; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=aarch64 | FileCheck %s --check-prefixes=CHECKLE
; RUN: llc < %s -mtriple=aarch64_be | FileCheck %s --check-prefixes=CHECKBE

define <8 x i8> @vtrni8(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrni8:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr d0, [x0]
; CHECKLE-NEXT:    ldr d1, [x1]
; CHECKLE-NEXT:    trn1 v2.8b, v0.8b, v1.8b
; CHECKLE-NEXT:    trn2 v0.8b, v0.8b, v1.8b
; CHECKLE-NEXT:    add v0.8b, v2.8b, v0.8b
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrni8:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.8b }, [x0]
; CHECKBE-NEXT:    ld1 { v1.8b }, [x1]
; CHECKBE-NEXT:    trn1 v2.8b, v0.8b, v1.8b
; CHECKBE-NEXT:    trn2 v0.8b, v0.8b, v1.8b
; CHECKBE-NEXT:    add v0.8b, v2.8b, v0.8b
; CHECKBE-NEXT:    rev64 v0.8b, v0.8b
; CHECKBE-NEXT:    ret
	%tmp1 = load <8 x i8>, ptr %A
	%tmp2 = load <8 x i8>, ptr %B
	%tmp3 = shufflevector <8 x i8> %tmp1, <8 x i8> %tmp2, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
	%tmp4 = shufflevector <8 x i8> %tmp1, <8 x i8> %tmp2, <8 x i32> <i32 1, i32 9, i32 3, i32 11, i32 5, i32 13, i32 7, i32 15>
        %tmp5 = add <8 x i8> %tmp3, %tmp4
	ret <8 x i8> %tmp5
}

define <4 x i16> @vtrni16(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrni16:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr d0, [x0]
; CHECKLE-NEXT:    ldr d1, [x1]
; CHECKLE-NEXT:    trn1 v2.4h, v0.4h, v1.4h
; CHECKLE-NEXT:    trn2 v0.4h, v0.4h, v1.4h
; CHECKLE-NEXT:    add v0.4h, v2.4h, v0.4h
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrni16:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.4h }, [x0]
; CHECKBE-NEXT:    ld1 { v1.4h }, [x1]
; CHECKBE-NEXT:    trn1 v2.4h, v0.4h, v1.4h
; CHECKBE-NEXT:    trn2 v0.4h, v0.4h, v1.4h
; CHECKBE-NEXT:    add v0.4h, v2.4h, v0.4h
; CHECKBE-NEXT:    rev64 v0.4h, v0.4h
; CHECKBE-NEXT:    ret
	%tmp1 = load <4 x i16>, ptr %A
	%tmp2 = load <4 x i16>, ptr %B
	%tmp3 = shufflevector <4 x i16> %tmp1, <4 x i16> %tmp2, <4 x i32> <i32 0, i32 4, i32 2, i32 6>
	%tmp4 = shufflevector <4 x i16> %tmp1, <4 x i16> %tmp2, <4 x i32> <i32 1, i32 5, i32 3, i32 7>
        %tmp5 = add <4 x i16> %tmp3, %tmp4
	ret <4 x i16> %tmp5
}

define <8 x i8> @vtrni16_viabitcast(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrni16_viabitcast:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr d0, [x0]
; CHECKLE-NEXT:    ldr d1, [x1]
; CHECKLE-NEXT:    trn1 v0.4h, v0.4h, v1.4h
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrni16_viabitcast:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.4h }, [x0]
; CHECKBE-NEXT:    ld1 { v1.4h }, [x1]
; CHECKBE-NEXT:    trn1 v0.4h, v0.4h, v1.4h
; CHECKBE-NEXT:    rev64 v0.4h, v0.4h
; CHECKBE-NEXT:    ret
  %l1 = load <4 x i16>, ptr %A
  %l2 = load <4 x i16>, ptr %B
  %b1 = bitcast <4 x i16> %l1 to <8 x i8>
  %b2 = bitcast <4 x i16> %l2 to <8 x i8>
  %tmp3 = shufflevector <8 x i8> %b1, <8 x i8> %b2, <8 x i32> <i32 0, i32 1, i32 8, i32 9, i32 4, i32 5, i32 12, i32 13>
  ret <8 x i8> %tmp3
}

