From: Sanjay Patel <spatel@rotateright.com>
Date: Wed, 19 Apr 2017 14:55:50 +0000 (+0000)
Subject: [ARM] add test and auto-generate checks; NFC
X-Git-Tag: llvmorg-5.0.0-rc1~7318
X-Git-Url: http://review.tizen.org/git/?a=commitdiff_plain;h=5a2235bbd069169fbc4ce2058f99627c0ed447bd;p=platform%2Fupstream%2Fllvm.git

[ARM] add test and auto-generate checks; NFC

llvm-svn: 300698
---

diff --git a/llvm/test/CodeGen/ARM/vbits.ll b/llvm/test/CodeGen/ARM/vbits.ll
index db9bc6c..040904b 100644
--- a/llvm/test/CodeGen/ARM/vbits.ll
+++ b/llvm/test/CodeGen/ARM/vbits.ll
@@ -1,8 +1,14 @@
-; RUN: llc -mtriple=arm-eabi -mattr=+neon -mcpu=cortex-a8 %s -o - | FileCheck %s
+; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
+; RUN: llc < %s -mtriple=arm-eabi -mattr=+neon -mcpu=cortex-a8 | FileCheck %s
 
 define <8 x i8> @v_andi8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
-;CHECK-LABEL: v_andi8:
-;CHECK: vand
+; CHECK-LABEL: v_andi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vand d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp2 = load <8 x i8>, <8 x i8>* %B
 	%tmp3 = and <8 x i8> %tmp1, %tmp2
@@ -10,8 +16,13 @@ define <8 x i8> @v_andi8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 }
 
 define <4 x i16> @v_andi16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
-;CHECK-LABEL: v_andi16:
-;CHECK: vand
+; CHECK-LABEL: v_andi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vand d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i16>, <4 x i16>* %A
 	%tmp2 = load <4 x i16>, <4 x i16>* %B
 	%tmp3 = and <4 x i16> %tmp1, %tmp2
@@ -19,8 +30,13 @@ define <4 x i16> @v_andi16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 }
 
 define <2 x i32> @v_andi32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
-;CHECK-LABEL: v_andi32:
-;CHECK: vand
+; CHECK-LABEL: v_andi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vand d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i32>, <2 x i32>* %A
 	%tmp2 = load <2 x i32>, <2 x i32>* %B
 	%tmp3 = and <2 x i32> %tmp1, %tmp2
@@ -28,8 +44,13 @@ define <2 x i32> @v_andi32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 }
 
 define <1 x i64> @v_andi64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
-;CHECK-LABEL: v_andi64:
-;CHECK: vand
+; CHECK-LABEL: v_andi64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vand d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <1 x i64>, <1 x i64>* %A
 	%tmp2 = load <1 x i64>, <1 x i64>* %B
 	%tmp3 = and <1 x i64> %tmp1, %tmp2
@@ -37,8 +58,14 @@ define <1 x i64> @v_andi64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
 }
 
 define <16 x i8> @v_andQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
-;CHECK-LABEL: v_andQi8:
-;CHECK: vand
+; CHECK-LABEL: v_andQi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vand q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp2 = load <16 x i8>, <16 x i8>* %B
 	%tmp3 = and <16 x i8> %tmp1, %tmp2
@@ -46,8 +73,14 @@ define <16 x i8> @v_andQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 }
 
 define <8 x i16> @v_andQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
-;CHECK-LABEL: v_andQi16:
-;CHECK: vand
+; CHECK-LABEL: v_andQi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vand q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i16>, <8 x i16>* %A
 	%tmp2 = load <8 x i16>, <8 x i16>* %B
 	%tmp3 = and <8 x i16> %tmp1, %tmp2
@@ -55,8 +88,14 @@ define <8 x i16> @v_andQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 }
 
 define <4 x i32> @v_andQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
-;CHECK-LABEL: v_andQi32:
-;CHECK: vand
+; CHECK-LABEL: v_andQi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vand q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i32>, <4 x i32>* %A
 	%tmp2 = load <4 x i32>, <4 x i32>* %B
 	%tmp3 = and <4 x i32> %tmp1, %tmp2
@@ -64,8 +103,14 @@ define <4 x i32> @v_andQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 }
 
