AArch64/ARM64: port more AArch64 tests to ARM64.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@206592 91177308-0d34-0410-b5e6-96231b3b80d8

2 files changed, 797 insertions, 1901 deletions

diff --git a/test/CodeGen/ARM64/aarch64-neon-misc.ll b/test/CodeGen/ARM64/aarch64-neon-misc.ll
deleted file mode 100644
index 718ac79830..0000000000
--- a/test/CodeGen/ARM64/aarch64-neon-misc.ll
+++ /dev/null
@@ -1,1901 +0,0 @@
-; RUN: llc < %s -verify-machineinstrs -mtriple=arm64-none-linux-gnu -mattr=+neon -fp-contract=fast | FileCheck %s
-
-define <8 x i8> @test_vrev16_s8(<8 x i8> %a) #0 {
-; CHECK: rev16 v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %shuffle.i = shufflevector <8 x i8> %a, <8 x i8> undef, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6>
-  ret <8 x i8> %shuffle.i
-}
-
-define <16 x i8> @test_vrev16q_s8(<16 x i8> %a) #0 {
-; CHECK: rev16 v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %shuffle.i = shufflevector <16 x i8> %a, <16 x i8> undef, <16 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6, i32 9, i32 8, i32 11, i32 10, i32 13, i32 12, i32 15, i32 14>
-  ret <16 x i8> %shuffle.i
-}
-
-define <8 x i8> @test_vrev32_s8(<8 x i8> %a) #0 {
-; CHECK: rev32 v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %shuffle.i = shufflevector <8 x i8> %a, <8 x i8> undef, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4>
-  ret <8 x i8> %shuffle.i
-}
-
-define <4 x i16> @test_vrev32_s16(<4 x i16> %a) #0 {
-; CHECK: rev32 v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %shuffle.i = shufflevector <4 x i16> %a, <4 x i16> undef, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
-  ret <4 x i16> %shuffle.i
-}
-
-define <16 x i8> @test_vrev32q_s8(<16 x i8> %a) #0 {
-; CHECK: rev32 v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %shuffle.i = shufflevector <16 x i8> %a, <16 x i8> undef, <16 x i32> <i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4, i32 11, i32 10, i32 9, i32 8, i32 15, i32 14, i32 13, i32 12>
-  ret <16 x i8> %shuffle.i
-}
-
-define <8 x i16> @test_vrev32q_s16(<8 x i16> %a) #0 {
-; CHECK: rev32 v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <8 x i32> <i32 1, i32 0, i32 3, i32 2, i32 5, i32 4, i32 7, i32 6>
-  ret <8 x i16> %shuffle.i
-}
-
-define <8 x i8> @test_vrev64_s8(<8 x i8> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %shuffle.i = shufflevector <8 x i8> %a, <8 x i8> undef, <8 x i32> <i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0>
-  ret <8 x i8> %shuffle.i
-}
-
-define <4 x i16> @test_vrev64_s16(<4 x i16> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %shuffle.i = shufflevector <4 x i16> %a, <4 x i16> undef, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
-  ret <4 x i16> %shuffle.i
-}
-
-define <2 x i32> @test_vrev64_s32(<2 x i32> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %shuffle.i = shufflevector <2 x i32> %a, <2 x i32> undef, <2 x i32> <i32 1, i32 0>
-  ret <2 x i32> %shuffle.i
-}
-
-define <2 x float> @test_vrev64_f32(<2 x float> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %shuffle.i = shufflevector <2 x float> %a, <2 x float> undef, <2 x i32> <i32 1, i32 0>
-  ret <2 x float> %shuffle.i
-}
-
-define <16 x i8> @test_vrev64q_s8(<16 x i8> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %shuffle.i = shufflevector <16 x i8> %a, <16 x i8> undef, <16 x i32> <i32 7, i32 6, i32 5, i32 4, i32 3, i32 2, i32 1, i32 0, i32 15, i32 14, i32 13, i32 12, i32 11, i32 10, i32 9, i32 8>
-  ret <16 x i8> %shuffle.i
-}
-
-define <8 x i16> @test_vrev64q_s16(<8 x i16> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <8 x i32> <i32 3, i32 2, i32 1, i32 0, i32 7, i32 6, i32 5, i32 4>
-  ret <8 x i16> %shuffle.i
-}
-
-define <4 x i32> @test_vrev64q_s32(<4 x i32> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
-  ret <4 x i32> %shuffle.i
-}
-
-define <4 x float> @test_vrev64q_f32(<4 x float> %a) #0 {
-; CHECK: rev64 v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %shuffle.i = shufflevector <4 x float> %a, <4 x float> undef, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
-  ret <4 x float> %shuffle.i
-}
-
-define <4 x i16> @test_vpaddl_s8(<8 x i8> %a) #0 {
-; CHECK: saddlp v{{[0-9]+}}.4h, v{{[0-9]+}}.8b
-  %vpaddl.i = tail call <4 x i16> @llvm.arm64.neon.saddlp.v4i16.v8i8(<8 x i8> %a) #4
-  ret <4 x i16> %vpaddl.i
-}
-
-define <2 x i32> @test_vpaddl_s16(<4 x i16> %a) #0 {
-; CHECK: saddlp v{{[0-9]+}}.2s, v{{[0-9]+}}.4h
-  %vpaddl1.i = tail call <2 x i32> @llvm.arm64.neon.saddlp.v2i32.v4i16(<4 x i16> %a) #4
-  ret <2 x i32> %vpaddl1.i
-}
-
-define <1 x i64> @test_vpaddl_s32(<2 x i32> %a) #0 {
-; CHECK: saddlp v{{[0-9]+}}.1d, v{{[0-9]+}}.2s
-  %vpaddl1.i = tail call <1 x i64> @llvm.arm64.neon.saddlp.v1i64.v2i32(<2 x i32> %a) #4
-  ret <1 x i64> %vpaddl1.i
-}
-
-define <4 x i16> @test_vpaddl_u8(<8 x i8> %a) #0 {
-; CHECK: uaddlp v{{[0-9]+}}.4h, v{{[0-9]+}}.8b
-  %vpaddl.i = tail call <4 x i16> @llvm.arm64.neon.uaddlp.v4i16.v8i8(<8 x i8> %a) #4
-  ret <4 x i16> %vpaddl.i
-}
-
-define <2 x i32> @test_vpaddl_u16(<4 x i16> %a) #0 {
-; CHECK: uaddlp v{{[0-9]+}}.2s, v{{[0-9]+}}.4h
-  %vpaddl1.i = tail call <2 x i32> @llvm.arm64.neon.uaddlp.v2i32.v4i16(<4 x i16> %a) #4
-  ret <2 x i32> %vpaddl1.i
-}
-
-define <1 x i64> @test_vpaddl_u32(<2 x i32> %a) #0 {
-; CHECK: uaddlp v{{[0-9]+}}.1d, v{{[0-9]+}}.2s
-  %vpaddl1.i = tail call <1 x i64> @llvm.arm64.neon.uaddlp.v1i64.v2i32(<2 x i32> %a) #4
-  ret <1 x i64> %vpaddl1.i
-}
-
-define <8 x i16> @test_vpaddlq_s8(<16 x i8> %a) #0 {
-; CHECK: saddlp v{{[0-9]+}}.8h, v{{[0-9]+}}.16b
-  %vpaddl.i = tail call <8 x i16> @llvm.arm64.neon.saddlp.v8i16.v16i8(<16 x i8> %a) #4
-  ret <8 x i16> %vpaddl.i
-}
-
-define <4 x i32> @test_vpaddlq_s16(<8 x i16> %a) #0 {
-; CHECK: saddlp v{{[0-9]+}}.4s, v{{[0-9]+}}.8h
-  %vpaddl1.i = tail call <4 x i32> @llvm.arm64.neon.saddlp.v4i32.v8i16(<8 x i16> %a) #4
-  ret <4 x i32> %vpaddl1.i
-}
-
-define <2 x i64> @test_vpaddlq_s32(<4 x i32> %a) #0 {
-; CHECK: saddlp v{{[0-9]+}}.2d, v{{[0-9]+}}.4s
-  %vpaddl1.i = tail call <2 x i64> @llvm.arm64.neon.saddlp.v2i64.v4i32(<4 x i32> %a) #4
-  ret <2 x i64> %vpaddl1.i
-}
-
-define <8 x i16> @test_vpaddlq_u8(<16 x i8> %a) #0 {
-; CHECK: uaddlp v{{[0-9]+}}.8h, v{{[0-9]+}}.16b
-  %vpaddl.i = tail call <8 x i16> @llvm.arm64.neon.uaddlp.v8i16.v16i8(<16 x i8> %a) #4
-  ret <8 x i16> %vpaddl.i
-}
-
-define <4 x i32> @test_vpaddlq_u16(<8 x i16> %a) #0 {
-; CHECK: uaddlp v{{[0-9]+}}.4s, v{{[0-9]+}}.8h
-  %vpaddl1.i = tail call <4 x i32> @llvm.arm64.neon.uaddlp.v4i32.v8i16(<8 x i16> %a) #4
-  ret <4 x i32> %vpaddl1.i
-}
-
-define <2 x i64> @test_vpaddlq_u32(<4 x i32> %a) #0 {
-; CHECK: uaddlp v{{[0-9]+}}.2d, v{{[0-9]+}}.4s
-  %vpaddl1.i = tail call <2 x i64> @llvm.arm64.neon.uaddlp.v2i64.v4i32(<4 x i32> %a) #4
-  ret <2 x i64> %vpaddl1.i
-}
-
-define <4 x i16> @test_vpadal_s8(<4 x i16> %a, <8 x i8> %b) #0 {
-; CHECK: sadalp v{{[0-9]+}}.4h, v{{[0-9]+}}.8b
-  %vpadal1.i = tail call <4 x i16> @llvm.arm64.neon.saddlp.v4i16.v8i8(<8 x i8> %b) #4
-  %sum = add <4 x i16> %a, %vpadal1.i
-  ret <4 x i16> %sum
-}
-
-define <2 x i32> @test_vpadal_s16(<2 x i32> %a, <4 x i16> %b) #0 {
-; CHECK: sadalp v{{[0-9]+}}.2s, v{{[0-9]+}}.4h
-  %vpadal2.i = tail call <2 x i32> @llvm.arm64.neon.saddlp.v2i32.v4i16(<4 x i16> %b) #4
-  %sum = add <2 x i32> %a, %vpadal2.i
-  ret <2 x i32> %sum
-}
-
-define <1 x i64> @test_vpadal_s32(<1 x i64> %a, <2 x i32> %b) #0 {
-; CHECK: sadalp v{{[0-9]+}}.1d, v{{[0-9]+}}.2s
-  %vpadal2.i = tail call <1 x i64> @llvm.arm64.neon.saddlp.v1i64.v2i32(<2 x i32> %b) #4
-  %sum = add <1 x i64> %a, %vpadal2.i
-  ret <1 x i64> %sum
-}
-
-define <4 x i16> @test_vpadal_u8(<4 x i16> %a, <8 x i8> %b) #0 {
-; CHECK: uadalp v{{[0-9]+}}.4h, v{{[0-9]+}}.8b
-  %vpadal1.i = tail call <4 x i16> @llvm.arm64.neon.uaddlp.v4i16.v8i8(<8 x i8> %b) #4
-  %sum = add <4 x i16> %a, %vpadal1.i
-  ret <4 x i16> %sum
-}
-
-define <2 x i32> @test_vpadal_u16(<2 x i32> %a, <4 x i16> %b) #0 {
-; CHECK: uadalp v{{[0-9]+}}.2s, v{{[0-9]+}}.4h
-  %vpadal2.i = tail call <2 x i32> @llvm.arm64.neon.uaddlp.v2i32.v4i16(<4 x i16> %b) #4
-  %sum = add <2 x i32> %a, %vpadal2.i
-  ret <2 x i32> %sum
-}
-
-define <1 x i64> @test_vpadal_u32(<1 x i64> %a, <2 x i32> %b) #0 {
-; CHECK: uadalp v{{[0-9]+}}.1d, v{{[0-9]+}}.2s
-  %vpadal2.i = tail call <1 x i64> @llvm.arm64.neon.uaddlp.v1i64.v2i32(<2 x i32> %b) #4
-  %sum = add <1 x i64> %a, %vpadal2.i
-  ret <1 x i64> %sum
-}
-
-define <8 x i16> @test_vpadalq_s8(<8 x i16> %a, <16 x i8> %b) #0 {
-; CHECK: sadalp v{{[0-9]+}}.8h, v{{[0-9]+}}.16b
-  %vpadal1.i = tail call <8 x i16> @llvm.arm64.neon.saddlp.v8i16.v16i8(<16 x i8> %b) #4
-  %sum = add <8 x i16> %a, %vpadal1.i
-  ret <8 x i16> %sum
-}
-
-define <4 x i32> @test_vpadalq_s16(<4 x i32> %a, <8 x i16> %b) #0 {
-; CHECK: sadalp v{{[0-9]+}}.4s, v{{[0-9]+}}.8h
-  %vpadal2.i = tail call <4 x i32> @llvm.arm64.neon.saddlp.v4i32.v8i16(<8 x i16> %b) #4
-  %sum = add <4 x i32> %a, %vpadal2.i
-  ret <4 x i32> %sum
-}
-
-define <2 x i64> @test_vpadalq_s32(<2 x i64> %a, <4 x i32> %b) #0 {
-; CHECK: sadalp v{{[0-9]+}}.2d, v{{[0-9]+}}.4s
-  %vpadal2.i = tail call <2 x i64> @llvm.arm64.neon.saddlp.v2i64.v4i32(<4 x i32> %b) #4
-  %sum = add <2 x i64> %a, %vpadal2.i
-  ret <2 x i64> %sum
-}
-
-define <8 x i16> @test_vpadalq_u8(<8 x i16> %a, <16 x i8> %b) #0 {
-; CHECK: uadalp v{{[0-9]+}}.8h, v{{[0-9]+}}.16b
-  %vpadal1.i = tail call <8 x i16> @llvm.arm64.neon.uaddlp.v8i16.v16i8(<16 x i8> %b) #4
-  %sum = add <8 x i16> %a, %vpadal1.i
-  ret <8 x i16> %sum
-}
-
-define <4 x i32> @test_vpadalq_u16(<4 x i32> %a, <8 x i16> %b) #0 {
-; CHECK: uadalp v{{[0-9]+}}.4s, v{{[0-9]+}}.8h
-  %vpadal2.i = tail call <4 x i32> @llvm.arm64.neon.uaddlp.v4i32.v8i16(<8 x i16> %b) #4
-  %sum = add <4 x i32> %a, %vpadal2.i
-  ret <4 x i32> %sum
-}
-
-define <2 x i64> @test_vpadalq_u32(<2 x i64> %a, <4 x i32> %b) #0 {
-; CHECK: uadalp v{{[0-9]+}}.2d, v{{[0-9]+}}.4s
-  %vpadal2.i = tail call <2 x i64> @llvm.arm64.neon.uaddlp.v2i64.v4i32(<4 x i32> %b) #4
-  %sum = add <2 x i64> %a, %vpadal2.i
-  ret <2 x i64> %sum
-}
-
-define <8 x i8> @test_vqabs_s8(<8 x i8> %a) #0 {
-; CHECK: sqabs v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vqabs.i = tail call <8 x i8> @llvm.arm64.neon.sqabs.v8i8(<8 x i8> %a) #4
-  ret <8 x i8> %vqabs.i
-}
-
-define <16 x i8> @test_vqabsq_s8(<16 x i8> %a) #0 {
-; CHECK: sqabs v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vqabs.i = tail call <16 x i8> @llvm.arm64.neon.sqabs.v16i8(<16 x i8> %a) #4
-  ret <16 x i8> %vqabs.i
-}
-
-define <4 x i16> @test_vqabs_s16(<4 x i16> %a) #0 {
-; CHECK: sqabs v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %vqabs1.i = tail call <4 x i16> @llvm.arm64.neon.sqabs.v4i16(<4 x i16> %a) #4
-  ret <4 x i16> %vqabs1.i
-}
-
-define <8 x i16> @test_vqabsq_s16(<8 x i16> %a) #0 {
-; CHECK: sqabs v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %vqabs1.i = tail call <8 x i16> @llvm.arm64.neon.sqabs.v8i16(<8 x i16> %a) #4
-  ret <8 x i16> %vqabs1.i
-}
-
-define <2 x i32> @test_vqabs_s32(<2 x i32> %a) #0 {
-; CHECK: sqabs v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vqabs1.i = tail call <2 x i32> @llvm.arm64.neon.sqabs.v2i32(<2 x i32> %a) #4
-  ret <2 x i32> %vqabs1.i
-}
-
-define <4 x i32> @test_vqabsq_s32(<4 x i32> %a) #0 {
