; RUN: llc < %s -march=arm -mattr=+neon | FileCheck %s

define <8 x i8> @vmins8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
;CHECK: vmins8:
;CHECK: vmin.s8
	%tmp1 = load <8 x i8>* %A
	%tmp2 = load <8 x i8>* %B
	%tmp3 = call <8 x i8> @llvm.arm.neon.vmins.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
	ret <8 x i8> %tmp3
}

define <4 x i16> @vmins16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
;CHECK: vmins16:
;CHECK: vmin.s16
	%tmp1 = load <4 x i16>* %A
	%tmp2 = load <4 x i16>* %B
	%tmp3 = call <4 x i16> @llvm.arm.neon.vmins.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
	ret <4 x i16> %tmp3
}

define <2 x i32> @vmins32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
;CHECK: vmins32:
;CHECK: vmin.s32
	%tmp1 = load <2 x i32>* %A
	%tmp2 = load <2 x i32>* %B
	%tmp3 = call <2 x i32> @llvm.arm.neon.vmins.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
	ret <2 x i32> %tmp3
}

define <8 x i8> @vminu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
;CHECK: vminu8:
;CHECK: vmin.u8
	%tmp1 = load <8 x i8>* %A
	%tmp2 = load <8 x i8>* %B
	%tmp3 = call <8 x i8> @llvm.arm.neon.vminu.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
	ret <8 x i8> %tmp3
}

define <4 x i16> @vminu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
;CHECK: vminu16:
;CHECK: vmin.u16
	%tmp1 = load <4 x i16>* %A
	%tmp2 = load <4 x i16>* %B
	%tmp3 = call <4 x i16> @llvm.arm.neon.vminu.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
	ret <4 x i16> %tmp3
}

define <2 x i32> @vminu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
;CHECK: vminu32:
;CHECK: vmin.u32
	%tmp1 = load <2 x i32>* %A
	%tmp2 = load <2 x i32>* %B
	%tmp3 = call <2 x i32> @llvm.arm.neon.vminu.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
	ret <2 x i32> %tmp3
}

define <2 x float> @vminf32(<2 x float>* %A, <2 x float>* %B) nounwind {
;CHECK: vminf32:
;CHECK: vmin.f32
	%tmp1 = load <2 x float>* %A
	%tmp2 = load <2 x float>* %B
	%tmp3 = call <2 x float> @llvm.arm.neon.vmins.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
	ret <2 x float> %tmp3
}

define <16 x i8> @vminQs8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
;CHECK: vminQs8:
;CHECK: vmin.s8
	%tmp1 = load <16 x i8>* %A
	%tmp2 = load <16 x i8>* %B
	%tmp3 = call <16 x i8> @llvm.arm.neon.vmins.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
	ret <16 x i8> %tmp3
}

define <8 x i16> @vminQs16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
;CHECK: vminQs16:
;CHECK: vmin.s16
	%tmp1 = load <8 x i16>* %A
	%tmp2 = load <8 x i16>* %B
	%tmp3 = call <8 x i16> @llvm.arm.neon.vmins.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
	ret <8 x i16> %tmp3
}

define <4 x i32> @vminQs32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
;CHECK: vminQs32:
;CHECK: vmin.s32
	%tmp1 = load <4 x i32>* %A
	%tmp2 = load <4 x i32>* %B
	%tmp3 = call <4 x i32> @llvm.arm.neon.vmins.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
	ret <4 x i32> %tmp3
}

define <16 x i8> @vminQu8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
;CHECK: vminQu8:
;CHECK: vmin.u8
	%tmp1 = load <16 x i8>* %A
	%tmp2 = load <16 x i8>* %B
	%tmp3 = call <16 x i8> @llvm.arm.neon.vminu.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
	ret <16 x i8> %tmp3
}

define <8 x i16> @vminQu16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
;CHECK: vminQu16:
;CHECK: vmin.u16
	%tmp1 = load <8 x i16>* %A
	%tmp2 = load <8 x i16>* %B
	%tmp3 = call <8 x i16> @llvm.arm.neon.vminu.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
	ret <8 x i16> %tmp3
}

define <4 x i32> @vminQu32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
;CHECK: vminQu32:
;CHECK: vmin.u32
	%tmp1 = load <4 x i32>* %A
	%tmp2 = load <4 x i32>* %B
	%tmp3 = call <4 x i32> @llvm.arm.neon.vminu.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
	ret <4 x i32> %tmp3
}