 define <2 x i64> @v_andQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
-;CHECK-LABEL: v_andQi64:
-;CHECK: vand
+; CHECK-LABEL: v_andQi64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vand q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i64>, <2 x i64>* %A
 	%tmp2 = load <2 x i64>, <2 x i64>* %B
 	%tmp3 = and <2 x i64> %tmp1, %tmp2
@@ -73,8 +118,13 @@ define <2 x i64> @v_andQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
 }
 
 define <8 x i8> @v_bici8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
-;CHECK-LABEL: v_bici8:
-;CHECK: vbic
+; CHECK-LABEL: v_bici8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vbic d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp2 = load <8 x i8>, <8 x i8>* %B
 	%tmp3 = xor <8 x i8> %tmp2, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
@@ -83,8 +133,13 @@ define <8 x i8> @v_bici8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 }
 
 define <4 x i16> @v_bici16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
-;CHECK-LABEL: v_bici16:
-;CHECK: vbic
+; CHECK-LABEL: v_bici16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vbic d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i16>, <4 x i16>* %A
 	%tmp2 = load <4 x i16>, <4 x i16>* %B
 	%tmp3 = xor <4 x i16> %tmp2, < i16 -1, i16 -1, i16 -1, i16 -1 >
@@ -93,8 +148,13 @@ define <4 x i16> @v_bici16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 }
 
 define <2 x i32> @v_bici32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
-;CHECK-LABEL: v_bici32:
-;CHECK: vbic
+; CHECK-LABEL: v_bici32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vbic d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i32>, <2 x i32>* %A
 	%tmp2 = load <2 x i32>, <2 x i32>* %B
 	%tmp3 = xor <2 x i32> %tmp2, < i32 -1, i32 -1 >
@@ -103,8 +163,13 @@ define <2 x i32> @v_bici32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 }
 
 define <1 x i64> @v_bici64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
-;CHECK-LABEL: v_bici64:
-;CHECK: vbic
+; CHECK-LABEL: v_bici64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vbic d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <1 x i64>, <1 x i64>* %A
 	%tmp2 = load <1 x i64>, <1 x i64>* %B
 	%tmp3 = xor <1 x i64> %tmp2, < i64 -1 >
@@ -113,8 +178,14 @@ define <1 x i64> @v_bici64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
 }
 
 define <16 x i8> @v_bicQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
-;CHECK-LABEL: v_bicQi8:
-;CHECK: vbic
+; CHECK-LABEL: v_bicQi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vbic q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp2 = load <16 x i8>, <16 x i8>* %B
 	%tmp3 = xor <16 x i8> %tmp2, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
@@ -123,8 +194,14 @@ define <16 x i8> @v_bicQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 }
 
 define <8 x i16> @v_bicQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
-;CHECK-LABEL: v_bicQi16:
-;CHECK: vbic
+; CHECK-LABEL: v_bicQi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vbic q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i16>, <8 x i16>* %A
 	%tmp2 = load <8 x i16>, <8 x i16>* %B
 	%tmp3 = xor <8 x i16> %tmp2, < i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1 >
@@ -133,8 +210,14 @@ define <8 x i16> @v_bicQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 }
 
 define <4 x i32> @v_bicQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
-;CHECK-LABEL: v_bicQi32:
-;CHECK: vbic
+; CHECK-LABEL: v_bicQi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vbic q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i32>, <4 x i32>* %A
 	%tmp2 = load <4 x i32>, <4 x i32>* %B
 	%tmp3 = xor <4 x i32> %tmp2, < i32 -1, i32 -1, i32 -1, i32 -1 >
@@ -143,8 +226,14 @@ define <4 x i32> @v_bicQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 }
 
 define <2 x i64> @v_bicQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
-;CHECK-LABEL: v_bicQi64:
-;CHECK: vbic
+; CHECK-LABEL: v_bicQi64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vbic q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i64>, <2 x i64>* %A
 	%tmp2 = load <2 x i64>, <2 x i64>* %B
 	%tmp3 = xor <2 x i64> %tmp2, < i64 -1, i64 -1 >
@@ -153,8 +242,13 @@ define <2 x i64> @v_bicQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
 }
 