-; CHECK: sqabs v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vqabs1.i = tail call <4 x i32> @llvm.arm64.neon.sqabs.v4i32(<4 x i32> %a) #4
-  ret <4 x i32> %vqabs1.i
-}
-
-define <2 x i64> @test_vqabsq_s64(<2 x i64> %a) #0 {
-; CHECK: sqabs v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vqabs1.i = tail call <2 x i64> @llvm.arm64.neon.sqabs.v2i64(<2 x i64> %a) #4
-  ret <2 x i64> %vqabs1.i
-}
-
-define <8 x i8> @test_vqneg_s8(<8 x i8> %a) #0 {
-; CHECK: sqneg v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vqneg.i = tail call <8 x i8> @llvm.arm64.neon.sqneg.v8i8(<8 x i8> %a) #4
-  ret <8 x i8> %vqneg.i
-}
-
-define <16 x i8> @test_vqnegq_s8(<16 x i8> %a) #0 {
-; CHECK: sqneg v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vqneg.i = tail call <16 x i8> @llvm.arm64.neon.sqneg.v16i8(<16 x i8> %a) #4
-  ret <16 x i8> %vqneg.i
-}
-
-define <4 x i16> @test_vqneg_s16(<4 x i16> %a) #0 {
-; CHECK: sqneg v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %vqneg1.i = tail call <4 x i16> @llvm.arm64.neon.sqneg.v4i16(<4 x i16> %a) #4
-  ret <4 x i16> %vqneg1.i
-}
-
-define <8 x i16> @test_vqnegq_s16(<8 x i16> %a) #0 {
-; CHECK: sqneg v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %vqneg1.i = tail call <8 x i16> @llvm.arm64.neon.sqneg.v8i16(<8 x i16> %a) #4
-  ret <8 x i16> %vqneg1.i
-}
-
-define <2 x i32> @test_vqneg_s32(<2 x i32> %a) #0 {
-; CHECK: sqneg v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vqneg1.i = tail call <2 x i32> @llvm.arm64.neon.sqneg.v2i32(<2 x i32> %a) #4
-  ret <2 x i32> %vqneg1.i
-}
-
-define <4 x i32> @test_vqnegq_s32(<4 x i32> %a) #0 {
-; CHECK: sqneg v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vqneg1.i = tail call <4 x i32> @llvm.arm64.neon.sqneg.v4i32(<4 x i32> %a) #4
-  ret <4 x i32> %vqneg1.i
-}
-
-define <2 x i64> @test_vqnegq_s64(<2 x i64> %a) #0 {
-; CHECK: sqneg v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vqneg1.i = tail call <2 x i64> @llvm.arm64.neon.sqneg.v2i64(<2 x i64> %a) #4
-  ret <2 x i64> %vqneg1.i
-}
-
-define <8 x i8> @test_vneg_s8(<8 x i8> %a) #0 {
-; CHECK: neg v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %sub.i = sub <8 x i8> zeroinitializer, %a
-  ret <8 x i8> %sub.i
-}
-
-define <16 x i8> @test_vnegq_s8(<16 x i8> %a) #0 {
-; CHECK: neg v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %sub.i = sub <16 x i8> zeroinitializer, %a
-  ret <16 x i8> %sub.i
-}
-
-define <4 x i16> @test_vneg_s16(<4 x i16> %a) #0 {
-; CHECK: neg v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %sub.i = sub <4 x i16> zeroinitializer, %a
-  ret <4 x i16> %sub.i
-}
-
-define <8 x i16> @test_vnegq_s16(<8 x i16> %a) #0 {
-; CHECK: neg v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %sub.i = sub <8 x i16> zeroinitializer, %a
-  ret <8 x i16> %sub.i
-}
-
-define <2 x i32> @test_vneg_s32(<2 x i32> %a) #0 {
-; CHECK: neg v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %sub.i = sub <2 x i32> zeroinitializer, %a
-  ret <2 x i32> %sub.i
-}
-
-define <4 x i32> @test_vnegq_s32(<4 x i32> %a) #0 {
-; CHECK: neg v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %sub.i = sub <4 x i32> zeroinitializer, %a
-  ret <4 x i32> %sub.i
-}
-
-define <2 x i64> @test_vnegq_s64(<2 x i64> %a) #0 {
-; CHECK: neg v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %sub.i = sub <2 x i64> zeroinitializer, %a
-  ret <2 x i64> %sub.i
-}
-
-define <2 x float> @test_vneg_f32(<2 x float> %a) #0 {
-; CHECK: fneg v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %sub.i = fsub <2 x float> <float -0.000000e+00, float -0.000000e+00>, %a
-  ret <2 x float> %sub.i
-}
-
-define <4 x float> @test_vnegq_f32(<4 x float> %a) #0 {
-; CHECK: fneg v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %sub.i = fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %a
-  ret <4 x float> %sub.i
-}
-
-define <2 x double> @test_vnegq_f64(<2 x double> %a) #0 {
-; CHECK: fneg v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %sub.i = fsub <2 x double> <double -0.000000e+00, double -0.000000e+00>, %a
-  ret <2 x double> %sub.i
-}
-
-define <8 x i8> @test_vabs_s8(<8 x i8> %a) #0 {
-; CHECK: abs v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vabs.i = tail call <8 x i8> @llvm.arm64.neon.abs.v8i8(<8 x i8> %a) #4
-  ret <8 x i8> %vabs.i
-}
-
-define <16 x i8> @test_vabsq_s8(<16 x i8> %a) #0 {
-; CHECK: abs v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vabs.i = tail call <16 x i8> @llvm.arm64.neon.abs.v16i8(<16 x i8> %a) #4
-  ret <16 x i8> %vabs.i
-}
-
-define <4 x i16> @test_vabs_s16(<4 x i16> %a) #0 {
-; CHECK: abs v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %vabs1.i = tail call <4 x i16> @llvm.arm64.neon.abs.v4i16(<4 x i16> %a) #4
-  ret <4 x i16> %vabs1.i
-}
-
-define <8 x i16> @test_vabsq_s16(<8 x i16> %a) #0 {
-; CHECK: abs v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %vabs1.i = tail call <8 x i16> @llvm.arm64.neon.abs.v8i16(<8 x i16> %a) #4
-  ret <8 x i16> %vabs1.i
-}
-
-define <2 x i32> @test_vabs_s32(<2 x i32> %a) #0 {
-; CHECK: abs v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vabs1.i = tail call <2 x i32> @llvm.arm64.neon.abs.v2i32(<2 x i32> %a) #4
-  ret <2 x i32> %vabs1.i
-}
-
-define <4 x i32> @test_vabsq_s32(<4 x i32> %a) #0 {
-; CHECK: abs v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vabs1.i = tail call <4 x i32> @llvm.arm64.neon.abs.v4i32(<4 x i32> %a) #4
-  ret <4 x i32> %vabs1.i
-}
-
-define <2 x i64> @test_vabsq_s64(<2 x i64> %a) #0 {
-; CHECK: abs v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vabs1.i = tail call <2 x i64> @llvm.arm64.neon.abs.v2i64(<2 x i64> %a) #4
-  ret <2 x i64> %vabs1.i
-}
-
-define <2 x float> @test_vabs_f32(<2 x float> %a) #1 {
-; CHECK: fabs v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vabs1.i = tail call <2 x float> @llvm.fabs.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vabs1.i
-}
-
-define <4 x float> @test_vabsq_f32(<4 x float> %a) #1 {
-; CHECK: fabs v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vabs1.i = tail call <4 x float> @llvm.fabs.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vabs1.i
-}
-
-define <2 x double> @test_vabsq_f64(<2 x double> %a) #1 {
-; CHECK: fabs v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vabs1.i = tail call <2 x double> @llvm.fabs.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vabs1.i
-}
-
-define <8 x i8> @test_vuqadd_s8(<8 x i8> %a, <8 x i8> %b) #0 {
-; CHECK: suqadd v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vuqadd.i = tail call <8 x i8> @llvm.arm64.neon.suqadd.v8i8(<8 x i8> %a, <8 x i8> %b) #4
-  ret <8 x i8> %vuqadd.i
-}
-
-define <16 x i8> @test_vuqaddq_s8(<16 x i8> %a, <16 x i8> %b) #0 {
-; CHECK: suqadd v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vuqadd.i = tail call <16 x i8> @llvm.arm64.neon.suqadd.v16i8(<16 x i8> %a, <16 x i8> %b) #4
-  ret <16 x i8> %vuqadd.i
-}
-
-define <4 x i16> @test_vuqadd_s16(<4 x i16> %a, <4 x i16> %b) #0 {
-; CHECK: suqadd v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %vuqadd2.i = tail call <4 x i16> @llvm.arm64.neon.suqadd.v4i16(<4 x i16> %a, <4 x i16> %b) #4
-  ret <4 x i16> %vuqadd2.i
-}
-
-define <8 x i16> @test_vuqaddq_s16(<8 x i16> %a, <8 x i16> %b) #0 {
-; CHECK: suqadd v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %vuqadd2.i = tail call <8 x i16> @llvm.arm64.neon.suqadd.v8i16(<8 x i16> %a, <8 x i16> %b) #4
-  ret <8 x i16> %vuqadd2.i
-}
-
-define <2 x i32> @test_vuqadd_s32(<2 x i32> %a, <2 x i32> %b) #0 {
-; CHECK: suqadd v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vuqadd2.i = tail call <2 x i32> @llvm.arm64.neon.suqadd.v2i32(<2 x i32> %a, <2 x i32> %b) #4
-  ret <2 x i32> %vuqadd2.i
-}
-
-define <4 x i32> @test_vuqaddq_s32(<4 x i32> %a, <4 x i32> %b) #0 {
-; CHECK: suqadd v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vuqadd2.i = tail call <4 x i32> @llvm.arm64.neon.suqadd.v4i32(<4 x i32> %a, <4 x i32> %b) #4
-  ret <4 x i32> %vuqadd2.i
-}
-
-define <2 x i64> @test_vuqaddq_s64(<2 x i64> %a, <2 x i64> %b) #0 {
-; CHECK: suqadd v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vuqadd2.i = tail call <2 x i64> @llvm.arm64.neon.suqadd.v2i64(<2 x i64> %a, <2 x i64> %b) #4
-  ret <2 x i64> %vuqadd2.i
-}
-
-define <8 x i8> @test_vcls_s8(<8 x i8> %a) #0 {
-; CHECK: cls v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vcls.i = tail call <8 x i8> @llvm.arm64.neon.cls.v8i8(<8 x i8> %a) #4
-  ret <8 x i8> %vcls.i
-}
-
-define <16 x i8> @test_vclsq_s8(<16 x i8> %a) #0 {
-; CHECK: cls v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vcls.i = tail call <16 x i8> @llvm.arm64.neon.cls.v16i8(<16 x i8> %a) #4
-  ret <16 x i8> %vcls.i
-}
-
-define <4 x i16> @test_vcls_s16(<4 x i16> %a) #0 {
-; CHECK: cls v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %vcls1.i = tail call <4 x i16> @llvm.arm64.neon.cls.v4i16(<4 x i16> %a) #4
-  ret <4 x i16> %vcls1.i
-}
-
-define <8 x i16> @test_vclsq_s16(<8 x i16> %a) #0 {
-; CHECK: cls v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %vcls1.i = tail call <8 x i16> @llvm.arm64.neon.cls.v8i16(<8 x i16> %a) #4
-  ret <8 x i16> %vcls1.i
-}
-
-define <2 x i32> @test_vcls_s32(<2 x i32> %a) #0 {
-; CHECK: cls v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcls1.i = tail call <2 x i32> @llvm.arm64.neon.cls.v2i32(<2 x i32> %a) #4
-  ret <2 x i32> %vcls1.i
-}
-
-define <4 x i32> @test_vclsq_s32(<4 x i32> %a) #0 {
-; CHECK: cls v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcls1.i = tail call <4 x i32> @llvm.arm64.neon.cls.v4i32(<4 x i32> %a) #4
-  ret <4 x i32> %vcls1.i
-}
-
-define <8 x i8> @test_vclz_s8(<8 x i8> %a) #0 {
-; CHECK: clz v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vclz.i = tail call <8 x i8> @llvm.ctlz.v8i8(<8 x i8> %a, i1 false) #4
-  ret <8 x i8> %vclz.i
-}
-
-define <16 x i8> @test_vclzq_s8(<16 x i8> %a) #0 {
-; CHECK: clz v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vclz.i = tail call <16 x i8> @llvm.ctlz.v16i8(<16 x i8> %a, i1 false) #4
-  ret <16 x i8> %vclz.i
-}
-
-define <4 x i16> @test_vclz_s16(<4 x i16> %a) #0 {
-; CHECK: clz v{{[0-9]+}}.4h, v{{[0-9]+}}.4h
-  %vclz1.i = tail call <4 x i16> @llvm.ctlz.v4i16(<4 x i16> %a, i1 false) #4
-  ret <4 x i16> %vclz1.i
-}
-
-define <8 x i16> @test_vclzq_s16(<8 x i16> %a) #0 {
-; CHECK: clz v{{[0-9]+}}.8h, v{{[0-9]+}}.8h
-  %vclz1.i = tail call <8 x i16> @llvm.ctlz.v8i16(<8 x i16> %a, i1 false) #4
-  ret <8 x i16> %vclz1.i
-}
-
-define <2 x i32> @test_vclz_s32(<2 x i32> %a) #0 {
-; CHECK: clz v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vclz1.i = tail call <2 x i32> @llvm.ctlz.v2i32(<2 x i32> %a, i1 false) #4
-  ret <2 x i32> %vclz1.i
-}
-
-define <4 x i32> @test_vclzq_s32(<4 x i32> %a) #0 {
-; CHECK: clz v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vclz1.i = tail call <4 x i32> @llvm.ctlz.v4i32(<4 x i32> %a, i1 false) #4
-  ret <4 x i32> %vclz1.i
-}
-
-define <8 x i8> @test_vcnt_s8(<8 x i8> %a) #0 {
-; CHECK: cnt v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vctpop.i = tail call <8 x i8> @llvm.ctpop.v8i8(<8 x i8> %a) #4
-  ret <8 x i8> %vctpop.i
-}
-
-define <16 x i8> @test_vcntq_s8(<16 x i8> %a) #0 {
-; CHECK: cnt v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vctpop.i = tail call <16 x i8> @llvm.ctpop.v16i8(<16 x i8> %a) #4
-  ret <16 x i8> %vctpop.i
-}
-
-define <8 x i8> @test_vmvn_s8(<8 x i8> %a) #0 {
