1 ; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t
2 ; RUN: grep {vmull\\.s8} %t | count 1
3 ; RUN: grep {vmull\\.s16} %t | count 1
4 ; RUN: grep {vmull\\.s32} %t | count 1
5 ; RUN: grep {vmull\\.u8} %t | count 1
6 ; RUN: grep {vmull\\.u16} %t | count 1
7 ; RUN: grep {vmull\\.u32} %t | count 1
8 ; RUN: grep {vmull\\.p8} %t | count 1
10 define <8 x i16> @vmulls8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
11 %tmp1 = load <8 x i8>* %A
12 %tmp2 = load <8 x i8>* %B
13 %tmp3 = call <8 x i16> @llvm.arm.neon.vmulls.v8i16(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 define <4 x i32> @vmulls16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
18 %tmp1 = load <4 x i16>* %A
19 %tmp2 = load <4 x i16>* %B
20 %tmp3 = call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> %tmp1, <4 x i16> %tmp2)
24 define <2 x i64> @vmulls32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
25 %tmp1 = load <2 x i32>* %A
26 %tmp2 = load <2 x i32>* %B
27 %tmp3 = call <2 x i64> @llvm.arm.neon.vmulls.v2i64(<2 x i32> %tmp1, <2 x i32> %tmp2)
31 define <8 x i16> @vmullu8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
32 %tmp1 = load <8 x i8>* %A
33 %tmp2 = load <8 x i8>* %B
34 %tmp3 = call <8 x i16> @llvm.arm.neon.vmullu.v8i16(<8 x i8> %tmp1, <8 x i8> %tmp2)
38 define <4 x i32> @vmullu16(<4 x i16>* %A, <4 x i16>* %B) nounwind {
39 %tmp1 = load <4 x i16>* %A
40 %tmp2 = load <4 x i16>* %B
41 %tmp3 = call <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16> %tmp1, <4 x i16> %tmp2)
45 define <2 x i64> @vmullu32(<2 x i32>* %A, <2 x i32>* %B) nounwind {
46 %tmp1 = load <2 x i32>* %A
47 %tmp2 = load <2 x i32>* %B
48 %tmp3 = call <2 x i64> @llvm.arm.neon.vmullu.v2i64(<2 x i32> %tmp1, <2 x i32> %tmp2)
52 define <8 x i16> @vmullp8(<8 x i8>* %A, <8 x i8>* %B) nounwind {
53 %tmp1 = load <8 x i8>* %A
54 %tmp2 = load <8 x i8>* %B
55 %tmp3 = call <8 x i16> @llvm.arm.neon.vmullp.v8i16(<8 x i8> %tmp1, <8 x i8> %tmp2)
59 declare <8 x i16> @llvm.arm.neon.vmulls.v8i16(<8 x i8>, <8 x i8>) nounwind readnone
60 declare <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16>, <4 x i16>) nounwind readnone
61 declare <2 x i64> @llvm.arm.neon.vmulls.v2i64(<2 x i32>, <2 x i32>) nounwind readnone
63 declare <8 x i16> @llvm.arm.neon.vmullu.v8i16(<8 x i8>, <8 x i8>) nounwind readnone
64 declare <4 x i32> @llvm.arm.neon.vmullu.v4i32(<4 x i16>, <4 x i16>) nounwind readnone
65 declare <2 x i64> @llvm.arm.neon.vmullu.v2i64(<2 x i32>, <2 x i32>) nounwind readnone
67 declare <8 x i16> @llvm.arm.neon.vmullp.v8i16(<8 x i8>, <8 x i8>) nounwind readnone