle_v4f64_4062:
-; AVX:       # BB#0:
-; AVX-NEXT:    vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
-; AVX-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) {
-; AVX-LABEL: shuffle_v4f64_5173:
-; AVX:       # BB#0:
-; AVX-NEXT:    vunpckhpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[3],ymm0[3]
-; AVX-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) {
-; AVX-LABEL: shuffle_v4f64_5163:
-; AVX:       # BB#0:
-; AVX-NEXT:    vshufpd {{.*#+}} ymm0 = ymm1[1],ymm0[1],ymm1[2],ymm0[3]
-; AVX-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
 }
@@ -371,10 +332,15 @@ define <4 x double> @shuffle_v4f64_4163(<4 x double> %a, <4 x double> %b) {
 }
 
 define <4 x double> @shuffle_v4f64_0145(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_0145:
-; AVX:       # BB#0:
-; AVX-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: shuffle_v4f64_0145:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v4f64_0145:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: shuffle_v4f64_0145:
 ; AVX512VL:       # BB#0:
@@ -385,10 +351,15 @@ define <4 x double> @shuffle_v4f64_0145(<4 x double> %a, <4 x double> %b) {
 }
 
 define <4 x double> @shuffle_v4f64_4501(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_4501:
-; AVX:       # BB#0:
-; AVX-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: shuffle_v4f64_4501:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v4f64_4501:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: shuffle_v4f64_4501:
 ; AVX512VL:       # BB#0:
@@ -408,65 +379,77 @@ define <4 x double> @shuffle_v4f64_0167(<4 x double> %a, <4 x double> %b) {
 }
 
 define <4 x double> @shuffle_v4f64_1054(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_1054:
-; AVX:       # BB#0:
-; AVX-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX-NEXT:    retq
+; AVX1-LABEL: shuffle_v4f64_1054:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v4f64_1054:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; AVX2-NEXT:    retq
 ;
 ; 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
 }
 
 define <4 x double> @shuffle_v4f64_3254(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_3254:
-; AVX:       # BB#0:
-; AVX-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[0,1]
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX-NEXT:    retq
+; AVX1-LABEL: shuffle_v4f64_3254:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[0,1]
+; AVX1-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v4f64_3254:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[0,1]
+; AVX2-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; AVX2-NEXT:    retq
 ;
 ; 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
 }
 
 define <4 x double> @shuffle_v4f64_3276(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_3276:
-; AVX:       # BB#0:
-; AVX-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX-NEXT:    retq
+; AVX1-LABEL: shuffle_v4f64_3276:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
+; AVX1-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v4f64_3276:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vperm2f128 {{.*#+}} ymm0 = ymm0[2,3],ymm1[2,3]
+; AVX2-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
+; AVX2-NEXT:    retq
 ;
 ; 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) {
-; AVX-LABEL: shuffle_v4f64_1076:
-; AVX:       # BB#0:
-; AVX-NEXT:    vblendpd {{.*#+}} ymm0 = ymm0[0,1],ymm1[2,3]
-; AVX-NEXT:    vpermilpd {{.*#+}} ymm0 = ymm0[1,0,3,2]
-; AVX-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
 }
@@ -497,16 +480,61 @@ define <4 x double> @shuffle_v4f64_0415(<4 x double> %a, <4 x double> %b) {
 }
 
 define <4 x double> @shuffle_v4f64_u062(<4 x double> %a, <4 x double> %b) {
-; AVX-LABEL: shuffle_v4f64_u062:
-; AVX:       # BB#0:
-; AVX-NEXT:    vunpcklpd {{.*#+}} ymm0 = ymm1[0],ymm0[0],ymm1[2],ymm0[2]
-; AVX-NEXT:    retq
+; 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:    vextractf128 $1, %ymm0, %xmm0
+; AVX1-NEXT:    vmovddup {{.*#+}} xmm0 = xmm0[0,0]
+; AVX1-NEXT:    retq
 ;
-; AVX512VL-LABEL: shuffle_v4f64_u062:
+; AVX2-LABEL: shuffle_v4f64_22uu:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vpermpd {{.*#+}} ymm0 = ymm0[2,2,2,3]
+; AVX2-NEXT:    retq
+;
+; 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
 }
 