define <4 x float> @vminQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
;CHECK: vminQf32:
;CHECK: vmin.f32
	%tmp1 = load <4 x float>* %A
	%tmp2 = load <4 x float>* %B
	%tmp3 = call <4 x float> @llvm.arm.neon.vmins.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
	ret <4 x float> %tmp3
}

declare <8 x i8>  @llvm.arm.neon.vmins.v8i8(<8 x i8>, <8 x i8>) nounwind readnone
declare <4 x i16> @llvm.arm.neon.vmins.v4i16(<4 x i16>, <4 x i16>) nounwind readnone
declare <2 x i32> @llvm.arm.neon.vmins.v2i32(<2 x i32>, <2 x i32>) nounwind readnone

declare <8 x i8>  @llvm.arm.neon.vminu.v8i8(<8 x i8>, <8 x i8>) nounwind readnone
declare <4 x i16> @llvm.arm.neon.vminu.v4i16(<4 x i16>, <4 x i16>) nounwind readnone
declare <2 x i32> @llvm.arm.neon.vminu.v2i32(<2 x i32>, <2 x i32>) nounwind readnone

declare <2 x float> @llvm.arm.neon.vmins.v2f32(<2 x float>, <2 x float>) nounwind readnone

declare <16 x i8> @llvm.arm.neon.vmins.v16i8(<16 x i8>, <16 x i8>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vmins.v8i16(<8 x i16>, <8 x i16>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vmins.v4i32(<4 x i32>, <4 x i32>) nounwind readnone

declare <16 x i8> @llvm.arm.neon.vminu.v16i8(<16 x i8>, <16 x i8>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vminu.v8i16(<8 x i16>, <8 x i16>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vminu.v4i32(<4 x i32>, <4 x i32>) nounwind readnone

declare <4 x float> @llvm.arm.neon.vmins.v4f32(<4 x float>, <4 x float>) nounwind readnone

define <8 x i8> @vmaxs8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
;CHECK: vmaxs8:
;CHECK: vmax.s8
	%tmp1 = load <8 x i8>* %A
	%tmp2 = load <8 x i8>* %B
	%tmp3 = call <8 x i8> @llvm.arm.neon.vmaxs.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
	ret <8 x i8> %tmp3
}

define <4 x i16> @vmaxs16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
;CHECK: vmaxs16:
;CHECK: vmax.s16
	%tmp1 = load <4 x i16>* %A
	%tmp2 = load <4 x i16>* %B
	%tmp3 = call <4 x i16> @llvm.arm.neon.vmaxs.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
	ret <4 x i16> %tmp3
}

define <2 x i32> @vmaxs32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
;CHECK: vmaxs32:
;CHECK: vmax.s32
	%tmp1 = load <2 x i32>* %A
	%tmp2 = load <2 x i32>* %B
	%tmp3 = call <2 x i32> @llvm.arm.neon.vmaxs.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
	ret <2 x i32> %tmp3
}

define <8 x i8> @vmaxu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
;CHECK: vmaxu8:
;CHECK: vmax.u8
	%tmp1 = load <8 x i8>* %A
	%tmp2 = load <8 x i8>* %B
	%tmp3 = call <8 x i8> @llvm.arm.neon.vmaxu.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
	ret <8 x i8> %tmp3
}

define <4 x i16> @vmaxu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
;CHECK: vmaxu16:
;CHECK: vmax.u16
	%tmp1 = load <4 x i16>* %A
	%tmp2 = load <4 x i16>* %B
	%tmp3 = call <4 x i16> @llvm.arm.neon.vmaxu.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
	ret <4 x i16> %tmp3
}

define <2 x i32> @vmaxu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
;CHECK: vmaxu32:
;CHECK: vmax.u32
	%tmp1 = load <2 x i32>* %A
	%tmp2 = load <2 x i32>* %B
	%tmp3 = call <2 x i32> @llvm.arm.neon.vmaxu.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
	ret <2 x i32> %tmp3
}

define <2 x float> @vmaxf32(<2 x float>* %A, <2 x float>* %B) nounwind {
;CHECK: vmaxf32:
;CHECK: vmax.f32
	%tmp1 = load <2 x float>* %A
	%tmp2 = load <2 x float>* %B
	%tmp3 = call <2 x float> @llvm.arm.neon.vmaxs.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
	ret <2 x float> %tmp3
}