; 2xi32 TRN is redundant with ZIP
define <2 x i32> @vtrni32(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrni32:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr d0, [x0]
; CHECKLE-NEXT:    ldr d1, [x1]
; CHECKLE-NEXT:    zip1 v2.2s, v0.2s, v1.2s
; CHECKLE-NEXT:    zip2 v0.2s, v0.2s, v1.2s
; CHECKLE-NEXT:    add v0.2s, v2.2s, v0.2s
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrni32:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.2s }, [x0]
; CHECKBE-NEXT:    ld1 { v1.2s }, [x1]
; CHECKBE-NEXT:    zip1 v2.2s, v0.2s, v1.2s
; CHECKBE-NEXT:    zip2 v0.2s, v0.2s, v1.2s
; CHECKBE-NEXT:    add v0.2s, v2.2s, v0.2s
; CHECKBE-NEXT:    rev64 v0.2s, v0.2s
; CHECKBE-NEXT:    ret
	%tmp1 = load <2 x i32>, ptr %A
	%tmp2 = load <2 x i32>, ptr %B
	%tmp3 = shufflevector <2 x i32> %tmp1, <2 x i32> %tmp2, <2 x i32> <i32 0, i32 2>
	%tmp4 = shufflevector <2 x i32> %tmp1, <2 x i32> %tmp2, <2 x i32> <i32 1, i32 3>
        %tmp5 = add <2 x i32> %tmp3, %tmp4
	ret <2 x i32> %tmp5
}

define <2 x float> @vtrnf(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrnf:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr d0, [x0]
; CHECKLE-NEXT:    ldr d1, [x1]
; CHECKLE-NEXT:    zip1 v2.2s, v0.2s, v1.2s
; CHECKLE-NEXT:    zip2 v0.2s, v0.2s, v1.2s
; CHECKLE-NEXT:    fadd v0.2s, v2.2s, v0.2s
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrnf:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.2s }, [x0]
; CHECKBE-NEXT:    ld1 { v1.2s }, [x1]
; CHECKBE-NEXT:    zip1 v2.2s, v0.2s, v1.2s
; CHECKBE-NEXT:    zip2 v0.2s, v0.2s, v1.2s
; CHECKBE-NEXT:    fadd v0.2s, v2.2s, v0.2s
; CHECKBE-NEXT:    rev64 v0.2s, v0.2s
; CHECKBE-NEXT:    ret
	%tmp1 = load <2 x float>, ptr %A
	%tmp2 = load <2 x float>, ptr %B
	%tmp3 = shufflevector <2 x float> %tmp1, <2 x float> %tmp2, <2 x i32> <i32 0, i32 2>
	%tmp4 = shufflevector <2 x float> %tmp1, <2 x float> %tmp2, <2 x i32> <i32 1, i32 3>
        %tmp5 = fadd <2 x float> %tmp3, %tmp4
	ret <2 x float> %tmp5
}

define <16 x i8> @vtrnQi8(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrnQi8:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr q0, [x0]
; CHECKLE-NEXT:    ldr q1, [x1]
; CHECKLE-NEXT:    trn1 v2.16b, v0.16b, v1.16b
; CHECKLE-NEXT:    trn2 v0.16b, v0.16b, v1.16b
; CHECKLE-NEXT:    add v0.16b, v2.16b, v0.16b
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrnQi8:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.16b }, [x0]
; CHECKBE-NEXT:    ld1 { v1.16b }, [x1]
; CHECKBE-NEXT:    trn1 v2.16b, v0.16b, v1.16b
; CHECKBE-NEXT:    trn2 v0.16b, v0.16b, v1.16b
; CHECKBE-NEXT:    add v0.16b, v2.16b, v0.16b
; CHECKBE-NEXT:    rev64 v0.16b, v0.16b
; CHECKBE-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
; CHECKBE-NEXT:    ret
	%tmp1 = load <16 x i8>, ptr %A
	%tmp2 = load <16 x i8>, ptr %B
	%tmp3 = shufflevector <16 x i8> %tmp1, <16 x i8> %tmp2, <16 x i32> <i32 0, i32 16, i32 2, i32 18, i32 4, i32 20, i32 6, i32 22, i32 8, i32 24, i32 10, i32 26, i32 12, i32 28, i32 14, i32 30>
	%tmp4 = shufflevector <16 x i8> %tmp1, <16 x i8> %tmp2, <16 x i32> <i32 1, i32 17, i32 3, i32 19, i32 5, i32 21, i32 7, i32 23, i32 9, i32 25, i32 11, i32 27, i32 13, i32 29, i32 15, i32 31>
        %tmp5 = add <16 x i8> %tmp3, %tmp4
	ret <16 x i8> %tmp5
}