@@ -797,10 +825,15 @@ define <4 x i64> @shuffle_v4i64_4012(<4 x i64> %a, <4 x i64> %b) {
 }
 
 define <4 x i64> @shuffle_v4i64_0145(<4 x i64> %a, <4 x i64> %b) {
-; AVX-LABEL: shuffle_v4i64_0145:
-; AVX:       # BB#0:
-; AVX-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: shuffle_v4i64_0145:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v4i64_0145:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: shuffle_v4i64_0145:
 ; AVX512VL:       # BB#0:
@@ -836,10 +869,15 @@ define <4 x i64> @shuffle_v4i64_0451(<4 x i64> %a, <4 x i64> %b) {
 }
 
 define <4 x i64> @shuffle_v4i64_4501(<4 x i64> %a, <4 x i64> %b) {
-; AVX-LABEL: shuffle_v4i64_4501:
-; AVX:       # BB#0:
-; AVX-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: shuffle_v4i64_4501:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: shuffle_v4i64_4501:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vinsertf128 $1, %xmm0, %ymm1, %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: shuffle_v4i64_4501:
 ; AVX512VL:       # BB#0:
@@ -1090,12 +1128,72 @@ define <4 x i64> @shuffle_v4i64_40u2(<4 x i64> %a, <4 x i64> %b) {
 ;
 ; 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
@@ -1118,10 +1216,15 @@ define <4 x i64> @insert_reg_and_zero_v4i64(i64 %a) {
 }
 
 define <4 x i64> @insert_mem_and_zero_v4i64(i64* %ptr) {
-; AVX-LABEL: insert_mem_and_zero_v4i64:
-; AVX:       # BB#0:
-; AVX-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
-; AVX-NEXT:    retq
+; AVX1-LABEL: insert_mem_and_zero_v4i64:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_mem_and_zero_v4i64:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vmovq {{.*#+}} xmm0 = mem[0],zero
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: insert_mem_and_zero_v4i64:
 ; AVX512VL:       # BB#0:
@@ -1134,11 +1237,17 @@ define <4 x i64> @insert_mem_and_zero_v4i64(i64* %ptr) {
 }
 
 define <4 x double> @insert_reg_and_zero_v4f64(double %a) {
-; AVX-LABEL: insert_reg_and_zero_v4f64:
-; AVX:       # BB#0:
-; AVX-NEXT:    vxorpd %ymm1, %ymm1, %ymm1
-; AVX-NEXT:    vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
-; AVX-NEXT:    retq
+; AVX1-LABEL: insert_reg_and_zero_v4f64:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vxorpd %ymm1, %ymm1, %ymm1
+; AVX1-NEXT:    vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_reg_and_zero_v4f64:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vxorpd %ymm1, %ymm1, %ymm1
+; AVX2-NEXT:    vblendpd {{.*#+}} ymm0 = ymm0[0],ymm1[1,2,3]
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: insert_reg_and_zero_v4f64:
 ; AVX512VL:       # BB#0:
@@ -1151,10 +1260,15 @@ define <4 x double> @insert_reg_and_zero_v4f64(double %a) {
 }
 
 define <4 x double> @insert_mem_and_zero_v4f64(double* %ptr) {
-; AVX-LABEL: insert_mem_and_zero_v4f64:
-; AVX:       # BB#0:
-; AVX-NEXT:    vmovsd {{.*#+}} xmm0 = mem[0],zero
-; AVX-NEXT:    retq
+; AVX1-LABEL: insert_mem_and_zero_v4f64:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vmovsd {{.*#+}} xmm0 = mem[0],zero
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_mem_and_zero_v4f64:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vmovsd {{.*#+}} xmm0 = mem[0],zero
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: insert_mem_and_zero_v4f64:
 ; AVX512VL:       # BB#0:
@@ -1178,10 +1292,15 @@ define <4 x double> @splat_mem_v4f64(double* %ptr) {
 }
 