-; CHECK: not v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %neg.i = xor <8 x i8> %a, <i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1, i8 -1>
-  ret <8 x i8> %neg.i
-}
-
-define <16 x i8> @test_vmvnq_s8(<16 x i8> %a) #0 {
-; CHECK: not v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %neg.i = xor <16 x i8> %a, <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> %neg.i
-}
-
-define <4 x i16> @test_vmvn_s16(<4 x i16> %a) #0 {
-; CHECK: not v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %neg.i = xor <4 x i16> %a, <i16 -1, i16 -1, i16 -1, i16 -1>
-  ret <4 x i16> %neg.i
-}
-
-define <8 x i16> @test_vmvnq_s16(<8 x i16> %a) #0 {
-; CHECK: not v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %neg.i = xor <8 x i16> %a, <i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1, i16 -1>
-  ret <8 x i16> %neg.i
-}
-
-define <2 x i32> @test_vmvn_s32(<2 x i32> %a) #0 {
-; CHECK: not v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %neg.i = xor <2 x i32> %a, <i32 -1, i32 -1>
-  ret <2 x i32> %neg.i
-}
-
-define <4 x i32> @test_vmvnq_s32(<4 x i32> %a) #0 {
-; CHECK: not v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %neg.i = xor <4 x i32> %a, <i32 -1, i32 -1, i32 -1, i32 -1>
-  ret <4 x i32> %neg.i
-}
-
-define <8 x i8> @test_vrbit_s8(<8 x i8> %a) #0 {
-; CHECK: rbit v{{[0-9]+}}.8b, v{{[0-9]+}}.8b
-  %vrbit.i = tail call <8 x i8> @llvm.arm64.neon.rbit.v8i8(<8 x i8> %a) #4
-  ret <8 x i8> %vrbit.i
-}
-
-define <16 x i8> @test_vrbitq_s8(<16 x i8> %a) #0 {
-; CHECK: rbit v{{[0-9]+}}.16b, v{{[0-9]+}}.16b
-  %vrbit.i = tail call <16 x i8> @llvm.arm64.neon.rbit.v16i8(<16 x i8> %a) #4
-  ret <16 x i8> %vrbit.i
-}
-
-define <8 x i8> @test_vmovn_s16(<8 x i16> %a) #0 {
-; CHECK: xtn v{{[0-9]+}}.8b, v{{[0-9]+}}.8h
-  %vmovn.i = trunc <8 x i16> %a to <8 x i8>
-  ret <8 x i8> %vmovn.i
-}
-
-define <4 x i16> @test_vmovn_s32(<4 x i32> %a) #0 {
-; CHECK: xtn v{{[0-9]+}}.4h, v{{[0-9]+}}.4s
-  %vmovn.i = trunc <4 x i32> %a to <4 x i16>
-  ret <4 x i16> %vmovn.i
-}
-
-define <2 x i32> @test_vmovn_s64(<2 x i64> %a) #0 {
-; CHECK: xtn v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vmovn.i = trunc <2 x i64> %a to <2 x i32>
-  ret <2 x i32> %vmovn.i
-}
-
-define <16 x i8> @test_vmovn_high_s16(<8 x i8> %a, <8 x i16> %b) #0 {
-; CHECK: xtn2 v{{[0-9]+}}.16b, v{{[0-9]+}}.8h
-  %vmovn.i.i = trunc <8 x i16> %b to <8 x i8>
-  %shuffle.i = shufflevector <8 x i8> %a, <8 x i8> %vmovn.i.i, <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> %shuffle.i
-}
-
-define <8 x i16> @test_vmovn_high_s32(<4 x i16> %a, <4 x i32> %b) #0 {
-; CHECK: xtn2 v{{[0-9]+}}.8h, v{{[0-9]+}}.4s
-  %vmovn.i.i = trunc <4 x i32> %b to <4 x i16>
-  %shuffle.i = shufflevector <4 x i16> %a, <4 x i16> %vmovn.i.i, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-  ret <8 x i16> %shuffle.i
-}
-
-define <4 x i32> @test_vmovn_high_s64(<2 x i32> %a, <2 x i64> %b) #0 {
-; CHECK: xtn2 v{{[0-9]+}}.4s, v{{[0-9]+}}.2d
-  %vmovn.i.i = trunc <2 x i64> %b to <2 x i32>
-  %shuffle.i = shufflevector <2 x i32> %a, <2 x i32> %vmovn.i.i, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-  ret <4 x i32> %shuffle.i
-}
-
-define <8 x i8> @test_vqmovun_s16(<8 x i16> %a) #0 {
-; CHECK: sqxtun v{{[0-9]+}}.8b, v{{[0-9]+}}.8h
-  %vqdmull1.i = tail call <8 x i8> @llvm.arm64.neon.sqxtun.v8i8(<8 x i16> %a) #4
-  ret <8 x i8> %vqdmull1.i
-}
-
-define <4 x i16> @test_vqmovun_s32(<4 x i32> %a) #0 {
-; CHECK: sqxtun v{{[0-9]+}}.4h, v{{[0-9]+}}.4s
-  %vqdmull1.i = tail call <4 x i16> @llvm.arm64.neon.sqxtun.v4i16(<4 x i32> %a) #4
-  ret <4 x i16> %vqdmull1.i
-}
-
-define <2 x i32> @test_vqmovun_s64(<2 x i64> %a) #0 {
-; CHECK: sqxtun v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vqdmull1.i = tail call <2 x i32> @llvm.arm64.neon.sqxtun.v2i32(<2 x i64> %a) #4
-  ret <2 x i32> %vqdmull1.i
-}
-
-define <16 x i8> @test_vqmovun_high_s16(<8 x i8> %a, <8 x i16> %b) #0 {
-; CHECK: sqxtun2 v{{[0-9]+}}.16b, v{{[0-9]+}}.8h
-  %vqdmull1.i.i = tail call <8 x i8> @llvm.arm64.neon.sqxtun.v8i8(<8 x i16> %b) #4
-  %shuffle.i = shufflevector <8 x i8> %a, <8 x i8> %vqdmull1.i.i, <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> %shuffle.i
-}
-
-define <8 x i16> @test_vqmovun_high_s32(<4 x i16> %a, <4 x i32> %b) #0 {
-; CHECK: sqxtun2 v{{[0-9]+}}.8h, v{{[0-9]+}}.4s
-  %vqdmull1.i.i = tail call <4 x i16> @llvm.arm64.neon.sqxtun.v4i16(<4 x i32> %b) #4
-  %shuffle.i = shufflevector <4 x i16> %a, <4 x i16> %vqdmull1.i.i, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-  ret <8 x i16> %shuffle.i
-}
-
-define <4 x i32> @test_vqmovun_high_s64(<2 x i32> %a, <2 x i64> %b) #0 {
-; CHECK: sqxtun2 v{{[0-9]+}}.4s, v{{[0-9]+}}.2d
-  %vqdmull1.i.i = tail call <2 x i32> @llvm.arm64.neon.sqxtun.v2i32(<2 x i64> %b) #4
-  %shuffle.i = shufflevector <2 x i32> %a, <2 x i32> %vqdmull1.i.i, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-  ret <4 x i32> %shuffle.i
-}
-
-define <8 x i8> @test_vqmovn_s16(<8 x i16> %a) #0 {
-; CHECK: sqxtn v{{[0-9]+}}.8b, v{{[0-9]+}}.8h
-  %vqmovn1.i = tail call <8 x i8> @llvm.arm64.neon.sqxtn.v8i8(<8 x i16> %a) #4
-  ret <8 x i8> %vqmovn1.i
-}
-
-define <4 x i16> @test_vqmovn_s32(<4 x i32> %a) #0 {
-; CHECK: sqxtn v{{[0-9]+}}.4h, v{{[0-9]+}}.4s
-  %vqmovn1.i = tail call <4 x i16> @llvm.arm64.neon.sqxtn.v4i16(<4 x i32> %a) #4
-  ret <4 x i16> %vqmovn1.i
-}
-
-define <2 x i32> @test_vqmovn_s64(<2 x i64> %a) #0 {
-; CHECK: sqxtn v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vqmovn1.i = tail call <2 x i32> @llvm.arm64.neon.sqxtn.v2i32(<2 x i64> %a) #4
-  ret <2 x i32> %vqmovn1.i
-}
-
-define <16 x i8> @test_vqmovn_high_s16(<8 x i8> %a, <8 x i16> %b) #0 {
-; CHECK: sqxtn2 v{{[0-9]+}}.16b, v{{[0-9]+}}.8h
-  %vqmovn1.i.i = tail call <8 x i8> @llvm.arm64.neon.sqxtn.v8i8(<8 x i16> %b) #4
-  %shuffle.i = shufflevector <8 x i8> %a, <8 x i8> %vqmovn1.i.i, <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> %shuffle.i
-}
-
-define <8 x i16> @test_vqmovn_high_s32(<4 x i16> %a, <4 x i32> %b) #0 {
-; CHECK: test_vqmovn_high_s32
-  %vqmovn1.i.i = tail call <4 x i16> @llvm.arm64.neon.sqxtn.v4i16(<4 x i32> %b) #4
-  %shuffle.i = shufflevector <4 x i16> %a, <4 x i16> %vqmovn1.i.i, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-  ret <8 x i16> %shuffle.i
-}
-
-define <4 x i32> @test_vqmovn_high_s64(<2 x i32> %a, <2 x i64> %b) #0 {
-; CHECK: test_vqmovn_high_s64
-  %vqmovn1.i.i = tail call <2 x i32> @llvm.arm64.neon.sqxtn.v2i32(<2 x i64> %b) #4
-  %shuffle.i = shufflevector <2 x i32> %a, <2 x i32> %vqmovn1.i.i, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-  ret <4 x i32> %shuffle.i
-}
-
-define <8 x i8> @test_vqmovn_u16(<8 x i16> %a) #0 {
-; CHECK: uqxtn v{{[0-9]+}}.8b, v{{[0-9]+}}.8h
-  %vqmovn1.i = tail call <8 x i8> @llvm.arm64.neon.uqxtn.v8i8(<8 x i16> %a) #4
-  ret <8 x i8> %vqmovn1.i
-}
-
-define <4 x i16> @test_vqmovn_u32(<4 x i32> %a) #0 {
-; CHECK: uqxtn v{{[0-9]+}}.4h, v{{[0-9]+}}.4s
-  %vqmovn1.i = tail call <4 x i16> @llvm.arm64.neon.uqxtn.v4i16(<4 x i32> %a) #4
-  ret <4 x i16> %vqmovn1.i
-}
-
-define <2 x i32> @test_vqmovn_u64(<2 x i64> %a) #0 {
-; CHECK: uqxtn v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vqmovn1.i = tail call <2 x i32> @llvm.arm64.neon.uqxtn.v2i32(<2 x i64> %a) #4
-  ret <2 x i32> %vqmovn1.i
-}
-
-define <16 x i8> @test_vqmovn_high_u16(<8 x i8> %a, <8 x i16> %b) #0 {
-; CHECK: uqxtn2 v{{[0-9]+}}.16b, v{{[0-9]+}}.8h
-  %vqmovn1.i.i = tail call <8 x i8> @llvm.arm64.neon.uqxtn.v8i8(<8 x i16> %b) #4
-  %shuffle.i = shufflevector <8 x i8> %a, <8 x i8> %vqmovn1.i.i, <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> %shuffle.i
-}
-
-define <8 x i16> @test_vqmovn_high_u32(<4 x i16> %a, <4 x i32> %b) #0 {
-; CHECK: uqxtn2 v{{[0-9]+}}.8h, v{{[0-9]+}}.4s
-  %vqmovn1.i.i = tail call <4 x i16> @llvm.arm64.neon.uqxtn.v4i16(<4 x i32> %b) #4
-  %shuffle.i = shufflevector <4 x i16> %a, <4 x i16> %vqmovn1.i.i, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-  ret <8 x i16> %shuffle.i
-}
-
-define <4 x i32> @test_vqmovn_high_u64(<2 x i32> %a, <2 x i64> %b) #0 {
-; CHECK: uqxtn2 v{{[0-9]+}}.4s, v{{[0-9]+}}.2d
-  %vqmovn1.i.i = tail call <2 x i32> @llvm.arm64.neon.uqxtn.v2i32(<2 x i64> %b) #4
-  %shuffle.i = shufflevector <2 x i32> %a, <2 x i32> %vqmovn1.i.i, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-  ret <4 x i32> %shuffle.i
-}
-
-define <8 x i16> @test_vshll_n_s8(<8 x i8> %a) #0 {
-; CHECK: shll {{v[0-9]+}}.8h, {{v[0-9]+}}.8b, #8
-  %1 = sext <8 x i8> %a to <8 x i16>
-  %vshll_n = shl <8 x i16> %1, <i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8>
-  ret <8 x i16> %vshll_n
-}
-
-define <4 x i32> @test_vshll_n_s16(<4 x i16> %a) #0 {
-; CHECK: shll {{v[0-9]+}}.4s, {{v[0-9]+}}.4h, #16
-  %1 = sext <4 x i16> %a to <4 x i32>
-  %vshll_n = shl <4 x i32> %1, <i32 16, i32 16, i32 16, i32 16>
-  ret <4 x i32> %vshll_n
-}
-
-define <2 x i64> @test_vshll_n_s32(<2 x i32> %a) #0 {
-; CHECK: shll {{v[0-9]+}}.2d, {{v[0-9]+}}.2s, #32
-  %1 = sext <2 x i32> %a to <2 x i64>
-  %vshll_n = shl <2 x i64> %1, <i64 32, i64 32>
-  ret <2 x i64> %vshll_n
-}
-
-define <8 x i16> @test_vshll_n_u8(<8 x i8> %a) #0 {
-; CHECK: shll {{v[0-9]+}}.8h, {{v[0-9]+}}.8b, #8
-  %1 = zext <8 x i8> %a to <8 x i16>
-  %vshll_n = shl <8 x i16> %1, <i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8>
-  ret <8 x i16> %vshll_n
-}
-
-define <4 x i32> @test_vshll_n_u16(<4 x i16> %a) #0 {
-; CHECK: shll {{v[0-9]+}}.4s, {{v[0-9]+}}.4h, #16
-  %1 = zext <4 x i16> %a to <4 x i32>
-  %vshll_n = shl <4 x i32> %1, <i32 16, i32 16, i32 16, i32 16>
-  ret <4 x i32> %vshll_n
-}
-
-define <2 x i64> @test_vshll_n_u32(<2 x i32> %a) #0 {
-; CHECK: shll {{v[0-9]+}}.2d, {{v[0-9]+}}.2s, #32
-  %1 = zext <2 x i32> %a to <2 x i64>
-  %vshll_n = shl <2 x i64> %1, <i64 32, i64 32>
-  ret <2 x i64> %vshll_n
-}
-
-define <8 x i16> @test_vshll_high_n_s8(<16 x i8> %a) #0 {
-; CHECK: shll2 {{v[0-9]+}}.8h, {{v[0-9]+}}.16b, #8
-  %shuffle.i = shufflevector <16 x i8> %a, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-  %1 = sext <8 x i8> %shuffle.i to <8 x i16>
-  %vshll_n = shl <8 x i16> %1, <i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8>
-  ret <8 x i16> %vshll_n
-}
-
-define <4 x i32> @test_vshll_high_n_s16(<8 x i16> %a) #0 {
-; CHECK: shll2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, #16
-  %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-  %1 = sext <4 x i16> %shuffle.i to <4 x i32>
-  %vshll_n = shl <4 x i32> %1, <i32 16, i32 16, i32 16, i32 16>
-  ret <4 x i32> %vshll_n
-}
-
-define <2 x i64> @test_vshll_high_n_s32(<4 x i32> %a) #0 {
-; CHECK: shll2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, #32
-  %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
-  %1 = sext <2 x i32> %shuffle.i to <2 x i64>
-  %vshll_n = shl <2 x i64> %1, <i64 32, i64 32>
-  ret <2 x i64> %vshll_n
-}
-
-define <8 x i16> @test_vshll_high_n_u8(<16 x i8> %a) #0 {
-; CHECK: shll2 {{v[0-9]+}}.8h, {{v[0-9]+}}.16b, #8
-  %shuffle.i = shufflevector <16 x i8> %a, <16 x i8> undef, <8 x i32> <i32 8, i32 9, i32 10, i32 11, i32 12, i32 13, i32 14, i32 15>
-  %1 = zext <8 x i8> %shuffle.i to <8 x i16>
-  %vshll_n = shl <8 x i16> %1, <i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8, i16 8>
-  ret <8 x i16> %vshll_n
-}
-
-define <4 x i32> @test_vshll_high_n_u16(<8 x i16> %a) #0 {
-; CHECK: shll2 {{v[0-9]+}}.4s, {{v[0-9]+}}.8h, #16