 define <8 x i8> @v_eori8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
-;CHECK-LABEL: v_eori8:
-;CHECK: veor
+; CHECK-LABEL: v_eori8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    veor d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp2 = load <8 x i8>, <8 x i8>* %B
 	%tmp3 = xor <8 x i8> %tmp1, %tmp2
@@ -162,8 +256,13 @@ define <8 x i8> @v_eori8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 }
 
 define <4 x i16> @v_eori16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
-;CHECK-LABEL: v_eori16:
-;CHECK: veor
+; CHECK-LABEL: v_eori16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    veor d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i16>, <4 x i16>* %A
 	%tmp2 = load <4 x i16>, <4 x i16>* %B
 	%tmp3 = xor <4 x i16> %tmp1, %tmp2
@@ -171,8 +270,13 @@ define <4 x i16> @v_eori16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 }
 
 define <2 x i32> @v_eori32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
-;CHECK-LABEL: v_eori32:
-;CHECK: veor
+; CHECK-LABEL: v_eori32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    veor d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i32>, <2 x i32>* %A
 	%tmp2 = load <2 x i32>, <2 x i32>* %B
 	%tmp3 = xor <2 x i32> %tmp1, %tmp2
@@ -180,8 +284,13 @@ define <2 x i32> @v_eori32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 }
 
 define <1 x i64> @v_eori64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
-;CHECK-LABEL: v_eori64:
-;CHECK: veor
+; CHECK-LABEL: v_eori64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    veor d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <1 x i64>, <1 x i64>* %A
 	%tmp2 = load <1 x i64>, <1 x i64>* %B
 	%tmp3 = xor <1 x i64> %tmp1, %tmp2
@@ -189,8 +298,14 @@ define <1 x i64> @v_eori64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
 }
 
 define <16 x i8> @v_eorQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
-;CHECK-LABEL: v_eorQi8:
-;CHECK: veor
+; CHECK-LABEL: v_eorQi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    veor q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp2 = load <16 x i8>, <16 x i8>* %B
 	%tmp3 = xor <16 x i8> %tmp1, %tmp2
@@ -198,8 +313,14 @@ define <16 x i8> @v_eorQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 }
 
 define <8 x i16> @v_eorQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
-;CHECK-LABEL: v_eorQi16:
-;CHECK: veor
+; CHECK-LABEL: v_eorQi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    veor q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i16>, <8 x i16>* %A
 	%tmp2 = load <8 x i16>, <8 x i16>* %B
 	%tmp3 = xor <8 x i16> %tmp1, %tmp2
@@ -207,8 +328,14 @@ define <8 x i16> @v_eorQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 }
 
 define <4 x i32> @v_eorQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
-;CHECK-LABEL: v_eorQi32:
-;CHECK: veor
+; CHECK-LABEL: v_eorQi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    veor q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i32>, <4 x i32>* %A
 	%tmp2 = load <4 x i32>, <4 x i32>* %B
 	%tmp3 = xor <4 x i32> %tmp1, %tmp2
@@ -216,8 +343,14 @@ define <4 x i32> @v_eorQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 }
 
 define <2 x i64> @v_eorQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
-;CHECK-LABEL: v_eorQi64:
-;CHECK: veor
+; CHECK-LABEL: v_eorQi64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    veor q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i64>, <2 x i64>* %A
 	%tmp2 = load <2 x i64>, <2 x i64>* %B
 	%tmp3 = xor <2 x i64> %tmp1, %tmp2
@@ -225,72 +358,113 @@ define <2 x i64> @v_eorQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
 }
 
 define <8 x i8> @v_mvni8(<8 x i8>* %A) nounwind {
-;CHECK-LABEL: v_mvni8:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvni8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r0]
+; CHECK-NEXT:    vmvn d16, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp2 = xor <8 x i8> %tmp1, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
 	ret <8 x i8> %tmp2
 }
 
 define <4 x i16> @v_mvni16(<4 x i16>* %A) nounwind {
-;CHECK-LABEL: v_mvni16:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvni16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r0]
+; CHECK-NEXT:    vmvn d16, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i16>, <4 x i16>* %A
 	%tmp2 = xor <4 x i16> %tmp1, < i16 -1, i16 -1, i16 -1, i16 -1 >
 	ret <4 x i16> %tmp2
 }
 
 define <2 x i32> @v_mvni32(<2 x i32>* %A) nounwind {
-;CHECK-LABEL: v_mvni32:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvni32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r0]
+; CHECK-NEXT:    vmvn d16, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i32>, <2 x i32>* %A
 	%tmp2 = xor <2 x i32> %tmp1, < i32 -1, i32 -1 >
 	ret <2 x i32> %tmp2
 }
 
 define <1 x i64> @v_mvni64(<1 x i64>* %A) nounwind {
-;CHECK-LABEL: v_mvni64:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvni64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r0]
+; CHECK-NEXT:    vmvn d16, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <1 x i64>, <1 x i64>* %A
 	%tmp2 = xor <1 x i64> %tmp1, < i64 -1 >
 	ret <1 x i64> %tmp2
 }
 
 define <16 x i8> @v_mvnQi8(<16 x i8>* %A) nounwind {
-;CHECK-LABEL: v_mvnQi8:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvnQi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r0]
+; CHECK-NEXT:    vmvn q8, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp2 = xor <16 x i8> %tmp1, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
 	ret <16 x i8> %tmp2
 }
 
 define <8 x i16> @v_mvnQi16(<8 x i16>* %A) nounwind {
-;CHECK-LABEL: v_mvnQi16:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvnQi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r0]
+; CHECK-NEXT:    vmvn q8, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i16>, <8 x i16>* %A
 	%tmp2 = xor <8 x i16> %tmp1, < i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1 >
 	ret <8 x i16> %tmp2
 }
 
 define <4 x i32> @v_mvnQi32(<4 x i32>* %A) nounwind {
-;CHECK-LABEL: v_mvnQi32:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvnQi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r0]
+; CHECK-NEXT:    vmvn q8, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i32>, <4 x i32>* %A
 	%tmp2 = xor <4 x i32> %tmp1, < i32 -1, i32 -1, i32 -1, i32 -1 >
 	ret <4 x i32> %tmp2
 }
 
 define <2 x i64> @v_mvnQi64(<2 x i64>* %A) nounwind {
-;CHECK-LABEL: v_mvnQi64:
-;CHECK: vmvn
+; CHECK-LABEL: v_mvnQi64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r0]
+; CHECK-NEXT:    vmvn q8, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i64>, <2 x i64>* %A
 	%tmp2 = xor <2 x i64> %tmp1, < i64 -1, i64 -1 >
 	ret <2 x i64> %tmp2
 }
 
 define <8 x i8> @v_orri8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
-;CHECK-LABEL: v_orri8:
-;CHECK: vorr
+; CHECK-LABEL: v_orri8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorr d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp2 = load <8 x i8>, <8 x i8>* %B
 	%tmp3 = or <8 x i8> %tmp1, %tmp2
@@ -298,8 +472,13 @@ define <8 x i8> @v_orri8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 }
 
 define <4 x i16> @v_orri16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
-;CHECK-LABEL: v_orri16:
-;CHECK: vorr
+; CHECK-LABEL: v_orri16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorr d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i16>, <4 x i16>* %A
 	%tmp2 = load <4 x i16>, <4 x i16>* %B
 	%tmp3 = or <4 x i16> %tmp1, %tmp2
@@ -307,8 +486,13 @@ define <4 x i16> @v_orri16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 }
 
 define <2 x i32> @v_orri32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
-;CHECK-LABEL: v_orri32:
-;CHECK: vorr
+; CHECK-LABEL: v_orri32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorr d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i32>, <2 x i32>* %A
 	%tmp2 = load <2 x i32>, <2 x i32>* %B
 	%tmp3 = or <2 x i32> %tmp1, %tmp2
@@ -316,8 +500,13 @@ define <2 x i32> @v_orri32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 }
 