 define <4 x i64> @splat_mem_v4i64(i64* %ptr) {
-; AVX-LABEL: splat_mem_v4i64:
-; AVX:       # BB#0:
-; AVX-NEXT:    vbroadcastsd (%rdi), %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: splat_mem_v4i64:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vbroadcastsd (%rdi), %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: splat_mem_v4i64:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vbroadcastsd (%rdi), %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: splat_mem_v4i64:
 ; AVX512VL:       # BB#0:
@@ -1268,11 +1387,17 @@ define <4 x double> @splat_mem_v4f64_from_v2f64(<2 x double>* %ptr) {
 }
 
 define <4 x i64> @splat128_mem_v4i64_from_v2i64(<2 x i64>* %ptr) {
-; AVX-LABEL: splat128_mem_v4i64_from_v2i64:
-; AVX:       # BB#0:
-; AVX-NEXT:    vmovaps (%rdi), %xmm0
-; AVX-NEXT:    vinsertf128 $1, %xmm0, %ymm0, %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: splat128_mem_v4i64_from_v2i64:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vmovaps (%rdi), %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: splat128_mem_v4i64_from_v2i64:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vmovaps (%rdi), %xmm0
+; AVX2-NEXT:    vinsertf128 $1, %xmm0, %ymm0, %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: splat128_mem_v4i64_from_v2i64:
 ; AVX512VL:       # BB#0:
@@ -1285,11 +1410,17 @@ define <4 x i64> @splat128_mem_v4i64_from_v2i64(<2 x i64>* %ptr) {
 }
 
 define <4 x double> @splat128_mem_v4f64_from_v2f64(<2 x double>* %ptr) {
-; AVX-LABEL: splat128_mem_v4f64_from_v2f64:
-; AVX:       # BB#0:
-; AVX-NEXT:    vmovaps (%rdi), %xmm0
-; AVX-NEXT:    vinsertf128 $1, %xmm0, %ymm0, %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: splat128_mem_v4f64_from_v2f64:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vmovaps (%rdi), %xmm0
+; AVX1-NEXT:    vinsertf128 $1, %xmm0, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: splat128_mem_v4f64_from_v2f64:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vmovaps (%rdi), %xmm0
+; AVX2-NEXT:    vinsertf128 $1, %xmm0, %ymm0, %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: splat128_mem_v4f64_from_v2f64:
 ; AVX512VL:       # BB#0:
@@ -1314,7 +1445,7 @@ define <4 x double> @bitcast_v4f64_0426(<4 x double> %a, <4 x double> %b) {
 ;
 ; 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>
@@ -1347,10 +1478,15 @@ define <4 x i64> @concat_v4i64_0167(<4 x i64> %a0, <4 x i64> %a1) {
 }
 
 define <4 x i64> @concat_v4i64_0145_bc(<4 x i64> %a0, <4 x i64> %a1) {
-; AVX-LABEL: concat_v4i64_0145_bc:
-; AVX:       # BB#0:
-; AVX-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: concat_v4i64_0145_bc:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: concat_v4i64_0145_bc:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vinsertf128 $1, %xmm1, %ymm0, %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: concat_v4i64_0145_bc:
 ; AVX512VL:       # BB#0:
@@ -1366,10 +1502,15 @@ define <4 x i64> @concat_v4i64_0145_bc(<4 x i64> %a0, <4 x i64> %a1) {
 }
 
 define <4 x i64> @insert_dup_mem_v4i64(i64* %ptr) {
-; AVX-LABEL: insert_dup_mem_v4i64:
-; AVX:       # BB#0:
-; AVX-NEXT:    vbroadcastsd (%rdi), %ymm0
-; AVX-NEXT:    retq
+; AVX1-LABEL: insert_dup_mem_v4i64:
+; AVX1:       # BB#0:
+; AVX1-NEXT:    vbroadcastsd (%rdi), %ymm0
+; AVX1-NEXT:    retq
+;
+; AVX2-LABEL: insert_dup_mem_v4i64:
+; AVX2:       # BB#0:
+; AVX2-NEXT:    vbroadcastsd (%rdi), %ymm0
+; AVX2-NEXT:    retq
 ;
 ; AVX512VL-LABEL: insert_dup_mem_v4i64:
 ; AVX512VL:       # BB#0: