| Bob Wilson | e60fee0 | 2009-06-22 23:27:02 +0000 | [diff] [blame^] | 1 | ; RUN: llvm-as < %s | llc -march=arm -mattr=+neon > %t |
| 2 | ; RUN: grep {vmlal\\.s8} %t | count 1 |
| 3 | ; RUN: grep {vmlal\\.s16} %t | count 1 |
| 4 | ; RUN: grep {vmlal\\.s32} %t | count 1 |
| 5 | ; RUN: grep {vmlal\\.u8} %t | count 1 |
| 6 | ; RUN: grep {vmlal\\.u16} %t | count 1 |
| 7 | ; RUN: grep {vmlal\\.u32} %t | count 1 |
| 8 | |
| 9 | define <8 x i16> @vmlals8(<8 x i16>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind { |
| 10 | %tmp1 = load <8 x i16>* %A |
| 11 | %tmp2 = load <8 x i8>* %B |
| 12 | %tmp3 = load <8 x i8>* %C |
| 13 | %tmp4 = call <8 x i16> @llvm.arm.neon.vmlals.v8i16(<8 x i16> %tmp1, <8 x i8> %tmp2, <8 x i8> %tmp3) |
| 14 | ret <8 x i16> %tmp4 |
| 15 | } |
| 16 | |
| 17 | define <4 x i32> @vmlals16(<4 x i32>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind { |
| 18 | %tmp1 = load <4 x i32>* %A |
| 19 | %tmp2 = load <4 x i16>* %B |
| 20 | %tmp3 = load <4 x i16>* %C |
| 21 | %tmp4 = call <4 x i32> @llvm.arm.neon.vmlals.v4i32(<4 x i32> %tmp1, <4 x i16> %tmp2, <4 x i16> %tmp3) |
| 22 | ret <4 x i32> %tmp4 |
| 23 | } |
| 24 | |
| 25 | define <2 x i64> @vmlals32(<2 x i64>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind { |
| 26 | %tmp1 = load <2 x i64>* %A |
| 27 | %tmp2 = load <2 x i32>* %B |
| 28 | %tmp3 = load <2 x i32>* %C |
| 29 | %tmp4 = call <2 x i64> @llvm.arm.neon.vmlals.v2i64(<2 x i64> %tmp1, <2 x i32> %tmp2, <2 x i32> %tmp3) |
| 30 | ret <2 x i64> %tmp4 |
| 31 | } |
| 32 | |
| 33 | define <8 x i16> @vmlalu8(<8 x i16>* %A, <8 x i8>* %B, <8 x i8>* %C) nounwind { |
| 34 | %tmp1 = load <8 x i16>* %A |
| 35 | %tmp2 = load <8 x i8>* %B |
| 36 | %tmp3 = load <8 x i8>* %C |
| 37 | %tmp4 = call <8 x i16> @llvm.arm.neon.vmlalu.v8i16(<8 x i16> %tmp1, <8 x i8> %tmp2, <8 x i8> %tmp3) |
| 38 | ret <8 x i16> %tmp4 |
| 39 | } |
| 40 | |
| 41 | define <4 x i32> @vmlalu16(<4 x i32>* %A, <4 x i16>* %B, <4 x i16>* %C) nounwind { |
| 42 | %tmp1 = load <4 x i32>* %A |
| 43 | %tmp2 = load <4 x i16>* %B |
| 44 | %tmp3 = load <4 x i16>* %C |
| 45 | %tmp4 = call <4 x i32> @llvm.arm.neon.vmlalu.v4i32(<4 x i32> %tmp1, <4 x i16> %tmp2, <4 x i16> %tmp3) |
| 46 | ret <4 x i32> %tmp4 |
| 47 | } |
| 48 | |
| 49 | define <2 x i64> @vmlalu32(<2 x i64>* %A, <2 x i32>* %B, <2 x i32>* %C) nounwind { |
| 50 | %tmp1 = load <2 x i64>* %A |
| 51 | %tmp2 = load <2 x i32>* %B |
| 52 | %tmp3 = load <2 x i32>* %C |
| 53 | %tmp4 = call <2 x i64> @llvm.arm.neon.vmlalu.v2i64(<2 x i64> %tmp1, <2 x i32> %tmp2, <2 x i32> %tmp3) |
| 54 | ret <2 x i64> %tmp4 |
| 55 | } |
| 56 | |
| 57 | declare <8 x i16> @llvm.arm.neon.vmlals.v8i16(<8 x i16>, <8 x i8>, <8 x i8>) nounwind readnone |
| 58 | declare <4 x i32> @llvm.arm.neon.vmlals.v4i32(<4 x i32>, <4 x i16>, <4 x i16>) nounwind readnone |
| 59 | declare <2 x i64> @llvm.arm.neon.vmlals.v2i64(<2 x i64>, <2 x i32>, <2 x i32>) nounwind readnone |
| 60 | |
| 61 | declare <8 x i16> @llvm.arm.neon.vmlalu.v8i16(<8 x i16>, <8 x i8>, <8 x i8>) nounwind readnone |
| 62 | declare <4 x i32> @llvm.arm.neon.vmlalu.v4i32(<4 x i32>, <4 x i16>, <4 x i16>) nounwind readnone |
| 63 | declare <2 x i64> @llvm.arm.neon.vmlalu.v2i64(<2 x i64>, <2 x i32>, <2 x i32>) nounwind readnone |