 define <1 x i64> @v_orri64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
-;CHECK-LABEL: v_orri64:
-;CHECK: vorr
+; CHECK-LABEL: v_orri64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorr d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <1 x i64>, <1 x i64>* %A
 	%tmp2 = load <1 x i64>, <1 x i64>* %B
 	%tmp3 = or <1 x i64> %tmp1, %tmp2
@@ -325,8 +514,14 @@ define <1 x i64> @v_orri64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
 }
 
 define <16 x i8> @v_orrQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
-;CHECK-LABEL: v_orrQi8:
-;CHECK: vorr
+; CHECK-LABEL: v_orrQi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorr q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp2 = load <16 x i8>, <16 x i8>* %B
 	%tmp3 = or <16 x i8> %tmp1, %tmp2
@@ -334,8 +529,14 @@ define <16 x i8> @v_orrQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 }
 
 define <8 x i16> @v_orrQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
-;CHECK-LABEL: v_orrQi16:
-;CHECK: vorr
+; CHECK-LABEL: v_orrQi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorr q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i16>, <8 x i16>* %A
 	%tmp2 = load <8 x i16>, <8 x i16>* %B
 	%tmp3 = or <8 x i16> %tmp1, %tmp2
@@ -343,8 +544,14 @@ define <8 x i16> @v_orrQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 }
 
 define <4 x i32> @v_orrQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
-;CHECK-LABEL: v_orrQi32:
-;CHECK: vorr
+; CHECK-LABEL: v_orrQi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorr q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i32>, <4 x i32>* %A
 	%tmp2 = load <4 x i32>, <4 x i32>* %B
 	%tmp3 = or <4 x i32> %tmp1, %tmp2
@@ -352,8 +559,14 @@ define <4 x i32> @v_orrQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 }
 
 define <2 x i64> @v_orrQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
-;CHECK-LABEL: v_orrQi64:
-;CHECK: vorr
+; CHECK-LABEL: v_orrQi64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorr q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i64>, <2 x i64>* %A
 	%tmp2 = load <2 x i64>, <2 x i64>* %B
 	%tmp3 = or <2 x i64> %tmp1, %tmp2
@@ -361,8 +574,13 @@ define <2 x i64> @v_orrQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
 }
 
 define <8 x i8> @v_orni8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
-;CHECK-LABEL: v_orni8:
-;CHECK: vorn
+; CHECK-LABEL: v_orni8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorn d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp2 = load <8 x i8>, <8 x i8>* %B
 	%tmp3 = xor <8 x i8> %tmp2, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
@@ -371,8 +589,13 @@ define <8 x i8> @v_orni8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 }
 
 define <4 x i16> @v_orni16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
-;CHECK-LABEL: v_orni16:
-;CHECK: vorn
+; CHECK-LABEL: v_orni16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorn d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i16>, <4 x i16>* %A
 	%tmp2 = load <4 x i16>, <4 x i16>* %B
 	%tmp3 = xor <4 x i16> %tmp2, < i16 -1, i16 -1, i16 -1, i16 -1 >
@@ -381,8 +604,13 @@ define <4 x i16> @v_orni16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 }
 
 define <2 x i32> @v_orni32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
-;CHECK-LABEL: v_orni32:
-;CHECK: vorn
+; CHECK-LABEL: v_orni32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorn d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i32>, <2 x i32>* %A
 	%tmp2 = load <2 x i32>, <2 x i32>* %B
 	%tmp3 = xor <2 x i32> %tmp2, < i32 -1, i32 -1 >
@@ -391,8 +619,13 @@ define <2 x i32> @v_orni32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 }
 
 define <1 x i64> @v_orni64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
-;CHECK-LABEL: v_orni64:
-;CHECK: vorn
+; CHECK-LABEL: v_orni64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vorn d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <1 x i64>, <1 x i64>* %A
 	%tmp2 = load <1 x i64>, <1 x i64>* %B
 	%tmp3 = xor <1 x i64> %tmp2, < i64 -1 >
@@ -401,8 +634,14 @@ define <1 x i64> @v_orni64(<1 x i64>* %A, <1 x i64>* %B) nounwind {
 }
 