define <16 x i8> @vmaxQs8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
;CHECK: vmaxQs8:
;CHECK: vmax.s8
	%tmp1 = load <16 x i8>* %A
	%tmp2 = load <16 x i8>* %B
	%tmp3 = call <16 x i8> @llvm.arm.neon.vmaxs.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
	ret <16 x i8> %tmp3
}

define <8 x i16> @vmaxQs16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
;CHECK: vmaxQs16:
;CHECK: vmax.s16
	%tmp1 = load <8 x i16>* %A
	%tmp2 = load <8 x i16>* %B
	%tmp3 = call <8 x i16> @llvm.arm.neon.vmaxs.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
	ret <8 x i16> %tmp3
}

define <4 x i32> @vmaxQs32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
;CHECK: vmaxQs32:
;CHECK: vmax.s32
	%tmp1 = load <4 x i32>* %A
	%tmp2 = load <4 x i32>* %B
	%tmp3 = call <4 x i32> @llvm.arm.neon.vmaxs.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
	ret <4 x i32> %tmp3
}

define <16 x i8> @vmaxQu8(<16 x i8>* %A, <16 x i8>* %B) nounwind {
;CHECK: vmaxQu8:
;CHECK: vmax.u8
	%tmp1 = load <16 x i8>* %A
	%tmp2 = load <16 x i8>* %B
	%tmp3 = call <16 x i8> @llvm.arm.neon.vmaxu.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
	ret <16 x i8> %tmp3
}

define <8 x i16> @vmaxQu16(<8 x i16>* %A, <8 x i16>* %B) nounwind {
;CHECK: vmaxQu16:
;CHECK: vmax.u16
	%tmp1 = load <8 x i16>* %A
	%tmp2 = load <8 x i16>* %B
	%tmp3 = call <8 x i16> @llvm.arm.neon.vmaxu.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
	ret <8 x i16> %tmp3
}

define <4 x i32> @vmaxQu32(<4 x i32>* %A, <4 x i32>* %B) nounwind {
;CHECK: vmaxQu32:
;CHECK: vmax.u32
	%tmp1 = load <4 x i32>* %A
	%tmp2 = load <4 x i32>* %B
	%tmp3 = call <4 x i32> @llvm.arm.neon.vmaxu.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
	ret <4 x i32> %tmp3
}

define <4 x float> @vmaxQf32(<4 x float>* %A, <4 x float>* %B) nounwind {
;CHECK: vmaxQf32:
;CHECK: vmax.f32
	%tmp1 = load <4 x float>* %A
	%tmp2 = load <4 x float>* %B
	%tmp3 = call <4 x float> @llvm.arm.neon.vmaxs.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
	ret <4 x float> %tmp3
}

declare <8 x i8>  @llvm.arm.neon.vmaxs.v8i8(<8 x i8>, <8 x i8>) nounwind readnone
declare <4 x i16> @llvm.arm.neon.vmaxs.v4i16(<4 x i16>, <4 x i16>) nounwind readnone
declare <2 x i32> @llvm.arm.neon.vmaxs.v2i32(<2 x i32>, <2 x i32>) nounwind readnone

declare <8 x i8>  @llvm.arm.neon.vmaxu.v8i8(<8 x i8>, <8 x i8>) nounwind readnone
declare <4 x i16> @llvm.arm.neon.vmaxu.v4i16(<4 x i16>, <4 x i16>) nounwind readnone
declare <2 x i32> @llvm.arm.neon.vmaxu.v2i32(<2 x i32>, <2 x i32>) nounwind readnone

declare <2 x float> @llvm.arm.neon.vmaxs.v2f32(<2 x float>, <2 x float>) nounwind readnone

declare <16 x i8> @llvm.arm.neon.vmaxs.v16i8(<16 x i8>, <16 x i8>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vmaxs.v8i16(<8 x i16>, <8 x i16>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vmaxs.v4i32(<4 x i32>, <4 x i32>) nounwind readnone

declare <16 x i8> @llvm.arm.neon.vmaxu.v16i8(<16 x i8>, <16 x i8>) nounwind readnone
declare <8 x i16> @llvm.arm.neon.vmaxu.v8i16(<8 x i16>, <8 x i16>) nounwind readnone
declare <4 x i32> @llvm.arm.neon.vmaxu.v4i32(<4 x i32>, <4 x i32>) nounwind readnone

declare <4 x float> @llvm.arm.neon.vmaxs.v4f32(<4 x float>, <4 x float>) nounwind readnone