Lines Matching full:tmp1
6 %tmp1 = load <8 x i8>, <8 x i8>* %A
8 %tmp3 = call <8 x i8> @llvm.arm.neon.vpadd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
15 %tmp1 = load <4 x i16>, <4 x i16>* %A
17 %tmp3 = call <4 x i16> @llvm.arm.neon.vpadd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
24 %tmp1 = load <2 x i32>, <2 x i32>* %A
26 %tmp3 = call <2 x i32> @llvm.arm.neon.vpadd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
33 %tmp1 = load <2 x float>, <2 x float>* %A
35 %tmp3 = call <2 x float> @llvm.arm.neon.vpadd.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
48 %tmp1 = load <8 x i8>, <8 x i8>* %A
49 %tmp2 = call <4 x i16> @llvm.arm.neon.vpaddls.v4i16.v8i8(<8 x i8> %tmp1)
56 %tmp1 = load <4 x i16>, <4 x i16>* %A
57 %tmp2 = call <2 x i32> @llvm.arm.neon.vpaddls.v2i32.v4i16(<4 x i16> %tmp1)
64 %tmp1 = load <2 x i32>, <2 x i32>* %A
65 %tmp2 = call <1 x i64> @llvm.arm.neon.vpaddls.v1i64.v2i32(<2 x i32> %tmp1)
72 %tmp1 = load <8 x i8>, <8 x i8>* %A
73 %tmp2 = call <4 x i16> @llvm.arm.neon.vpaddlu.v4i16.v8i8(<8 x i8> %tmp1)
80 %tmp1 = load <4 x i16>, <4 x i16>* %A
81 %tmp2 = call <2 x i32> @llvm.arm.neon.vpaddlu.v2i32.v4i16(<4 x i16> %tmp1)
88 %tmp1 = load <2 x i32>, <2 x i32>* %A
89 %tmp2 = call <1 x i64> @llvm.arm.neon.vpaddlu.v1i64.v2i32(<2 x i32> %tmp1)
96 %tmp1 = load <16 x i8>, <16 x i8>* %A
97 %tmp2 = call <8 x i16> @llvm.arm.neon.vpaddls.v8i16.v16i8(<16 x i8> %tmp1)
104 %tmp1 = load <8 x i16>, <8 x i16>* %A
105 %tmp2 = call <4 x i32> @llvm.arm.neon.vpaddls.v4i32.v8i16(<8 x i16> %tmp1)
112 %tmp1 = load <4 x i32>, <4 x i32>* %A
113 %tmp2 = call <2 x i64> @llvm.arm.neon.vpaddls.v2i64.v4i32(<4 x i32> %tmp1)
120 %tmp1 = load <16 x i8>, <16 x i8>* %A
121 %tmp2 = call <8 x i16> @llvm.arm.neon.vpaddlu.v8i16.v16i8(<16 x i8> %tmp1)
128 %tmp1 = load <8 x i16>, <8 x i16>* %A
129 %tmp2 = call <4 x i32> @llvm.arm.neon.vpaddlu.v4i32.v8i16(<8 x i16> %tmp1)
136 %tmp1 = load <4 x i32>, <4 x i32>* %A
137 %tmp2 = call <2 x i64> @llvm.arm.neon.vpaddlu.v2i64.v4i32(<4 x i32> %tmp1)
147 %tmp1 = shufflevector <16 x i8> %tmp, <16 x i8> undef, <8 x i32> <i32 0, i32 2, i32 4, i32 6, i32 8, i32 10, i32 12, i32 14>
150 %add = add <8 x i8> %tmp3, %tmp1
160 %tmp1 = shufflevector <4 x i16> %in, <4 x i16> undef, <2 x i32> <i32 0, i32 2>
162 %x = add <2 x i16> %tmp2, %tmp1
