declare <4 x i32> @llvm.arm.neon.vqdmull.v4i32(<4 x i16>, <4 x i16>) nounwind readnone
declare <2 x i64> @llvm.arm.neon.vqdmull.v2i64(<2 x i32>, <2 x i32>) nounwind readnone
-define <4 x i32> @vqdmlals16(<4 x i32>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind {
-;CHECK-LABEL: vqdmlals16:
-;CHECK: vqdmlal.s16
- %tmp1 = load <4 x i32>* %A
- %tmp2 = load <4 x i16>* %B
- %tmp3 = load <4 x i16>* %C
- %tmp4 = call <4 x i32> @llvm.arm.neon.vqdmlal.v4i32(<4 x i32> %tmp1, <4 x i16> %tmp2, <4 x i16> %tmp3)
- ret <4 x i32> %tmp4
-}
-
-define <2 x i64> @vqdmlals32(<2 x i64>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind {
-;CHECK-LABEL: vqdmlals32:
-;CHECK: vqdmlal.s32
- %tmp1 = load <2 x i64>* %A
- %tmp2 = load <2 x i32>* %B
- %tmp3 = load <2 x i32>* %C
- %tmp4 = call <2 x i64> @llvm.arm.neon.vqdmlal.v2i64(<2 x i64> %tmp1, <2 x i32> %tmp2, <2 x i32> %tmp3)
- ret <2 x i64> %tmp4
-}
-
-define arm_aapcs_vfpcc <4 x i32> @test_vqdmlal_lanes16(<4 x i32> %arg0_int32x4_t, <4 x i16> %arg1_int16x4_t, <4 x i16> %arg2_int16x4_t) nounwind readnone {
-entry:
-; CHECK: test_vqdmlal_lanes16
-; CHECK: vqdmlal.s16 q0, d2, d3[1]
- %0 = shufflevector <4 x i16> %arg2_int16x4_t, <4 x i16> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1> ; <<4 x i16>> [#uses=1]
- %1 = tail call <4 x i32> @llvm.arm.neon.vqdmlal.v4i32(<4 x i32> %arg0_int32x4_t, <4 x i16> %arg1_int16x4_t, <4 x i16> %0) ; <<4 x i32>> [#uses=1]
- ret <4 x i32> %1
-}
-
-define arm_aapcs_vfpcc <2 x i64> @test_vqdmlal_lanes32(<2 x i64> %arg0_int64x2_t, <2 x i32> %arg1_int32x2_t, <2 x i32> %arg2_int32x2_t) nounwind readnone {
-entry:
-; CHECK: test_vqdmlal_lanes32
-; CHECK: vqdmlal.s32 q0, d2, d3[1]
- %0 = shufflevector <2 x i32> %arg2_int32x2_t, <2 x i32> undef, <2 x i32> <i32 1, i32 1> ; <<2 x i32>> [#uses=1]
- %1 = tail call <2 x i64> @llvm.arm.neon.vqdmlal.v2i64(<2 x i64> %arg0_int64x2_t, <2 x i32> %arg1_int32x2_t, <2 x i32> %0) ; <<2 x i64>> [#uses=1]
- ret <2 x i64> %1
-}
-
-declare <4 x i32> @llvm.arm.neon.vqdmlal.v4i32(<4 x i32>, <4 x i16>, <4 x i16>) nounwind readnone
-declare <2 x i64> @llvm.arm.neon.vqdmlal.v2i64(<2 x i64>, <2 x i32>, <2 x i32>) nounwind readnone
-
define <4 x i32> @vqdmlals16_natural(<4 x i32>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind {
;CHECK-LABEL: vqdmlals16_natural:
;CHECK: vqdmlal.s16
declare <4 x i32> @llvm.arm.neon.vqadds.v4i32(<4 x i32>, <4 x i32>) nounwind readnone
declare <2 x i64> @llvm.arm.neon.vqadds.v2i64(<2 x i64>, <2 x i64>) nounwind readnone
-define <4 x i32> @vqdmlsls16(<4 x i32>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind {
-;CHECK-LABEL: vqdmlsls16:
-;CHECK: vqdmlsl.s16
- %tmp1 = load <4 x i32>* %A
- %tmp2 = load <4 x i16>* %B
- %tmp3 = load <4 x i16>* %C
- %tmp4 = call <4 x i32> @llvm.arm.neon.vqdmlsl.v4i32(<4 x i32> %tmp1, <4 x i16> %tmp2, <4 x i16> %tmp3)
- ret <4 x i32> %tmp4
-}
-
-define <2 x i64> @vqdmlsls32(<2 x i64>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind {
-;CHECK-LABEL: vqdmlsls32:
-;CHECK: vqdmlsl.s32
- %tmp1 = load <2 x i64>* %A
- %tmp2 = load <2 x i32>* %B
- %tmp3 = load <2 x i32>* %C
- %tmp4 = call <2 x i64> @llvm.arm.neon.vqdmlsl.v2i64(<2 x i64> %tmp1, <2 x i32> %tmp2, <2 x i32> %tmp3)
- ret <2 x i64> %tmp4
-}
-
-define arm_aapcs_vfpcc <4 x i32> @test_vqdmlsl_lanes16(<4 x i32> %arg0_int32x4_t, <4 x i16> %arg1_int16x4_t, <4 x i16> %arg2_int16x4_t) nounwind readnone {
-entry:
-; CHECK: test_vqdmlsl_lanes16
-; CHECK: vqdmlsl.s16 q0, d2, d3[1]
- %0 = shufflevector <4 x i16> %arg2_int16x4_t, <4 x i16> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1> ; <<4 x i16>> [#uses=1]
- %1 = tail call <4 x i32> @llvm.arm.neon.vqdmlsl.v4i32(<4 x i32> %arg0_int32x4_t, <4 x i16> %arg1_int16x4_t, <4 x i16> %0) ; <<4 x i32>> [#uses=1]
- ret <4 x i32> %1
-}
-
-define arm_aapcs_vfpcc <2 x i64> @test_vqdmlsl_lanes32(<2 x i64> %arg0_int64x2_t, <2 x i32> %arg1_int32x2_t, <2 x i32> %arg2_int32x2_t) nounwind readnone {
-entry:
-; CHECK: test_vqdmlsl_lanes32
-; CHECK: vqdmlsl.s32 q0, d2, d3[1]
- %0 = shufflevector <2 x i32> %arg2_int32x2_t, <2 x i32> undef, <2 x i32> <i32 1, i32 1> ; <<2 x i32>> [#uses=1]
- %1 = tail call <2 x i64> @llvm.arm.neon.vqdmlsl.v2i64(<2 x i64> %arg0_int64x2_t, <2 x i32> %arg1_int32x2_t, <2 x i32> %0) ; <<2 x i64>> [#uses=1]
- ret <2 x i64> %1
-}
-
-declare <4 x i32> @llvm.arm.neon.vqdmlsl.v4i32(<4 x i32>, <4 x i16>, <4 x i16>) nounwind readnone
-declare <2 x i64> @llvm.arm.neon.vqdmlsl.v2i64(<2 x i64>, <2 x i32>, <2 x i32>) nounwind readnone
-
define <4 x i32> @vqdmlsls16_natural(<4 x i32>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind {
;CHECK-LABEL: vqdmlsls16_natural:
;CHECK: vqdmlsl.s16