 define <16 x i8> @v_ornQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
-;CHECK-LABEL: v_ornQi8:
-;CHECK: vorn
+; CHECK-LABEL: v_ornQi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorn q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp2 = load <16 x i8>, <16 x i8>* %B
 	%tmp3 = xor <16 x i8> %tmp2, < i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1 >
@@ -411,8 +650,14 @@ define <16 x i8> @v_ornQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 }
 
 define <8 x i16> @v_ornQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
-;CHECK-LABEL: v_ornQi16:
-;CHECK: vorn
+; CHECK-LABEL: v_ornQi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorn q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i16>, <8 x i16>* %A
 	%tmp2 = load <8 x i16>, <8 x i16>* %B
 	%tmp3 = xor <8 x i16> %tmp2, < i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1 >
@@ -421,8 +666,14 @@ define <8 x i16> @v_ornQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 }
 
 define <4 x i32> @v_ornQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
-;CHECK-LABEL: v_ornQi32:
-;CHECK: vorn
+; CHECK-LABEL: v_ornQi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorn q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i32>, <4 x i32>* %A
 	%tmp2 = load <4 x i32>, <4 x i32>* %B
 	%tmp3 = xor <4 x i32> %tmp2, < i32 -1, i32 -1, i32 -1, i32 -1 >
@@ -431,8 +682,14 @@ define <4 x i32> @v_ornQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 }
 
 define <2 x i64> @v_ornQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
-;CHECK-LABEL: v_ornQi64:
-;CHECK: vorn
+; CHECK-LABEL: v_ornQi64:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vorn q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i64>, <2 x i64>* %A
 	%tmp2 = load <2 x i64>, <2 x i64>* %B
 	%tmp3 = xor <2 x i64> %tmp2, < i64 -1, i64 -1 >
@@ -441,8 +698,13 @@ define <2 x i64> @v_ornQi64(<2 x i64>* %A, <2 x i64>* %B) nounwind {
 }
 
 define <8 x i8> @vtsti8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
-;CHECK-LABEL: vtsti8:
-;CHECK: vtst.8
+; CHECK-LABEL: vtsti8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vtst.8 d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp2 = load <8 x i8>, <8 x i8>* %B
 	%tmp3 = and <8 x i8> %tmp1, %tmp2
@@ -452,8 +714,13 @@ define <8 x i8> @vtsti8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
 }
 
 define <4 x i16> @vtsti16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
-;CHECK-LABEL: vtsti16:
-;CHECK: vtst.16
+; CHECK-LABEL: vtsti16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vtst.16 d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i16>, <4 x i16>* %A
 	%tmp2 = load <4 x i16>, <4 x i16>* %B
 	%tmp3 = and <4 x i16> %tmp1, %tmp2
@@ -463,8 +730,13 @@ define <4 x i16> @vtsti16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
 }
 
 define <2 x i32> @vtsti32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
-;CHECK-LABEL: vtsti32:
-;CHECK: vtst.32
+; CHECK-LABEL: vtsti32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r1]
+; CHECK-NEXT:    vldr d17, [r0]
+; CHECK-NEXT:    vtst.32 d16, d17, d16
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <2 x i32>, <2 x i32>* %A
 	%tmp2 = load <2 x i32>, <2 x i32>* %B
 	%tmp3 = and <2 x i32> %tmp1, %tmp2
@@ -474,8 +746,14 @@ define <2 x i32> @vtsti32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
 }
 
 define <16 x i8> @vtstQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
-;CHECK-LABEL: vtstQi8:
-;CHECK: vtst.8
+; CHECK-LABEL: vtstQi8:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vtst.8 q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp2 = load <16 x i8>, <16 x i8>* %B
 	%tmp3 = and <16 x i8> %tmp1, %tmp2
@@ -485,8 +763,14 @@ define <16 x i8> @vtstQi8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
 }
 
 define <8 x i16> @vtstQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
-;CHECK-LABEL: vtstQi16:
-;CHECK: vtst.16
+; CHECK-LABEL: vtstQi16:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vtst.16 q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i16>, <8 x i16>* %A
 	%tmp2 = load <8 x i16>, <8 x i16>* %B
 	%tmp3 = and <8 x i16> %tmp1, %tmp2
@@ -496,8 +780,14 @@ define <8 x i16> @vtstQi16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
 }
 