-  %shuffle.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-  %1 = zext <4 x i16> %shuffle.i to <4 x i32>
-  %vshll_n = shl <4 x i32> %1, <i32 16, i32 16, i32 16, i32 16>
-  ret <4 x i32> %vshll_n
-}
-
-define <2 x i64> @test_vshll_high_n_u32(<4 x i32> %a) #0 {
-; CHECK: shll2 {{v[0-9]+}}.2d, {{v[0-9]+}}.4s, #32
-  %shuffle.i = shufflevector <4 x i32> %a, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
-  %1 = zext <2 x i32> %shuffle.i to <2 x i64>
-  %vshll_n = shl <2 x i64> %1, <i64 32, i64 32>
-  ret <2 x i64> %vshll_n
-}
-
-define <4 x i16> @test_vcvt_f16_f32(<4 x float> %a) #0 {
-; CHECK: fcvtn v{{[0-9]+}}.4h, v{{[0-9]+}}.4s
-  %vcvt1.i = tail call <4 x i16> @llvm.arm64.neon.vcvtfp2hf(<4 x float> %a) #4
-  ret <4 x i16> %vcvt1.i
-}
-
-define <8 x i16> @test_vcvt_high_f16_f32(<4 x i16> %a, <4 x float> %b) #0 {
-; CHECK: fcvtn2 v{{[0-9]+}}.8h, v{{[0-9]+}}.4s
-  %vcvt1.i.i = tail call <4 x i16> @llvm.arm64.neon.vcvtfp2hf(<4 x float> %b) #4
-  %shuffle.i = shufflevector <4 x i16> %a, <4 x i16> %vcvt1.i.i, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
-  ret <8 x i16> %shuffle.i
-}
-
-define <4 x float> @test_vcvt_f32_f16(<4 x i16> %a) #0 {
-; CHECK: fcvtl v{{[0-9]+}}.4s, v{{[0-9]+}}.4h
-  %vcvt1.i = tail call <4 x float> @llvm.arm64.neon.vcvthf2fp(<4 x i16> %a) #4
-  ret <4 x float> %vcvt1.i
-}
-
-define <4 x float> @test_vcvt_high_f32_f16(<8 x i16> %a) #0 {
-; CHECK: fcvtl2 v{{[0-9]+}}.4s, v{{[0-9]+}}.8h
-  %shuffle.i.i = shufflevector <8 x i16> %a, <8 x i16> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
-  %vcvt1.i.i = tail call <4 x float> @llvm.arm64.neon.vcvthf2fp(<4 x i16> %shuffle.i.i) #4
-  ret <4 x float> %vcvt1.i.i
-}
-
-define <2 x float> @test_vcvt_f32_f64(<2 x double> %a) #0 {
-; CHECK: fcvtn v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vcvt.i = fptrunc <2 x double> %a to <2 x float>
-  ret <2 x float> %vcvt.i
-}
-
-define <4 x float> @test_vcvt_high_f32_f64(<2 x float> %a, <2 x double> %b) #0 {
-; CHECK: fcvtn2 v{{[0-9]+}}.4s, v{{[0-9]+}}.2d
-  %vcvt.i.i = fptrunc <2 x double> %b to <2 x float>
-  %shuffle.i = shufflevector <2 x float> %a, <2 x float> %vcvt.i.i, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-  ret <4 x float> %shuffle.i
-}
-
-define <2 x float> @test_vcvtx_f32_f64(<2 x double> %a) #0 {
-; CHECK: fcvtxn v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vcvtx_f32_f641.i = call <2 x float> @llvm.arm64.neon.fcvtxn.v2f32.v2f64(<2 x double> %a) #4
-  ret <2 x float> %vcvtx_f32_f641.i
-}
-
-define <4 x float> @test_vcvtx_high_f32_f64(<2 x float> %a, <2 x double> %b) #0 {
-; CHECK: fcvtxn2 v{{[0-9]+}}.4s, v{{[0-9]+}}.2d
-  %vcvtx_f32_f641.i.i = tail call <2 x float> @llvm.arm64.neon.fcvtxn.v2f32.v2f64(<2 x double> %b) #4
-  %shuffle.i = shufflevector <2 x float> %a, <2 x float> %vcvtx_f32_f641.i.i, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
-  ret <4 x float> %shuffle.i
-}
-
-define <2 x double> @test_vcvt_f64_f32(<2 x float> %a) #0 {
-; CHECK: fcvtl v{{[0-9]+}}.2d, v{{[0-9]+}}.2s
-  %vcvt.i = fpext <2 x float> %a to <2 x double>
-  ret <2 x double> %vcvt.i
-}
-
-define <2 x double> @test_vcvt_high_f64_f32(<4 x float> %a) #0 {
-; CHECK: fcvtl2 v{{[0-9]+}}.2d, v{{[0-9]+}}.4s
-  %shuffle.i.i = shufflevector <4 x float> %a, <4 x float> undef, <2 x i32> <i32 2, i32 3>
-  %vcvt.i.i = fpext <2 x float> %shuffle.i.i to <2 x double>
-  ret <2 x double> %vcvt.i.i
-}
-
-define <2 x float> @test_vrndn_f32(<2 x float> %a) #0 {
-; CHECK: frintn v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrndn1.i = tail call <2 x float> @llvm.arm64.neon.frintn.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrndn1.i
-}
-
-define <4 x float> @test_vrndnq_f32(<4 x float> %a) #0 {
-; CHECK: frintn v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrndn1.i = tail call <4 x float> @llvm.arm64.neon.frintn.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrndn1.i
-}
-
-define <2 x double> @test_vrndnq_f64(<2 x double> %a) #0 {
-; CHECK: frintn v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrndn1.i = tail call <2 x double> @llvm.arm64.neon.frintn.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrndn1.i
-}
-
-define <2 x float> @test_vrnda_f32(<2 x float> %a) #0 {
-; CHECK: frinta v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrnda1.i = tail call <2 x float> @llvm.round.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrnda1.i
-}
-
-define <4 x float> @test_vrndaq_f32(<4 x float> %a) #0 {
-; CHECK: frinta v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-   %vrnda1.i = tail call <4 x float> @llvm.round.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrnda1.i
-}
-
-define <2 x double> @test_vrndaq_f64(<2 x double> %a) #0 {
-; CHECK: frinta v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrnda1.i = tail call <2 x double> @llvm.round.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrnda1.i
-}
-
-define <2 x float> @test_vrndp_f32(<2 x float> %a) #0 {
-; CHECK: frintp v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrndp1.i = tail call <2 x float> @llvm.ceil.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrndp1.i
-}
-
-define <4 x float> @test_vrndpq_f32(<4 x float> %a) #0 {
-; CHECK: frintp v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
- %vrndp1.i = tail call <4 x float> @llvm.ceil.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrndp1.i
-}
-
-define <2 x double> @test_vrndpq_f64(<2 x double> %a) #0 {
-; CHECK: frintp v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrndp1.i = tail call <2 x double> @llvm.ceil.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrndp1.i
-}
-
-define <2 x float> @test_vrndm_f32(<2 x float> %a) #0 {
-; CHECK: frintm v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrndm1.i = tail call <2 x float> @llvm.floor.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrndm1.i
-}
-
-define <4 x float> @test_vrndmq_f32(<4 x float> %a) #0 {
-; CHECK: frintm v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrndm1.i = tail call <4 x float> @llvm.floor.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrndm1.i
-}
-
-define <2 x double> @test_vrndmq_f64(<2 x double> %a) #0 {
-; CHECK: frintm v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-   %vrndm1.i = tail call <2 x double> @llvm.floor.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrndm1.i
-}
-
-define <2 x float> @test_vrndx_f32(<2 x float> %a) #0 {
-; CHECK: frintx v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrndx1.i = tail call <2 x float> @llvm.rint.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrndx1.i
-}
-
-define <4 x float> @test_vrndxq_f32(<4 x float> %a) #0 {
-; CHECK: frintx v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrndx1.i = tail call <4 x float> @llvm.rint.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrndx1.i
-}
-
-define <2 x double> @test_vrndxq_f64(<2 x double> %a) #0 {
-; CHECK: frintx v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrndx1.i = tail call <2 x double> @llvm.rint.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrndx1.i
-}
-
-define <2 x float> @test_vrnd_f32(<2 x float> %a) #0 {
-; CHECK: frintz v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-   %vrnd1.i = tail call <2 x float> @llvm.trunc.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrnd1.i
-}
-
-define <4 x float> @test_vrndq_f32(<4 x float> %a) #0 {
-; CHECK: frintz v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrnd1.i = tail call <4 x float> @llvm.trunc.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrnd1.i
-}
-
-define <2 x double> @test_vrndq_f64(<2 x double> %a) #0 {
-; CHECK: frintz v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrnd1.i = tail call <2 x double> @llvm.trunc.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrnd1.i
-}
-
-define <2 x float> @test_vrndi_f32(<2 x float> %a) #0 {
-; CHECK: frinti v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrndi1.i = tail call <2 x float> @llvm.nearbyint.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrndi1.i
-}
-
-define <4 x float> @test_vrndiq_f32(<4 x float> %a) #0 {
-; CHECK: frinti v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrndi1.i = tail call <4 x float> @llvm.nearbyint.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrndi1.i
-}
-
-define <2 x double> @test_vrndiq_f64(<2 x double> %a) #0 {
-; CHECK: frinti v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrndi1.i = tail call <2 x double> @llvm.nearbyint.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrndi1.i
-}
-
-define <2 x i32> @test_vcvt_s32_f32(<2 x float> %a) #0 {
-; CHECK: fcvtzs v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvt.i = fptosi <2 x float> %a to <2 x i32>
-  ret <2 x i32> %vcvt.i
-}
-
-define <4 x i32> @test_vcvtq_s32_f32(<4 x float> %a) #0 {
-; CHECK: fcvtzs v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvt.i = fptosi <4 x float> %a to <4 x i32>
-  ret <4 x i32> %vcvt.i
-}
-
-define <2 x i64> @test_vcvtq_s64_f64(<2 x double> %a) #0 {
-; CHECK: fcvtzs v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = fptosi <2 x double> %a to <2 x i64>
-  ret <2 x i64> %vcvt.i
-}
-
-define <2 x i32> @test_vcvt_u32_f32(<2 x float> %a) #0 {
-; CHECK: fcvtzu v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvt.i = fptoui <2 x float> %a to <2 x i32>
-  ret <2 x i32> %vcvt.i
-}
-
-define <4 x i32> @test_vcvtq_u32_f32(<4 x float> %a) #0 {
-; CHECK: fcvtzu v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvt.i = fptoui <4 x float> %a to <4 x i32>
-  ret <4 x i32> %vcvt.i
-}
-
-define <2 x i64> @test_vcvtq_u64_f64(<2 x double> %a) #0 {
-; CHECK: fcvtzu v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = fptoui <2 x double> %a to <2 x i64>
-  ret <2 x i64> %vcvt.i
-}
-
-define <2 x i64> @test_vcvt_s64_f32(<2 x float> %a) #0 {
-; CHECK: fcvtl v{{[0-9]+}}.2d, v{{[0-9]+}}.2s
-; CHECK: fcvtzs v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = fptosi <2 x float> %a to <2 x i64>
-  ret <2 x i64> %vcvt.i
-}
-
-define <2 x i64> @test_vcvt_u64_f32(<2 x float> %a) #0 {
-; CHECK: fcvtl v{{[0-9]+}}.2d, v{{[0-9]+}}.2s
-; CHECK: fcvtzu v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = fptoui <2 x float> %a to <2 x i64>
-  ret <2 x i64> %vcvt.i
-}
-
-define <4 x i16> @test_vcvt_s16_f32(<4 x float> %a) #0 {
-; CHECK: fcvtzs v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-; CHECK: xtn v{{[0-9]+}}.4h, v{{[0-9]+}}.4s
-  %vcvt.i = fptosi <4 x float> %a to <4 x i16>
-  ret <4 x i16> %vcvt.i
-}
-
-define <4 x i16> @test_vcvt_u16_f32(<4 x float> %a) #0 {
-; CHECK: fcvtzu v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-; CHECK: xtn v{{[0-9]+}}.4h, v{{[0-9]+}}.4s
-  %vcvt.i = fptoui <4 x float> %a to <4 x i16>
-  ret <4 x i16> %vcvt.i
-}
-
-define <2 x i32> @test_vcvt_s32_f64(<2 x double> %a) #0 {
-; CHECK: fcvtzs v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-; CHECK: xtn v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vcvt.i = fptosi <2 x double> %a to <2 x i32>
-  ret <2 x i32> %vcvt.i
-}
-
-define <2 x i32> @test_vcvt_u32_f64(<2 x double> %a) #0 {
-; CHECK: fcvtzu v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-; CHECK: xtn v{{[0-9]+}}.2s, v{{[0-9]+}}.2d
-  %vcvt.i = fptoui <2 x double> %a to <2 x i32>
-  ret <2 x i32> %vcvt.i
-}
-
-define <2 x i32> @test_vcvtn_s32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvtn_s32_f32
