; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=arm-eabi -mattr=+neon | FileCheck %s

define <8 x i8> @vceqi8(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqi8:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vldr d16, [r1]
; CHECK-NEXT:    vldr d17, [r0]
; CHECK-NEXT:    vceq.i8 d16, d17, d16
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <8 x i8>, ptr %A
	%tmp2 = load <8 x i8>, ptr %B
	%tmp3 = icmp eq <8 x i8> %tmp1, %tmp2
	%tmp4 = sext <8 x i1> %tmp3 to <8 x i8>
	ret <8 x i8> %tmp4
}

define <4 x i16> @vceqi16(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqi16:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vldr d16, [r1]
; CHECK-NEXT:    vldr d17, [r0]
; CHECK-NEXT:    vceq.i16 d16, d17, d16
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <4 x i16>, ptr %A
	%tmp2 = load <4 x i16>, ptr %B
	%tmp3 = icmp eq <4 x i16> %tmp1, %tmp2
	%tmp4 = sext <4 x i1> %tmp3 to <4 x i16>
	ret <4 x i16> %tmp4
}

define <2 x i32> @vceqi32(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqi32:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vldr d16, [r1]
; CHECK-NEXT:    vldr d17, [r0]
; CHECK-NEXT:    vceq.i32 d16, d17, d16
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <2 x i32>, ptr %A
	%tmp2 = load <2 x i32>, ptr %B
	%tmp3 = icmp eq <2 x i32> %tmp1, %tmp2
	%tmp4 = sext <2 x i1> %tmp3 to <2 x i32>
	ret <2 x i32> %tmp4
}

define <2 x i32> @vceqf32(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqf32:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vldr d16, [r1]
; CHECK-NEXT:    vldr d17, [r0]
; CHECK-NEXT:    vceq.f32 d16, d17, d16
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <2 x float>, ptr %A
	%tmp2 = load <2 x float>, ptr %B
	%tmp3 = fcmp oeq <2 x float> %tmp1, %tmp2
	%tmp4 = sext <2 x i1> %tmp3 to <2 x i32>
	ret <2 x i32> %tmp4
}

define <16 x i8> @vceqQi8(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqQi8:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
; CHECK-NEXT:    vceq.i8 q8, q9, q8
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    vmov r2, r3, d17
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <16 x i8>, ptr %A
	%tmp2 = load <16 x i8>, ptr %B
	%tmp3 = icmp eq <16 x i8> %tmp1, %tmp2
	%tmp4 = sext <16 x i1> %tmp3 to <16 x i8>
	ret <16 x i8> %tmp4
}

define <8 x i16> @vceqQi16(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqQi16:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
; CHECK-NEXT:    vceq.i16 q8, q9, q8
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    vmov r2, r3, d17
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <8 x i16>, ptr %A
	%tmp2 = load <8 x i16>, ptr %B
	%tmp3 = icmp eq <8 x i16> %tmp1, %tmp2
	%tmp4 = sext <8 x i1> %tmp3 to <8 x i16>
	ret <8 x i16> %tmp4
}

define <4 x i32> @vceqQi32(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqQi32:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
; CHECK-NEXT:    vceq.i32 q8, q9, q8
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    vmov r2, r3, d17
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <4 x i32>, ptr %A
	%tmp2 = load <4 x i32>, ptr %B
	%tmp3 = icmp eq <4 x i32> %tmp1, %tmp2
	%tmp4 = sext <4 x i1> %tmp3 to <4 x i32>
	ret <4 x i32> %tmp4
}

define <4 x i32> @vceqQf32(ptr %A, ptr %B) nounwind {
; CHECK-LABEL: vceqQf32:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vld1.64 {d16, d17}, [r1]
; CHECK-NEXT:    vld1.64 {d18, d19}, [r0]
; CHECK-NEXT:    vceq.f32 q8, q9, q8
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    vmov r2, r3, d17
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <4 x float>, ptr %A
	%tmp2 = load <4 x float>, ptr %B
	%tmp3 = fcmp oeq <4 x float> %tmp1, %tmp2
	%tmp4 = sext <4 x i1> %tmp3 to <4 x i32>
	ret <4 x i32> %tmp4
}

define <8 x i8> @vceqi8Z(ptr %A) nounwind {
; CHECK-LABEL: vceqi8Z:
; CHECK:       @ %bb.0:
; CHECK-NEXT:    vldr d16, [r0]
; CHECK-NEXT:    vceq.i8 d16, d16, #0
; CHECK-NEXT:    vmov r0, r1, d16
; CHECK-NEXT:    mov pc, lr
	%tmp1 = load <8 x i8>, ptr %A
	%tmp3 = icmp eq <8 x i8> %tmp1, <i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0, i8 0>
	%tmp4 = sext <8 x i1> %tmp3 to <8 x i8>
	ret <8 x i8> %tmp4
}