define <8 x i16> @vtrnQi16(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrnQi16:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr q0, [x0]
; CHECKLE-NEXT:    ldr q1, [x1]
; CHECKLE-NEXT:    trn1 v2.8h, v0.8h, v1.8h
; CHECKLE-NEXT:    trn2 v0.8h, v0.8h, v1.8h
; CHECKLE-NEXT:    add v0.8h, v2.8h, v0.8h
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrnQi16:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.8h }, [x0]
; CHECKBE-NEXT:    ld1 { v1.8h }, [x1]
; CHECKBE-NEXT:    trn1 v2.8h, v0.8h, v1.8h
; CHECKBE-NEXT:    trn2 v0.8h, v0.8h, v1.8h
; CHECKBE-NEXT:    add v0.8h, v2.8h, v0.8h
; CHECKBE-NEXT:    rev64 v0.8h, v0.8h
; CHECKBE-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
; CHECKBE-NEXT:    ret
	%tmp1 = load <8 x i16>, ptr %A
	%tmp2 = load <8 x i16>, ptr %B
	%tmp3 = shufflevector <8 x i16> %tmp1, <8 x i16> %tmp2, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
	%tmp4 = shufflevector <8 x i16> %tmp1, <8 x i16> %tmp2, <8 x i32> <i32 1, i32 9, i32 3, i32 11, i32 5, i32 13, i32 7, i32 15>
        %tmp5 = add <8 x i16> %tmp3, %tmp4
	ret <8 x i16> %tmp5
}

define <4 x i32> @vtrnQi32(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrnQi32:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr q0, [x0]
; CHECKLE-NEXT:    ldr q1, [x1]
; CHECKLE-NEXT:    trn1 v2.4s, v0.4s, v1.4s
; CHECKLE-NEXT:    trn2 v0.4s, v0.4s, v1.4s
; CHECKLE-NEXT:    add v0.4s, v2.4s, v0.4s
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrnQi32:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.4s }, [x0]
; CHECKBE-NEXT:    ld1 { v1.4s }, [x1]
; CHECKBE-NEXT:    trn1 v2.4s, v0.4s, v1.4s
; CHECKBE-NEXT:    trn2 v0.4s, v0.4s, v1.4s
; CHECKBE-NEXT:    add v0.4s, v2.4s, v0.4s
; CHECKBE-NEXT:    rev64 v0.4s, v0.4s
; CHECKBE-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
; CHECKBE-NEXT:    ret
	%tmp1 = load <4 x i32>, ptr %A
	%tmp2 = load <4 x i32>, ptr %B
	%tmp3 = shufflevector <4 x i32> %tmp1, <4 x i32> %tmp2, <4 x i32> <i32 0, i32 4, i32 2, i32 6>
	%tmp4 = shufflevector <4 x i32> %tmp1, <4 x i32> %tmp2, <4 x i32> <i32 1, i32 5, i32 3, i32 7>
        %tmp5 = add <4 x i32> %tmp3, %tmp4
	ret <4 x i32> %tmp5
}

define <4 x float> @vtrnQf(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrnQf:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr q0, [x0]
; CHECKLE-NEXT:    ldr q1, [x1]
; CHECKLE-NEXT:    trn1 v2.4s, v0.4s, v1.4s
; CHECKLE-NEXT:    trn2 v0.4s, v0.4s, v1.4s
; CHECKLE-NEXT:    fadd v0.4s, v2.4s, v0.4s
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrnQf:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.4s }, [x0]
; CHECKBE-NEXT:    ld1 { v1.4s }, [x1]
; CHECKBE-NEXT:    trn1 v2.4s, v0.4s, v1.4s
; CHECKBE-NEXT:    trn2 v0.4s, v0.4s, v1.4s
; CHECKBE-NEXT:    fadd v0.4s, v2.4s, v0.4s
; CHECKBE-NEXT:    rev64 v0.4s, v0.4s
; CHECKBE-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
; CHECKBE-NEXT:    ret
	%tmp1 = load <4 x float>, ptr %A
	%tmp2 = load <4 x float>, ptr %B
	%tmp3 = shufflevector <4 x float> %tmp1, <4 x float> %tmp2, <4 x i32> <i32 0, i32 4, i32 2, i32 6>
	%tmp4 = shufflevector <4 x float> %tmp1, <4 x float> %tmp2, <4 x i32> <i32 1, i32 5, i32 3, i32 7>
        %tmp5 = fadd <4 x float> %tmp3, %tmp4
	ret <4 x float> %tmp5
}