-; CHECK: fcvtns v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtns_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtns.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtns_f321.i
-}
-
-define <4 x i32> @test_vcvtnq_s32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtnq_s32_f32
-; CHECK: fcvtns v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtns_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtns.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtns_f321.i
-}
-
-define <2 x i64> @test_vcvtnq_s64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtnq_s64_f64
-; CHECK: fcvtns v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtns_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtns.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtns_f641.i
-}
-
-define <2 x i32> @test_vcvtn_u32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvtn_u32_f32
-; CHECK: fcvtnu v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtnu_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtnu.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtnu_f321.i
-}
-
-define <4 x i32> @test_vcvtnq_u32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtnq_u32_f32
-; CHECK: fcvtnu v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtnu_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtnu.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtnu_f321.i
-}
-
-define <2 x i64> @test_vcvtnq_u64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtnq_u64_f64
-; CHECK: fcvtnu v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtnu_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtnu.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtnu_f641.i
-}
-
-define <2 x i32> @test_vcvtp_s32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvtp_s32_f32
-; CHECK: fcvtps v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtps_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtps.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtps_f321.i
-}
-
-define <4 x i32> @test_vcvtpq_s32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtpq_s32_f32
-; CHECK: fcvtps v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtps_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtps.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtps_f321.i
-}
-
-define <2 x i64> @test_vcvtpq_s64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtpq_s64_f64
-; CHECK: fcvtps v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtps_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtps.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtps_f641.i
-}
-
-define <2 x i32> @test_vcvtp_u32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvtp_u32_f32
-; CHECK: fcvtpu v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtpu_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtpu.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtpu_f321.i
-}
-
-define <4 x i32> @test_vcvtpq_u32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtpq_u32_f32
-; CHECK: fcvtpu v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtpu_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtpu.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtpu_f321.i
-}
-
-define <2 x i64> @test_vcvtpq_u64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtpq_u64_f64
-; CHECK: fcvtpu v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtpu_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtpu.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtpu_f641.i
-}
-
-define <2 x i32> @test_vcvtm_s32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvtm_s32_f32
-; CHECK: fcvtms v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtms_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtms.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtms_f321.i
-}
-
-define <4 x i32> @test_vcvtmq_s32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtmq_s32_f32
-; CHECK: fcvtms v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtms_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtms.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtms_f321.i
-}
-
-define <2 x i64> @test_vcvtmq_s64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtmq_s64_f64
-; CHECK: fcvtms v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtms_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtms.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtms_f641.i
-}
-
-define <2 x i32> @test_vcvtm_u32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvtm_u32_f32
-; CHECK: fcvtmu v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtmu_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtmu.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtmu_f321.i
-}
-
-define <4 x i32> @test_vcvtmq_u32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtmq_u32_f32
-; CHECK: fcvtmu v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtmu_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtmu.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtmu_f321.i
-}
-
-define <2 x i64> @test_vcvtmq_u64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtmq_u64_f64
-; CHECK: fcvtmu v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtmu_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtmu.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtmu_f641.i
-}
-
-define <2 x i32> @test_vcvta_s32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvta_s32_f32
-; CHECK: fcvtas v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtas_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtas.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtas_f321.i
-}
-
-define <4 x i32> @test_vcvtaq_s32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtaq_s32_f32
-; CHECK: fcvtas v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtas_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtas.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtas_f321.i
-}
-
-define <2 x i64> @test_vcvtaq_s64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtaq_s64_f64
-; CHECK: fcvtas v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtas_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtas.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtas_f641.i
-}
-
-define <2 x i32> @test_vcvta_u32_f32(<2 x float> %a) {
-; CHECK-LABEL: test_vcvta_u32_f32
-; CHECK: fcvtau v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvtau_f321.i = call <2 x i32> @llvm.arm64.neon.fcvtau.v2i32.v2f32(<2 x float> %a)
-  ret <2 x i32> %vcvtau_f321.i
-}
-
-define <4 x i32> @test_vcvtaq_u32_f32(<4 x float> %a) {
-; CHECK-LABEL: test_vcvtaq_u32_f32
-; CHECK: fcvtau v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvtau_f321.i = call <4 x i32> @llvm.arm64.neon.fcvtau.v4i32.v4f32(<4 x float> %a)
-  ret <4 x i32> %vcvtau_f321.i
-}
-
-define <2 x i64> @test_vcvtaq_u64_f64(<2 x double> %a) {
-; CHECK-LABEL: test_vcvtaq_u64_f64
-; CHECK: fcvtau v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvtau_f641.i = call <2 x i64> @llvm.arm64.neon.fcvtau.v2i64.v2f64(<2 x double> %a)
-  ret <2 x i64> %vcvtau_f641.i
-}
-
-define <2 x float> @test_vrsqrte_f32(<2 x float> %a) #0 {
-; CHECK: frsqrte v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrsqrte1.i = tail call <2 x float> @llvm.arm64.neon.frsqrte.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrsqrte1.i
-}
-
-define <4 x float> @test_vrsqrteq_f32(<4 x float> %a) #0 {
-; CHECK: frsqrte v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrsqrte1.i = tail call <4 x float> @llvm.arm64.neon.frsqrte.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrsqrte1.i
-}
-
-define <2 x double> @test_vrsqrteq_f64(<2 x double> %a) #0 {
-; CHECK: frsqrte v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrsqrte1.i = tail call <2 x double> @llvm.arm64.neon.frsqrte.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrsqrte1.i
-}
-
-declare <2 x float> @llvm.arm64.neon.frecpe.v2f32(<2 x float>)
-declare <4 x float> @llvm.arm64.neon.frecpe.v4f32(<4 x float>)
-declare <2 x double> @llvm.arm64.neon.frecpe.v2f64(<2 x double>)
-
-define <2 x float> @test_vrecpe_f32(<2 x float> %a) #0 {
-; CHECK: frecpe v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrecpe1.i = tail call <2 x float> @llvm.arm64.neon.frecpe.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vrecpe1.i
-}
-
-define <4 x float> @test_vrecpeq_f32(<4 x float> %a) #0 {
-; CHECK: frecpe v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrecpe1.i = tail call <4 x float> @llvm.arm64.neon.frecpe.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vrecpe1.i
-}
-
-define <2 x double> @test_vrecpeq_f64(<2 x double> %a) #0 {
-; CHECK: frecpe v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vrecpe1.i = tail call <2 x double> @llvm.arm64.neon.frecpe.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vrecpe1.i
-}
-
-define <2 x i32> @test_vrecpe_u32(<2 x i32> %a) #0 {
-; CHECK: urecpe v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vrecpe1.i = tail call <2 x i32> @llvm.arm64.neon.urecpe.v2i32(<2 x i32> %a) #4
-  ret <2 x i32> %vrecpe1.i
-}
-
-define <4 x i32> @test_vrecpeq_u32(<4 x i32> %a) #0 {
-; CHECK: urecpe v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vrecpe1.i = tail call <4 x i32> @llvm.arm64.neon.urecpe.v4i32(<4 x i32> %a) #4
-  ret <4 x i32> %vrecpe1.i
-}
-
-define <2 x float> @test_vsqrt_f32(<2 x float> %a) #0 {
-; CHECK: fsqrt v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vsqrt1.i = tail call <2 x float> @llvm.sqrt.v2f32(<2 x float> %a) #4
-  ret <2 x float> %vsqrt1.i
-}
-
-define <4 x float> @test_vsqrtq_f32(<4 x float> %a) #0 {
-; CHECK: fsqrt v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vsqrt1.i = tail call <4 x float> @llvm.sqrt.v4f32(<4 x float> %a) #4
-  ret <4 x float> %vsqrt1.i
-}
-
-define <2 x double> @test_vsqrtq_f64(<2 x double> %a) #0 {
-; CHECK: fsqrt v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vsqrt1.i = tail call <2 x double> @llvm.sqrt.v2f64(<2 x double> %a) #4
-  ret <2 x double> %vsqrt1.i
-}
-
-define <2 x float> @test_vcvt_f32_s32(<2 x i32> %a) #0 {
-; CHECK: scvtf v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvt.i = sitofp <2 x i32> %a to <2 x float>
-  ret <2 x float> %vcvt.i
-}
-
-define <2 x float> @test_vcvt_f32_u32(<2 x i32> %a) #0 {
-; CHECK: ucvtf v{{[0-9]+}}.2s, v{{[0-9]+}}.2s
-  %vcvt.i = uitofp <2 x i32> %a to <2 x float>
-  ret <2 x float> %vcvt.i
-}
-
-define <4 x float> @test_vcvtq_f32_s32(<4 x i32> %a) #0 {
-; CHECK: scvtf v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvt.i = sitofp <4 x i32> %a to <4 x float>
-  ret <4 x float> %vcvt.i
-}
-
-define <4 x float> @test_vcvtq_f32_u32(<4 x i32> %a) #0 {
-; CHECK: ucvtf v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvt.i = uitofp <4 x i32> %a to <4 x float>
-  ret <4 x float> %vcvt.i
-}
-
-define <2 x double> @test_vcvtq_f64_s64(<2 x i64> %a) #0 {
-; CHECK: scvtf v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = sitofp <2 x i64> %a to <2 x double>
-  ret <2 x double> %vcvt.i
-}
-
-define <2 x double> @test_vcvtq_f64_u64(<2 x i64> %a) #0 {
-; CHECK: ucvtf v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = uitofp <2 x i64> %a to <2 x double>
-  ret <2 x double> %vcvt.i
-}
-
-define <4 x float> @test_vcvt_f32_s16(<4 x i16> %a) #0 {
-; CHECK: sshll v{{[0-9]+}}.4s, v{{[0-9]+}}.4h, #0