 define <4 x i32> @vtstQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
-;CHECK-LABEL: vtstQi32:
-;CHECK: vtst.32
+; CHECK-LABEL: vtstQi32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
+; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
+; CHECK-NEXT:    vtst.32 q8, q9, q8
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <4 x i32>, <4 x i32>* %A
 	%tmp2 = load <4 x i32>, <4 x i32>* %B
 	%tmp3 = and <4 x i32> %tmp1, %tmp2
@@ -508,19 +798,24 @@ define <4 x i32> @vtstQi32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
 
 define <8 x i8> @v_orrimm(<8 x i8>* %A) nounwind {
 ; CHECK-LABEL: v_orrimm:
-; CHECK-NOT: vmov
-; CHECK-NOT: vmvn
-; CHECK: vorr
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r0]
+; CHECK-NEXT:    vorr.i32 d16, #0x1000000
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp3 = or <8 x i8> %tmp1, <i8 0, i8 0, i8 0, i8 1, i8 0, i8 0, i8 0, i8 1>
 	ret <8 x i8> %tmp3
 }
 
 define <16 x i8> @v_orrimmQ(<16 x i8>* %A) nounwind {
-; CHECK: v_orrimmQ
-; CHECK-NOT: vmov
-; CHECK-NOT: vmvn
-; CHECK: vorr
+; CHECK-LABEL: v_orrimmQ:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r0]
+; CHECK-NEXT:    vorr.i32 q8, #0x1000000
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp3 = or <16 x i8> %tmp1, <i8 0, i8 0, i8 0, i8 1, i8 0, i8 0, i8 0, i8 1, i8 0, i8 0, i8 0, i8 1, i8 0, i8 0, i8 0, i8 1>
 	ret <16 x i8> %tmp3
@@ -528,9 +823,11 @@ define <16 x i8> @v_orrimmQ(<16 x i8>* %A) nounwind {
 
 define <8 x i8> @v_bicimm(<8 x i8>* %A) nounwind {
 ; CHECK-LABEL: v_bicimm:
-; CHECK-NOT: vmov
-; CHECK-NOT: vmvn
-; CHECK: vbic
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vldr d16, [r0]
+; CHECK-NEXT:    vbic.i32 d16, #0xff000000
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <8 x i8>, <8 x i8>* %A
 	%tmp3 = and <8 x i8> %tmp1, < i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0 >
 	ret <8 x i8> %tmp3
@@ -538,10 +835,31 @@ define <8 x i8> @v_bicimm(<8 x i8>* %A) nounwind {
 
 define <16 x i8> @v_bicimmQ(<16 x i8>* %A) nounwind {
 ; CHECK-LABEL: v_bicimmQ:
-; CHECK-NOT: vmov
-; CHECK-NOT: vmvn
-; CHECK: vbic
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vld1.64 {d16, d17}, [r0]
+; CHECK-NEXT:    vbic.i32 q8, #0xff000000
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
 	%tmp1 = load <16 x i8>, <16 x i8>* %A
 	%tmp3 = and <16 x i8> %tmp1, < i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0, i8 -1, i8 -1, i8 -1, i8 0 >
 	ret <16 x i8> %tmp3
 }
+
+define <4 x i32> @hidden_not_v4i32(<4 x i32> %x) nounwind {
+; CHECK-LABEL: hidden_not_v4i32:
+; CHECK:       @ BB#0:
+; CHECK-NEXT:    vmov d19, r2, r3
+; CHECK-NEXT:    vmov.i32 q8, #0xf
+; CHECK-NEXT:    vmov d18, r0, r1
+; CHECK-NEXT:    vmov.i32 q10, #0x6
+; CHECK-NEXT:    veor q8, q9, q8
+; CHECK-NEXT:    vand q8, q8, q10
+; CHECK-NEXT:    vmov r0, r1, d16
+; CHECK-NEXT:    vmov r2, r3, d17
+; CHECK-NEXT:    bx lr
+  %xor = xor <4 x i32> %x, <i32 15, i32 15, i32 15, i32 15>
+  %and = and <4 x i32> %xor, <i32 6, i32 6, i32 6, i32 6>
+  ret <4 x i32> %and
+}
+