; Undef shuffle indices should not prevent matching to VTRN:

define <8 x i8> @vtrni8_undef(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrni8_undef:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr d0, [x0]
; CHECKLE-NEXT:    ldr d1, [x1]
; CHECKLE-NEXT:    trn1 v2.8b, v0.8b, v1.8b
; CHECKLE-NEXT:    trn2 v0.8b, v0.8b, v1.8b
; CHECKLE-NEXT:    add v0.8b, v2.8b, v0.8b
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrni8_undef:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.8b }, [x0]
; CHECKBE-NEXT:    ld1 { v1.8b }, [x1]
; CHECKBE-NEXT:    trn1 v2.8b, v0.8b, v1.8b
; CHECKBE-NEXT:    trn2 v0.8b, v0.8b, v1.8b
; CHECKBE-NEXT:    add v0.8b, v2.8b, v0.8b
; CHECKBE-NEXT:    rev64 v0.8b, v0.8b
; CHECKBE-NEXT:    ret
	%tmp1 = load <8 x i8>, ptr %A
	%tmp2 = load <8 x i8>, ptr %B
	%tmp3 = shufflevector <8 x i8> %tmp1, <8 x i8> %tmp2, <8 x i32> <i32 0, i32 undef, i32 2, i32 10, i32 undef, i32 12, i32 6, i32 14>
	%tmp4 = shufflevector <8 x i8> %tmp1, <8 x i8> %tmp2, <8 x i32> <i32 1, i32 9, i32 3, i32 11, i32 5, i32 undef, i32 undef, i32 15>
        %tmp5 = add <8 x i8> %tmp3, %tmp4
	ret <8 x i8> %tmp5
}

define <8 x i16> @vtrnQi16_undef(ptr %A, ptr %B) nounwind {
; CHECKLE-LABEL: vtrnQi16_undef:
; CHECKLE:       // %bb.0:
; CHECKLE-NEXT:    ldr q0, [x0]
; CHECKLE-NEXT:    ldr q1, [x1]
; CHECKLE-NEXT:    trn1 v2.8h, v0.8h, v1.8h
; CHECKLE-NEXT:    trn2 v0.8h, v0.8h, v1.8h
; CHECKLE-NEXT:    add v0.8h, v2.8h, v0.8h
; CHECKLE-NEXT:    ret
;
; CHECKBE-LABEL: vtrnQi16_undef:
; CHECKBE:       // %bb.0:
; CHECKBE-NEXT:    ld1 { v0.8h }, [x0]
; CHECKBE-NEXT:    ld1 { v1.8h }, [x1]
; CHECKBE-NEXT:    trn1 v2.8h, v0.8h, v1.8h
; CHECKBE-NEXT:    trn2 v0.8h, v0.8h, v1.8h
; CHECKBE-NEXT:    add v0.8h, v2.8h, v0.8h
; CHECKBE-NEXT:    rev64 v0.8h, v0.8h
; CHECKBE-NEXT:    ext v0.16b, v0.16b, v0.16b, #8
; CHECKBE-NEXT:    ret
	%tmp1 = load <8 x i16>, ptr %A
	%tmp2 = load <8 x i16>, ptr %B
	%tmp3 = shufflevector <8 x i16> %tmp1, <8 x i16> %tmp2, <8 x i32> <i32 0, i32 8, i32 undef, i32 undef, i32 4, i32 12, i32 6, i32 14>
	%tmp4 = shufflevector <8 x i16> %tmp1, <8 x i16> %tmp2, <8 x i32> <i32 1, i32 undef, i32 3, i32 11, i32 5, i32 13, i32 undef, i32 undef>
        %tmp5 = add <8 x i16> %tmp3, %tmp4
	ret <8 x i16> %tmp5
}