-; CHECK: scvtf v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvt.i = sitofp <4 x i16> %a to <4 x float>
-  ret <4 x float> %vcvt.i
-}
-
-define <4 x float> @test_vcvt_f32_u16(<4 x i16> %a) #0 {
-; CHECK: ushll v{{[0-9]+}}.4s, v{{[0-9]+}}.4h, #0
-; CHECK: ucvtf v{{[0-9]+}}.4s, v{{[0-9]+}}.4s
-  %vcvt.i = uitofp <4 x i16> %a to <4 x float>
-  ret <4 x float> %vcvt.i
-}
-
-define <2 x double> @test_vcvt_f64_s32(<2 x i32> %a) #0 {
-; CHECK: sshll v{{[0-9]+}}.2d, v{{[0-9]+}}.2s, #0
-; CHECK: scvtf v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = sitofp <2 x i32> %a to <2 x double>
-  ret <2 x double> %vcvt.i
-}
-
-define <2 x double> @test_vcvt_f64_u32(<2 x i32> %a) #0 {
-; CHECK: ushll v{{[0-9]+}}.2d, v{{[0-9]+}}.2s, #0
-; CHECK: ucvtf v{{[0-9]+}}.2d, v{{[0-9]+}}.2d
-  %vcvt.i = uitofp <2 x i32> %a to <2 x double>
-  ret <2 x double> %vcvt.i
-}
-
-declare <2 x double> @llvm.sqrt.v2f64(<2 x double>) #2
-
-declare <4 x float> @llvm.sqrt.v4f32(<4 x float>) #2
-
-declare <2 x float> @llvm.sqrt.v2f32(<2 x float>) #2
-
-declare <4 x i32> @llvm.arm64.neon.urecpe.v4i32(<4 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.urecpe.v2i32(<2 x i32>) #2
-
-declare <2 x double> @llvm.arm64.neon.urecpe.v2f64(<2 x double>) #2
-
-declare <4 x float> @llvm.arm64.neon.urecpe.v4f32(<4 x float>) #2
-
-declare <2 x float> @llvm.arm64.neon.urecpe.v2f32(<2 x float>) #2
-
-declare <2 x double> @llvm.arm64.neon.frsqrte.v2f64(<2 x double>) #2
-
-declare <4 x float> @llvm.arm64.neon.frsqrte.v4f32(<4 x float>) #2
-
-declare <2 x float> @llvm.arm64.neon.frsqrte.v2f32(<2 x float>) #2
-
-declare <2 x i64> @llvm.arm64.neon.fcvtau.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtau.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtau.v2i32.v2f32(<2 x float>)
-
-declare <2 x i64> @llvm.arm64.neon.fcvtas.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtas.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtas.v2i32.v2f32(<2 x float>)
-
-declare <2 x i64> @llvm.arm64.neon.fcvtmu.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtmu.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtmu.v2i32.v2f32(<2 x float>)
-
-declare <2 x i64> @llvm.arm64.neon.fcvtms.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtms.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtms.v2i32.v2f32(<2 x float>)
-
-declare <2 x i64> @llvm.arm64.neon.fcvtpu.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtpu.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtpu.v2i32.v2f32(<2 x float>)
-
-declare <2 x i64> @llvm.arm64.neon.fcvtps.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtps.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtps.v2i32.v2f32(<2 x float>)
-
-declare <2 x i64> @llvm.arm64.neon.fcvtnu.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtnu.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtnu.v2i32.v2f32(<2 x float>)
-
-declare <2 x i64> @llvm.arm64.neon.fcvtns.v2i64.v2f64(<2 x double>)
-
-declare <4 x i32> @llvm.arm64.neon.fcvtns.v4i32.v4f32(<4 x float>)
-
-declare <2 x i32> @llvm.arm64.neon.fcvtns.v2i32.v2f32(<2 x float>)
-
-declare <2 x double> @llvm.nearbyint.v2f64(<2 x double>) #3
-
-declare <4 x float> @llvm.nearbyint.v4f32(<4 x float>) #3
-
-declare <2 x float> @llvm.nearbyint.v2f32(<2 x float>) #3
-
-declare <2 x double> @llvm.trunc.v2f64(<2 x double>) #3
-
-declare <4 x float> @llvm.trunc.v4f32(<4 x float>) #3
-
-declare <2 x float> @llvm.trunc.v2f32(<2 x float>) #3
-
-declare <2 x double> @llvm.rint.v2f64(<2 x double>) #3
-
-declare <4 x float> @llvm.rint.v4f32(<4 x float>) #3
-
-declare <2 x float> @llvm.rint.v2f32(<2 x float>) #3
-
-declare <2 x double> @llvm.floor.v2f64(<2 x double>) #3
-
-declare <4 x float> @llvm.floor.v4f32(<4 x float>) #3
-
-declare <2 x float> @llvm.floor.v2f32(<2 x float>) #3
-
-declare <2 x double> @llvm.ceil.v2f64(<2 x double>) #3
-
-declare <4 x float> @llvm.ceil.v4f32(<4 x float>) #3
-
-declare <2 x float> @llvm.ceil.v2f32(<2 x float>) #3
-
-declare <2 x double> @llvm.round.v2f64(<2 x double>) #3
-
-declare <4 x float> @llvm.round.v4f32(<4 x float>) #3
-
-declare <2 x float> @llvm.round.v2f32(<2 x float>) #3
-
-declare <2 x double> @llvm.arm64.neon.frintn.v2f64(<2 x double>) #2
-
-declare <4 x float> @llvm.arm64.neon.frintn.v4f32(<4 x float>) #2
-
-declare <2 x float> @llvm.arm64.neon.frintn.v2f32(<2 x float>) #2
-
-declare <2 x float> @llvm.arm64.neon.fcvtxn.v2f32.v2f64(<2 x double>) #2
-
-declare <2 x float> @llvm.arm64.neon.fcvtn.v2f32.v2f64(<2 x double>) #2
-
-declare <2 x i32> @llvm.arm64.neon.uqxtn.v2i32(<2 x i64>) #2
-
-declare <4 x i16> @llvm.arm64.neon.uqxtn.v4i16(<4 x i32>) #2
-
-declare <8 x i8> @llvm.arm64.neon.uqxtn.v8i8(<8 x i16>) #2
-
-declare <2 x i32> @llvm.arm64.neon.sqxtn.v2i32(<2 x i64>) #2
-
-declare <4 x i16> @llvm.arm64.neon.sqxtn.v4i16(<4 x i32>) #2
-
-declare <8 x i8> @llvm.arm64.neon.sqxtn.v8i8(<8 x i16>) #2
-
-declare <2 x i32> @llvm.arm64.neon.sqxtun.v2i32(<2 x i64>) #2
-
-declare <4 x i16> @llvm.arm64.neon.sqxtun.v4i16(<4 x i32>) #2
-
-declare <8 x i8> @llvm.arm64.neon.sqxtun.v8i8(<8 x i16>) #2
-
-declare <16 x i8> @llvm.arm64.neon.rbit.v16i8(<16 x i8>) #2
-
-declare <8 x i8> @llvm.arm64.neon.rbit.v8i8(<8 x i8>) #2
-
-declare <16 x i8> @llvm.ctpop.v16i8(<16 x i8>) #2
-
-declare <8 x i8> @llvm.ctpop.v8i8(<8 x i8>) #2
-
-declare <4 x i32> @llvm.ctlz.v4i32(<4 x i32>, i1) #2
-
-declare <2 x i32> @llvm.ctlz.v2i32(<2 x i32>, i1) #2
-
-declare <8 x i16> @llvm.ctlz.v8i16(<8 x i16>, i1) #2
-
-declare <4 x i16> @llvm.ctlz.v4i16(<4 x i16>, i1) #2
-
-declare <16 x i8> @llvm.ctlz.v16i8(<16 x i8>, i1) #2
-
-declare <8 x i8> @llvm.ctlz.v8i8(<8 x i8>, i1) #2
-
-declare <4 x i32> @llvm.arm64.neon.cls.v4i32(<4 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.cls.v2i32(<2 x i32>) #2
-
-declare <8 x i16> @llvm.arm64.neon.cls.v8i16(<8 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.cls.v4i16(<4 x i16>) #2
-
-declare <16 x i8> @llvm.arm64.neon.cls.v16i8(<16 x i8>) #2
-
-declare <8 x i8> @llvm.arm64.neon.cls.v8i8(<8 x i8>) #2
-
-declare <2 x i64> @llvm.arm64.neon.suqadd.v2i64(<2 x i64>, <2 x i64>) #2
-
-declare <4 x i32> @llvm.arm64.neon.suqadd.v4i32(<4 x i32>, <4 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.suqadd.v2i32(<2 x i32>, <2 x i32>) #2
-
-declare <8 x i16> @llvm.arm64.neon.suqadd.v8i16(<8 x i16>, <8 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.suqadd.v4i16(<4 x i16>, <4 x i16>) #2
-
-declare <16 x i8> @llvm.arm64.neon.suqadd.v16i8(<16 x i8>, <16 x i8>) #2
-
-declare <8 x i8> @llvm.arm64.neon.suqadd.v8i8(<8 x i8>, <8 x i8>) #2
-
-declare <2 x double> @llvm.fabs.v2f64(<2 x double>) #3
-
-declare <4 x float> @llvm.fabs.v4f32(<4 x float>) #3
-
-declare <2 x float> @llvm.fabs.v2f32(<2 x float>) #3
-
-declare <2 x i64> @llvm.arm64.neon.abs.v2i64(<2 x i64>) #2
-
-declare <4 x i32> @llvm.arm64.neon.abs.v4i32(<4 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.abs.v2i32(<2 x i32>) #2
-
-declare <8 x i16> @llvm.arm64.neon.abs.v8i16(<8 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.abs.v4i16(<4 x i16>) #2
-
-declare <16 x i8> @llvm.arm64.neon.abs.v16i8(<16 x i8>) #2
-
-declare <8 x i8> @llvm.arm64.neon.abs.v8i8(<8 x i8>) #2
-
-declare <2 x i64> @llvm.arm64.neon.sqneg.v2i64(<2 x i64>) #2
-
-declare <4 x i32> @llvm.arm64.neon.sqneg.v4i32(<4 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.sqneg.v2i32(<2 x i32>) #2
-
-declare <8 x i16> @llvm.arm64.neon.sqneg.v8i16(<8 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.sqneg.v4i16(<4 x i16>) #2
-
-declare <16 x i8> @llvm.arm64.neon.sqneg.v16i8(<16 x i8>) #2
-
-declare <8 x i8> @llvm.arm64.neon.sqneg.v8i8(<8 x i8>) #2
-
-declare <2 x i64> @llvm.arm64.neon.sqabs.v2i64(<2 x i64>) #2
-
-declare <4 x i32> @llvm.arm64.neon.sqabs.v4i32(<4 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.sqabs.v2i32(<2 x i32>) #2
-
-declare <8 x i16> @llvm.arm64.neon.sqabs.v8i16(<8 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.sqabs.v4i16(<4 x i16>) #2
-
-declare <16 x i8> @llvm.arm64.neon.sqabs.v16i8(<16 x i8>) #2
-
-declare <8 x i8> @llvm.arm64.neon.sqabs.v8i8(<8 x i8>) #2
-
-declare <2 x i64> @llvm.arm64.neon.uaddlp.v2i64.v4i32(<4 x i32>) #2
-
-declare <4 x i32> @llvm.arm64.neon.uaddlp.v4i32.v8i16(<8 x i16>) #2
-
-declare <8 x i16> @llvm.arm64.neon.uaddlp.v8i16.v16i8(<16 x i8>) #2
-
-declare <2 x i64> @llvm.arm64.neon.saddlp.v2i64.v4i32(<4 x i32>) #2
-
-declare <4 x i32> @llvm.arm64.neon.saddlp.v4i32.v8i16(<8 x i16>) #2
-
-declare <8 x i16> @llvm.arm64.neon.saddlp.v8i16.v16i8(<16 x i8>) #2
-
-declare <1 x i64> @llvm.arm64.neon.uaddlp.v1i64.v2i32(<2 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.uaddlp.v2i32.v4i16(<4 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.uaddlp.v4i16.v8i8(<8 x i8>) #2
-
-declare <1 x i64> @llvm.arm64.neon.saddlp.v1i64.v2i32(<2 x i32>) #2
-
-declare <2 x i32> @llvm.arm64.neon.saddlp.v2i32.v4i16(<4 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.saddlp.v4i16.v8i8(<8 x i8>) #2
-
-declare <4 x float> @llvm.arm64.neon.vcvthf2fp(<4 x i16>) #2
-
-declare <4 x i16> @llvm.arm64.neon.vcvtfp2hf(<4 x float>) #2
-
-
-define <1 x i64> @test_vcvt_s64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvt_s64_f64
-; CHECK: fcvtzs {{[xd][0-9]+}}, d{{[0-9]+}}
-  %1 = fptosi <1 x double> %a to <1 x i64>
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvt_u64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvt_u64_f64
-; CHECK: fcvtzu {{[xd][0-9]+}}, d{{[0-9]+}}
-  %1 = fptoui <1 x double> %a to <1 x i64>
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvtn_s64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvtn_s64_f64
-; CHECK: fcvtns d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtns.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvtn_u64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvtn_u64_f64
-; CHECK: fcvtnu d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtnu.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvtp_s64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvtp_s64_f64
-; CHECK: fcvtps d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtps.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvtp_u64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvtp_u64_f64
-; CHECK: fcvtpu d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtpu.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvtm_s64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvtm_s64_f64
-; CHECK: fcvtms d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtms.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvtm_u64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvtm_u64_f64
-; CHECK: fcvtmu d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtmu.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvta_s64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvta_s64_f64
-; CHECK: fcvtas d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtas.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x i64> @test_vcvta_u64_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vcvta_u64_f64
-; CHECK: fcvtau d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = call <1 x i64> @llvm.arm64.neon.fcvtau.v1i64.v1f64(<1 x double> %a)
-  ret <1 x i64> %1
-}
-
-define <1 x double> @test_vcvt_f64_s64(<1 x i64> %a) {
-; CHECK-LABEL: test_vcvt_f64_s64
-; CHECK: scvtf d{{[0-9]+}}, {{[xd][0-9]+}}
-  %1 = sitofp <1 x i64> %a to <1 x double>
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vcvt_f64_u64(<1 x i64> %a) {
-; CHECK-LABEL: test_vcvt_f64_u64
-; CHECK: ucvtf d{{[0-9]+}}, {{[xd][0-9]+}}
-  %1 = uitofp <1 x i64> %a to <1 x double>
-  ret <1 x double> %1
-}
-
-declare <1 x i64> @llvm.arm64.neon.fcvtau.v1i64.v1f64(<1 x double>)
-declare <1 x i64> @llvm.arm64.neon.fcvtas.v1i64.v1f64(<1 x double>)
-declare <1 x i64> @llvm.arm64.neon.fcvtmu.v1i64.v1f64(<1 x double>)
-declare <1 x i64> @llvm.arm64.neon.fcvtms.v1i64.v1f64(<1 x double>)
-declare <1 x i64> @llvm.arm64.neon.fcvtpu.v1i64.v1f64(<1 x double>)
-declare <1 x i64> @llvm.arm64.neon.fcvtps.v1i64.v1f64(<1 x double>)
-declare <1 x i64> @llvm.arm64.neon.fcvtnu.v1i64.v1f64(<1 x double>)
-declare <1 x i64> @llvm.arm64.neon.fcvtns.v1i64.v1f64(<1 x double>)
-
-define <1 x double> @test_vrndn_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrndn_f64
-; CHECK: frintn d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.arm64.neon.frintn.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrnda_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrnda_f64
-; CHECK: frinta d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.round.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrndp_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrndp_f64
-; CHECK: frintp d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.ceil.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrndm_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrndm_f64
-; CHECK: frintm d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.floor.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrndx_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrndx_f64
-; CHECK: frintx d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.rint.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrnd_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrnd_f64
-; CHECK: frintz d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.trunc.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrndi_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrndi_f64
-; CHECK: frinti d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.nearbyint.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-declare <1 x double> @llvm.nearbyint.v1f64(<1 x double>)
-declare <1 x double> @llvm.trunc.v1f64(<1 x double>)
-declare <1 x double> @llvm.rint.v1f64(<1 x double>)
-declare <1 x double> @llvm.floor.v1f64(<1 x double>)
-declare <1 x double> @llvm.ceil.v1f64(<1 x double>)
-declare <1 x double> @llvm.round.v1f64(<1 x double>)
-declare <1 x double> @llvm.arm64.neon.frintn.v1f64(<1 x double>)
-
-define <1 x double> @test_vrsqrte_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrsqrte_f64
-; CHECK: frsqrte d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.arm64.neon.frsqrte.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrecpe_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vrecpe_f64
-; CHECK: frecpe d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.arm64.neon.frecpe.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vsqrt_f64(<1 x double> %a) {
-; CHECK-LABEL: test_vsqrt_f64
-; CHECK: fsqrt d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.sqrt.v1f64(<1 x double> %a)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrecps_f64(<1 x double> %a, <1 x double> %b) {
-; CHECK-LABEL: test_vrecps_f64
-; CHECK: frecps d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.arm64.neon.frecps.v1f64(<1 x double> %a, <1 x double> %b)
-  ret <1 x double> %1
-}
-
-define <1 x double> @test_vrsqrts_f64(<1 x double> %a, <1 x double> %b) {
-; CHECK-LABEL: test_vrsqrts_f64
-; CHECK: frsqrts d{{[0-9]+}}, d{{[0-9]+}}, d{{[0-9]+}}
-  %1 = tail call <1 x double> @llvm.arm64.neon.frsqrts.v1f64(<1 x double> %a, <1 x double> %b)
-  ret <1 x double> %1
-}
-
-declare <1 x double> @llvm.arm64.neon.frsqrts.v1f64(<1 x double>, <1 x double>)
-declare <1 x double> @llvm.arm64.neon.frecps.v1f64(<1 x double>, <1 x double>)
-declare <1 x double> @llvm.sqrt.v1f64(<1 x double>)
-declare <1 x double> @llvm.arm64.neon.frecpe.v1f64(<1 x double>)
-declare <1 x double> @llvm.arm64.neon.frsqrte.v1f64(<1 x double>)
-
-define i64 @test_vaddlv_s32(<2 x i32> %a) {
-; CHECK-LABEL: test_vaddlv_s32
-; CHECK: saddlp {{v[0-9]+}}.1d, {{v[0-9]+}}.2s
-  %1 = tail call i64 @llvm.arm64.neon.saddlv.i64.v2i32(<2 x i32> %a)
-  ret i64 %1
-}
-
-define i64 @test_vaddlv_u32(<2 x i32> %a) {
-; CHECK-LABEL: test_vaddlv_u32
-; CHECK: uaddlp {{v[0-9]+}}.1d, {{v[0-9]+}}.2s
-  %1 = tail call i64 @llvm.arm64.neon.uaddlv.i64.v2i32(<2 x i32> %a)
-  ret i64 %1
-}
-
-declare i64 @llvm.arm64.neon.saddlv.i64.v2i32(<2 x i32>)
-declare i64 @llvm.arm64.neon.uaddlv.i64.v2i32(<2 x i32>)
diff --git a/test/CodeGen/ARM64/aarch64-neon-mul-div.ll b/test/CodeGen/ARM64/aarch64-neon-mul-div.ll
new file mode 100644
index 0000000000..f3a9766319
--- /dev/null
+++ b/test/CodeGen/ARM64/aarch64-neon-mul-div.ll
@@ -0,0 +1,797 @@
+; RUN: llc < %s -verify-machineinstrs -mtriple=arm64-none-linux-gnu -mattr=+neon | FileCheck %s
+; arm64 has its own copy of this because of the intrinsics
+
+define <8 x i8> @mul8xi8(<8 x i8> %A, <8 x i8> %B) {
+; CHECK-LABEL: mul8xi8:
+; CHECK: mul {{v[0-9]+}}.8b, {{v[0-9]+}}.8b, {{v[0-9]+}}.8b
+	%tmp3 = mul <8 x i8> %A, %B;
+	ret <8 x i8> %tmp3
+}
+
+define <16 x i8> @mul16xi8(<16 x i8> %A, <16 x i8> %B) {
+; CHECK-LABEL: mul16xi8:
+; CHECK: mul {{v[0-9]+}}.16b, {{v[0-9]+}}.16b, {{v[0-9]+}}.16b
+	%tmp3 = mul <16 x i8> %A, %B;
+	ret <16 x i8> %tmp3
+}
+
+define <4 x i16> @mul4xi16(<4 x i16> %A, <4 x i16> %B) {
+; CHECK-LABEL: mul4xi16:
+; CHECK: mul {{v[0-9]+}}.4h, {{v[0-9]+}}.4h, {{v[0-9]+}}.4h
+	%tmp3 = mul <4 x i16> %A, %B;
+	ret <4 x i16> %tmp3
+}
+
+define <8 x i16> @mul8xi16(<8 x i16> %A, <8 x i16> %B) {
+; CHECK-LABEL: mul8xi16:
+; CHECK: mul {{v[0-9]+}}.8h, {{v[0-9]+}}.8h, {{v[0-9]+}}.8h
+	%tmp3 = mul <8 x i16> %A, %B;
+	ret <8 x i16> %tmp3
+}
+
+define <2 x i32> @mul2xi32(<2 x i32> %A, <2 x i32> %B) {
+; CHECK-LABEL: mul2xi32:
+; CHECK: mul {{v[0-9]+}}.2s, {{v[0-9]+}}.2s, {{v[0-9]+}}.2s
+	%tmp3 = mul <2 x i32> %A, %B;
+	ret <2 x i32> %tmp3
+}
+
+define <4 x i32> @mul4x32(<4 x i32> %A, <4 x i32> %B) {
+; CHECK-LABEL: mul4x32:
+; CHECK: mul {{v[0-9]+}}.4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s
+	%tmp3 = mul <4 x i32> %A, %B;
+	ret <4 x i32> %tmp3
+}
+
+define <1 x i64> @mul1xi64(<1 x i64> %A, <1 x i64> %B) {
+; CHECK-LABEL: mul1xi64:
+; CHECK: mul x{{[0-9]+}}, x{{[0-9]+}}, x{{[0-9]+}}
+  %tmp3 = mul <1 x i64> %A, %B;
+  ret <1 x i64> %tmp3
+}
+
+define <2 x i64> @mul2xi64(<2 x i64> %A, <2 x i64> %B) {
+; CHECK-LABEL: mul2xi64:
+; CHECK: mul x{{[0-9]+}}, x{{[0-9]+}}, x{{[0-9]+}}
+; CHECK: mul x{{[0-9]+}}, x{{[0-9]+}}, x{{[0-9]+}}
+  %tmp3 = mul <2 x i64> %A, %B;
+  ret <2 x i64> %tmp3
+}
+
+ define <2 x float> @mul2xfloat(<2 x float> %A, <2 x float> %B) {
+; CHECK-LABEL: mul2xfloat:
+; CHECK: fmul {{v[0-9]+}}.2s, {{v[0-9]+}}.2s, {{v[0-9]+}}.2s
+	%tmp3 = fmul <2 x float> %A, %B;
+	ret <2 x float> %tmp3
+}
+
+define <4 x float> @mul4xfloat(<4 x float> %A, <4 x float> %B) {
+; CHECK-LABEL: mul4xfloat:
+; CHECK: fmul {{v[0-9]+}}.4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s
+	%tmp3 = fmul <4 x float> %A, %B;
+	ret <4 x float> %tmp3
+}
+define <2 x double> @mul2xdouble(<2 x double> %A, <2 x double> %B) {
+; CHECK-LABEL: mul2xdouble:
+; CHECK: fmul {{v[0-9]+}}.2d, {{v[0-9]+}}.2d, {{v[0-9]+}}.2d
+	%tmp3 = fmul <2 x double> %A, %B;
+	ret <2 x double> %tmp3
+}
+
+
+ define <2 x float> @div2xfloat(<2 x float> %A, <2 x float> %B) {
+; CHECK-LABEL: div2xfloat:
+; CHECK: fdiv {{v[0-9]+}}.2s, {{v[0-9]+}}.2s, {{v[0-9]+}}.2s
+	%tmp3 = fdiv <2 x float> %A, %B;
+	ret <2 x float> %tmp3
+}
+
+define <4 x float> @div4xfloat(<4 x float> %A, <4 x float> %B) {
+; CHECK-LABEL: div4xfloat:
+; CHECK: fdiv {{v[0-9]+}}.4s, {{v[0-9]+}}.4s, {{v[0-9]+}}.4s
+	%tmp3 = fdiv <4 x float> %A, %B;
+	ret <4 x float> %tmp3
+}
+define <2 x double> @div2xdouble(<2 x double> %A, <2 x double> %B) {
+; CHECK-LABEL: div2xdouble:
+; CHECK: fdiv {{v[0-9]+}}.2d, {{v[0-9]+}}.2d, {{v[0-9]+}}.2d
+	%tmp3 = fdiv <2 x double> %A, %B;
+	ret <2 x double> %tmp3
+}
+
+define <1 x i8> @sdiv1x8(<1 x i8> %A, <1 x i8> %B) {
+; CHECK-LABEL: sdiv1x8:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <1 x i8> %A, %B;
+	ret <1 x i8> %tmp3
+}
+
+define <8 x i8> @sdiv8x8(<8 x i8> %A, <8 x i8> %B) {
+; CHECK-LABEL: sdiv8x8:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <8 x i8> %A, %B;
+	ret <8 x i8> %tmp3
+}
+
+define <16 x i8> @sdiv16x8(<16 x i8> %A, <16 x i8> %B) {
+; CHECK-LABEL: sdiv16x8:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <16 x i8> %A, %B;
+	ret <16 x i8> %tmp3
+}
+
+define <1 x i16> @sdiv1x16(<1 x i16> %A, <1 x i16> %B) {
+; CHECK-LABEL: sdiv1x16:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <1 x i16> %A, %B;
+	ret <1 x i16> %tmp3
+}
+
+define <4 x i16> @sdiv4x16(<4 x i16> %A, <4 x i16> %B) {
+; CHECK-LABEL: sdiv4x16:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <4 x i16> %A, %B;
+	ret <4 x i16> %tmp3
+}
+
+define <8 x i16> @sdiv8x16(<8 x i16> %A, <8 x i16> %B) {
+; CHECK-LABEL: sdiv8x16:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <8 x i16> %A, %B;
+	ret <8 x i16> %tmp3
+}
+
+define <1 x i32> @sdiv1x32(<1 x i32> %A, <1 x i32> %B) {
+; CHECK-LABEL: sdiv1x32:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <1 x i32> %A, %B;
+	ret <1 x i32> %tmp3
+}
+
+define <2 x i32> @sdiv2x32(<2 x i32> %A, <2 x i32> %B) {
+; CHECK-LABEL: sdiv2x32:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <2 x i32> %A, %B;
+	ret <2 x i32> %tmp3
+}
+
+define <4 x i32> @sdiv4x32(<4 x i32> %A, <4 x i32> %B) {
+; CHECK-LABEL: sdiv4x32:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = sdiv <4 x i32> %A, %B;
+	ret <4 x i32> %tmp3
+}
+
+define <1 x i64> @sdiv1x64(<1 x i64> %A, <1 x i64> %B) {
+; CHECK-LABEL: sdiv1x64:
+; CHECK: sdiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = sdiv <1 x i64> %A, %B;
+	ret <1 x i64> %tmp3
+}
+
+define <2 x i64> @sdiv2x64(<2 x i64> %A, <2 x i64> %B) {
+; CHECK-LABEL: sdiv2x64:
+; CHECK: sdiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: sdiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = sdiv <2 x i64> %A, %B;
+	ret <2 x i64> %tmp3
+}
+
+define <1 x i8> @udiv1x8(<1 x i8> %A, <1 x i8> %B) {
+; CHECK-LABEL: udiv1x8:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <1 x i8> %A, %B;
+	ret <1 x i8> %tmp3
+}
+
+define <8 x i8> @udiv8x8(<8 x i8> %A, <8 x i8> %B) {
+; CHECK-LABEL: udiv8x8:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <8 x i8> %A, %B;
+	ret <8 x i8> %tmp3
+}
+
+define <16 x i8> @udiv16x8(<16 x i8> %A, <16 x i8> %B) {
+; CHECK-LABEL: udiv16x8:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <16 x i8> %A, %B;
+	ret <16 x i8> %tmp3
+}
+
+define <1 x i16> @udiv1x16(<1 x i16> %A, <1 x i16> %B) {
+; CHECK-LABEL: udiv1x16:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <1 x i16> %A, %B;
+	ret <1 x i16> %tmp3
+}
+
+define <4 x i16> @udiv4x16(<4 x i16> %A, <4 x i16> %B) {
+; CHECK-LABEL: udiv4x16:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <4 x i16> %A, %B;
+	ret <4 x i16> %tmp3
+}
+
+define <8 x i16> @udiv8x16(<8 x i16> %A, <8 x i16> %B) {
+; CHECK-LABEL: udiv8x16:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <8 x i16> %A, %B;
+	ret <8 x i16> %tmp3
+}
+
+define <1 x i32> @udiv1x32(<1 x i32> %A, <1 x i32> %B) {
+; CHECK-LABEL: udiv1x32:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <1 x i32> %A, %B;
+	ret <1 x i32> %tmp3
+}
+
+define <2 x i32> @udiv2x32(<2 x i32> %A, <2 x i32> %B) {
+; CHECK-LABEL: udiv2x32:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <2 x i32> %A, %B;
+	ret <2 x i32> %tmp3
+}
+
+define <4 x i32> @udiv4x32(<4 x i32> %A, <4 x i32> %B) {
+; CHECK-LABEL: udiv4x32:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = udiv <4 x i32> %A, %B;
+	ret <4 x i32> %tmp3
+}
+
+define <1 x i64> @udiv1x64(<1 x i64> %A, <1 x i64> %B) {
+; CHECK-LABEL: udiv1x64:
+; CHECK: udiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = udiv <1 x i64> %A, %B;
+	ret <1 x i64> %tmp3
+}
+
+define <2 x i64> @udiv2x64(<2 x i64> %A, <2 x i64> %B) {
+; CHECK-LABEL: udiv2x64:
+; CHECK: udiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: udiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = udiv <2 x i64> %A, %B;
+	ret <2 x i64> %tmp3
+}
+
+define <1 x i8> @srem1x8(<1 x i8> %A, <1 x i8> %B) {
+; CHECK-LABEL: srem1x8:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <1 x i8> %A, %B;
+	ret <1 x i8> %tmp3
+}
+
+define <8 x i8> @srem8x8(<8 x i8> %A, <8 x i8> %B) {
+; CHECK-LABEL: srem8x8:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <8 x i8> %A, %B;
+	ret <8 x i8> %tmp3
+}
+
+define <16 x i8> @srem16x8(<16 x i8> %A, <16 x i8> %B) {
+; CHECK-LABEL: srem16x8:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <16 x i8> %A, %B;
+	ret <16 x i8> %tmp3
+}
+
+define <1 x i16> @srem1x16(<1 x i16> %A, <1 x i16> %B) {
+; CHECK-LABEL: srem1x16:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <1 x i16> %A, %B;
+	ret <1 x i16> %tmp3
+}
+
+define <4 x i16> @srem4x16(<4 x i16> %A, <4 x i16> %B) {
+; CHECK-LABEL: srem4x16:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <4 x i16> %A, %B;
+	ret <4 x i16> %tmp3
+}
+
+define <8 x i16> @srem8x16(<8 x i16> %A, <8 x i16> %B) {
+; CHECK-LABEL: srem8x16:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <8 x i16> %A, %B;
+	ret <8 x i16> %tmp3
+}
+
+define <1 x i32> @srem1x32(<1 x i32> %A, <1 x i32> %B) {
+; CHECK-LABEL: srem1x32:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <1 x i32> %A, %B;
+	ret <1 x i32> %tmp3
+}
+
+define <2 x i32> @srem2x32(<2 x i32> %A, <2 x i32> %B) {
+; CHECK-LABEL: srem2x32:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <2 x i32> %A, %B;
+	ret <2 x i32> %tmp3
+}
+
+define <4 x i32> @srem4x32(<4 x i32> %A, <4 x i32> %B) {
+; CHECK-LABEL: srem4x32:
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: sdiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = srem <4 x i32> %A, %B;
+	ret <4 x i32> %tmp3
+}
+
+define <1 x i64> @srem1x64(<1 x i64> %A, <1 x i64> %B) {
+; CHECK-LABEL: srem1x64:
+; CHECK: sdiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: msub {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = srem <1 x i64> %A, %B;
+	ret <1 x i64> %tmp3
+}
+
+define <2 x i64> @srem2x64(<2 x i64> %A, <2 x i64> %B) {
+; CHECK-LABEL: srem2x64:
+; CHECK: sdiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: sdiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: msub {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: msub {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = srem <2 x i64> %A, %B;
+	ret <2 x i64> %tmp3
+}
+
+define <1 x i8> @urem1x8(<1 x i8> %A, <1 x i8> %B) {
+; CHECK-LABEL: urem1x8:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <1 x i8> %A, %B;
+	ret <1 x i8> %tmp3
+}
+
+define <8 x i8> @urem8x8(<8 x i8> %A, <8 x i8> %B) {
+; CHECK-LABEL: urem8x8:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <8 x i8> %A, %B;
+	ret <8 x i8> %tmp3
+}
+
+define <16 x i8> @urem16x8(<16 x i8> %A, <16 x i8> %B) {
+; CHECK-LABEL: urem16x8:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <16 x i8> %A, %B;
+	ret <16 x i8> %tmp3
+}
+
+define <1 x i16> @urem1x16(<1 x i16> %A, <1 x i16> %B) {
+; CHECK-LABEL: urem1x16:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <1 x i16> %A, %B;
+	ret <1 x i16> %tmp3
+}
+
+define <4 x i16> @urem4x16(<4 x i16> %A, <4 x i16> %B) {
+; CHECK-LABEL: urem4x16:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <4 x i16> %A, %B;
+	ret <4 x i16> %tmp3
+}
+
+define <8 x i16> @urem8x16(<8 x i16> %A, <8 x i16> %B) {
+; CHECK-LABEL: urem8x16:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <8 x i16> %A, %B;
+	ret <8 x i16> %tmp3
+}
+
+define <1 x i32> @urem1x32(<1 x i32> %A, <1 x i32> %B) {
+; CHECK-LABEL: urem1x32:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <1 x i32> %A, %B;
+	ret <1 x i32> %tmp3
+}
+
+define <2 x i32> @urem2x32(<2 x i32> %A, <2 x i32> %B) {
+; CHECK-LABEL: urem2x32:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <2 x i32> %A, %B;
+	ret <2 x i32> %tmp3
+}
+
+define <4 x i32> @urem4x32(<4 x i32> %A, <4 x i32> %B) {
+; CHECK-LABEL: urem4x32:
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: udiv {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+; CHECK: msub {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}, {{w[0-9]+}}
+	%tmp3 = urem <4 x i32> %A, %B;
+	ret <4 x i32> %tmp3
+}
+
+define <1 x i64> @urem1x64(<1 x i64> %A, <1 x i64> %B) {
+; CHECK-LABEL: urem1x64:
+; CHECK: udiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: msub {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = urem <1 x i64> %A, %B;
+	ret <1 x i64> %tmp3
+}
+
+define <2 x i64> @urem2x64(<2 x i64> %A, <2 x i64> %B) {
+; CHECK-LABEL: urem2x64:
+; CHECK: udiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: udiv {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: msub {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+; CHECK: msub {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}, {{x[0-9]+}}
+	%tmp3 = urem <2 x i64> %A, %B;
+	ret <2 x i64> %tmp3
+}
+
+define <2 x float> @frem2f32(<2 x float> %A, <2 x float> %B) {
+; CHECK-LABEL: frem2f32:
+; CHECK: bl fmodf
+; CHECK: bl fmodf
+	%tmp3 = frem <2 x float> %A, %B;
+	ret <2 x float> %tmp3
+}
+
+define <4 x float> @frem4f32(<4 x float> %A, <4 x float> %B) {
+; CHECK-LABEL: frem4f32:
+; CHECK: bl fmodf
+; CHECK: bl fmodf
+; CHECK: bl fmodf
+; CHECK: bl fmodf
+	%tmp3 = frem <4 x float> %A, %B;
+	ret <4 x float> %tmp3
+}
+
+define <1 x double> @frem1d64(<1 x double> %A, <1 x double> %B) {
+; CHECK-LABEL: frem1d64:
+; CHECK: bl fmod
+	%tmp3 = frem <1 x double> %A, %B;
+	ret <1 x double> %tmp3
+}
+
+define <2 x double> @frem2d64(<2 x double> %A, <2 x double> %B) {
+; CHECK-LABEL: frem2d64:
+; CHECK: bl fmod
+; CHECK: bl fmod
+	%tmp3 = frem <2 x double> %A, %B;
+	ret <2 x double> %tmp3
+}
+
+declare <8 x i8> @llvm.arm64.neon.pmul.v8i8(<8 x i8>, <8 x i8>)
+declare <16 x i8> @llvm.arm64.neon.pmul.v16i8(<16 x i8>, <16 x i8>)
+
+define <8 x i8> @poly_mulv8i8(<8 x i8> %lhs, <8 x i8> %rhs) {
+; CHECK-LABEL: poly_mulv8i8:
+   %prod = call <8 x i8> @llvm.arm64.neon.pmul.v8i8(<8 x i8> %lhs, <8 x i8> %rhs)
+; CHECK: pmul v0.8b, v0.8b, v1.8b
+   ret <8 x i8> %prod
+}
+
+define <16 x i8> @poly_mulv16i8(<16 x i8> %lhs, <16 x i8> %rhs) {
+; CHECK-LABEL: poly_mulv16i8:
+   %prod = call <16 x i8> @llvm.arm64.neon.pmul.v16i8(<16 x i8> %lhs, <16 x i8> %rhs)
+; CHECK: pmul v0.16b, v0.16b, v1.16b
+   ret <16 x i8> %prod
+}
+
+declare <4 x i16> @llvm.arm64.neon.sqdmulh.v4i16(<4 x i16>, <4 x i16>)
+declare <8 x i16> @llvm.arm64.neon.sqdmulh.v8i16(<8 x i16>, <8 x i16>)
+declare <2 x i32> @llvm.arm64.neon.sqdmulh.v2i32(<2 x i32>, <2 x i32>)
+declare <4 x i32> @llvm.arm64.neon.sqdmulh.v4i32(<4 x i32>, <4 x i32>)
+
+define <4 x i16> @test_sqdmulh_v4i16(<4 x i16> %lhs, <4 x i16> %rhs) {
+; CHECK-LABEL: test_sqdmulh_v4i16:
+   %prod = call <4 x i16> @llvm.arm64.neon.sqdmulh.v4i16(<4 x i16> %lhs, <4 x i16> %rhs)
+; CHECK: sqdmulh v0.4h, v0.4h, v1.4h
+   ret <4 x i16> %prod
+}
+
+define <8 x i16> @test_sqdmulh_v8i16(<8 x i16> %lhs, <8 x i16> %rhs) {
+; CHECK-LABEL: test_sqdmulh_v8i16:
+   %prod = call <8 x i16> @llvm.arm64.neon.sqdmulh.v8i16(<8 x i16> %lhs, <8 x i16> %rhs)
+; CHECK: sqdmulh v0.8h, v0.8h, v1.8h
+   ret <8 x i16> %prod
+}
+
+define <2 x i32> @test_sqdmulh_v2i32(<2 x i32> %lhs, <2 x i32> %rhs) {
+; CHECK-LABEL: test_sqdmulh_v2i32:
+   %prod = call <2 x i32> @llvm.arm64.neon.sqdmulh.v2i32(<2 x i32> %lhs, <2 x i32> %rhs)
+; CHECK: sqdmulh v0.2s, v0.2s, v1.2s
+   ret <2 x i32> %prod
+}
+
+define <4 x i32> @test_sqdmulh_v4i32(<4 x i32> %lhs, <4 x i32> %rhs) {
+; CHECK-LABEL: test_sqdmulh_v4i32:
+   %prod = call <4 x i32> @llvm.arm64.neon.sqdmulh.v4i32(<4 x i32> %lhs, <4 x i32> %rhs)
+; CHECK: sqdmulh v0.4s, v0.4s, v1.4s
+   ret <4 x i32> %prod
+}
+
+declare <4 x i16> @llvm.arm64.neon.sqrdmulh.v4i16(<4 x i16>, <4 x i16>)
+declare <8 x i16> @llvm.arm64.neon.sqrdmulh.v8i16(<8 x i16>, <8 x i16>)
+declare <2 x i32> @llvm.arm64.neon.sqrdmulh.v2i32(<2 x i32>, <2 x i32>)
+declare <4 x i32> @llvm.arm64.neon.sqrdmulh.v4i32(<4 x i32>, <4 x i32>)
+
+define <4 x i16> @test_sqrdmulh_v4i16(<4 x i16> %lhs, <4 x i16> %rhs) {
+; CHECK-LABEL: test_sqrdmulh_v4i16:
+   %prod = call <4 x i16> @llvm.arm64.neon.sqrdmulh.v4i16(<4 x i16> %lhs, <4 x i16> %rhs)
+; CHECK: sqrdmulh v0.4h, v0.4h, v1.4h
+   ret <4 x i16> %prod
+}
+
+define <8 x i16> @test_sqrdmulh_v8i16(<8 x i16> %lhs, <8 x i16> %rhs) {
+; CHECK-LABEL: test_sqrdmulh_v8i16:
+   %prod = call <8 x i16> @llvm.arm64.neon.sqrdmulh.v8i16(<8 x i16> %lhs, <8 x i16> %rhs)
+; CHECK: sqrdmulh v0.8h, v0.8h, v1.8h
+   ret <8 x i16> %prod
+}
+
+define <2 x i32> @test_sqrdmulh_v2i32(<2 x i32> %lhs, <2 x i32> %rhs) {
+; CHECK-LABEL: test_sqrdmulh_v2i32:
+   %prod = call <2 x i32> @llvm.arm64.neon.sqrdmulh.v2i32(<2 x i32> %lhs, <2 x i32> %rhs)
+; CHECK: sqrdmulh v0.2s, v0.2s, v1.2s
+   ret <2 x i32> %prod
+}
+
+define <4 x i32> @test_sqrdmulh_v4i32(<4 x i32> %lhs, <4 x i32> %rhs) {
+; CHECK-LABEL: test_sqrdmulh_v4i32:
+   %prod = call <4 x i32> @llvm.arm64.neon.sqrdmulh.v4i32(<4 x i32> %lhs, <4 x i32> %rhs)
+; CHECK: sqrdmulh v0.4s, v0.4s, v1.4s
+   ret <4 x i32> %prod
+}
+
+declare <2 x float> @llvm.arm64.neon.fmulx.v2f32(<2 x float>, <2 x float>)
+declare <4 x float> @llvm.arm64.neon.fmulx.v4f32(<4 x float>, <4 x float>)
+declare <2 x double> @llvm.arm64.neon.fmulx.v2f64(<2 x double>, <2 x double>)
+
+define <2 x float> @fmulx_v2f32(<2 x float> %lhs, <2 x float> %rhs) {
+; CHECK-LABEL: fmulx_v2f32:
+; Using registers other than v0, v1 and v2 are possible, but would be odd.
+; CHECK: fmulx v0.2s, v0.2s, v1.2s
+        %val = call <2 x float> @llvm.arm64.neon.fmulx.v2f32(<2 x float> %lhs, <2 x float> %rhs)
+        ret <2 x float> %val
+}
+
+define <4 x float> @fmulx_v4f32(<4 x float> %lhs, <4 x float> %rhs) {
+; CHECK-LABEL: fmulx_v4f32:
+; Using registers other than v0, v1 and v2 are possible, but would be odd.
+; CHECK: fmulx v0.4s, v0.4s, v1.4s
+        %val = call <4 x float> @llvm.arm64.neon.fmulx.v4f32(<4 x float> %lhs, <4 x float> %rhs)
+        ret <4 x float> %val
+}
+
+define <2 x double> @fmulx_v2f64(<2 x double> %lhs, <2 x double> %rhs) {
+; CHECK-LABEL: fmulx_v2f64:
+; Using registers other than v0, v1 and v2 are possible, but would be odd.
+; CHECK: fmulx v0.2d, v0.2d, v1.2d
+        %val = call <2 x double> @llvm.arm64.neon.fmulx.v2f64(<2 x double> %lhs, <2 x double> %rhs)
+        ret <2 x double> %val
+}
+