| /external/llvm/test/CodeGen/AArch64/ |
| D | arm64-cvt.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
10 %tmp3 = call i32 @llvm.aarch64.neon.fcvtas.i32.f32(float %A)
18 %tmp3 = call i64 @llvm.aarch64.neon.fcvtas.i64.f32(float %A)
26 %tmp3 = call i32 @llvm.aarch64.neon.fcvtas.i32.f64(double %A)
34 %tmp3 = call i64 @llvm.aarch64.neon.fcvtas.i64.f64(double %A)
38 declare i32 @llvm.aarch64.neon.fcvtas.i32.f32(float) nounwind readnone
39 declare i64 @llvm.aarch64.neon.fcvtas.i64.f32(float) nounwind readnone
40 declare i32 @llvm.aarch64.neon.fcvtas.i32.f64(double) nounwind readnone
41 declare i64 @llvm.aarch64.neon.fcvtas.i64.f64(double) nounwind readnone
50 %tmp3 = call i32 @llvm.aarch64.neon.fcvtau.i32.f32(float %A)
[all …]
|
| D | arm64-neon-across.ll | 1 ; RUN: llc < %s -verify-machineinstrs -mtriple=arm64-none-linux-gnu -mattr=+neon | FileCheck %s
3 declare float @llvm.aarch64.neon.fminnmv.f32.v4f32(<4 x float>)
5 declare float @llvm.aarch64.neon.fmaxnmv.f32.v4f32(<4 x float>)
7 declare float @llvm.aarch64.neon.fminv.f32.v4f32(<4 x float>)
9 declare float @llvm.aarch64.neon.fmaxv.f32.v4f32(<4 x float>)
11 declare i32 @llvm.aarch64.neon.saddv.i32.v4i32(<4 x i32>)
13 declare i32 @llvm.aarch64.neon.saddv.i32.v8i16(<8 x i16>)
15 declare i32 @llvm.aarch64.neon.saddv.i32.v16i8(<16 x i8>)
17 declare i32 @llvm.aarch64.neon.saddv.i32.v4i16(<4 x i16>)
19 declare i32 @llvm.aarch64.neon.saddv.i32.v8i8(<8 x i8>)
[all …]
|
| D | arm64-fminv.ll | 6 %min = call float @llvm.aarch64.neon.fminv.f32.v2f32(<2 x float> %in)
13 %min = call float @llvm.aarch64.neon.fminv.f32.v4f32(<4 x float> %in)
20 %min = call double @llvm.aarch64.neon.fminv.f64.v2f64(<2 x double> %in)
24 declare float @llvm.aarch64.neon.fminv.f32.v2f32(<2 x float>)
25 declare float @llvm.aarch64.neon.fminv.f32.v4f32(<4 x float>)
26 declare double @llvm.aarch64.neon.fminv.f64.v2f64(<2 x double>)
31 %max = call float @llvm.aarch64.neon.fmaxv.f32.v2f32(<2 x float> %in)
38 %max = call float @llvm.aarch64.neon.fmaxv.f32.v4f32(<4 x float> %in)
45 %max = call double @llvm.aarch64.neon.fmaxv.f64.v2f64(<2 x double> %in)
49 declare float @llvm.aarch64.neon.fmaxv.f32.v2f32(<2 x float>)
[all …]
|
| D | arm64-vsqrt.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
8 %tmp3 = call <2 x float> @llvm.aarch64.neon.frecps.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
17 %tmp3 = call <4 x float> @llvm.aarch64.neon.frecps.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
26 %tmp3 = call <2 x double> @llvm.aarch64.neon.frecps.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
30 declare <2 x float> @llvm.aarch64.neon.frecps.v2f32(<2 x float>, <2 x float>) nounwind readnone
31 declare <4 x float> @llvm.aarch64.neon.frecps.v4f32(<4 x float>, <4 x float>) nounwind readnone
32 declare <2 x double> @llvm.aarch64.neon.frecps.v2f64(<2 x double>, <2 x double>) nounwind readnone
40 %tmp3 = call <2 x float> @llvm.aarch64.neon.frsqrts.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
49 %tmp3 = call <4 x float> @llvm.aarch64.neon.frsqrts.v4f32(<4 x float> %tmp1, <4 x float> %tmp2)
58 %tmp3 = call <2 x double> @llvm.aarch64.neon.frsqrts.v2f64(<2 x double> %tmp1, <2 x double> %tmp2)
[all …]
|
| D | arm64-vqadd.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
8 %tmp3 = call <8 x i8> @llvm.aarch64.neon.sqadd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 %tmp3 = call <4 x i16> @llvm.aarch64.neon.sqadd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
26 %tmp3 = call <2 x i32> @llvm.aarch64.neon.sqadd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
35 %tmp3 = call <8 x i8> @llvm.aarch64.neon.uqadd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
44 %tmp3 = call <4 x i16> @llvm.aarch64.neon.uqadd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
53 %tmp3 = call <2 x i32> @llvm.aarch64.neon.uqadd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
62 %tmp3 = call <16 x i8> @llvm.aarch64.neon.sqadd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
71 %tmp3 = call <8 x i16> @llvm.aarch64.neon.sqadd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
80 %tmp3 = call <4 x i32> @llvm.aarch64.neon.sqadd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
[all …]
|
| D | arm64-vmax.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
8 %tmp3 = call <8 x i8> @llvm.aarch64.neon.smax.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 %tmp3 = call <16 x i8> @llvm.aarch64.neon.smax.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
26 %tmp3 = call <4 x i16> @llvm.aarch64.neon.smax.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
35 %tmp3 = call <8 x i16> @llvm.aarch64.neon.smax.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
44 %tmp3 = call <2 x i32> @llvm.aarch64.neon.smax.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
53 %tmp3 = call <4 x i32> @llvm.aarch64.neon.smax.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
57 declare <8 x i8> @llvm.aarch64.neon.smax.v8i8(<8 x i8>, <8 x i8>) nounwind readnone
58 declare <16 x i8> @llvm.aarch64.neon.smax.v16i8(<16 x i8>, <16 x i8>) nounwind readnone
59 declare <4 x i16> @llvm.aarch64.neon.smax.v4i16(<4 x i16>, <4 x i16>) nounwind readnone
[all …]
|
| D | arm64-neon-add-pairwise.ll | 1 ; RUN: llc -mtriple=arm64-none-linux-gnu -mattr=+neon < %s | FileCheck %s
3 declare <8 x i8> @llvm.aarch64.neon.addp.v8i8(<8 x i8>, <8 x i8>)
8 %tmp1 = call <8 x i8> @llvm.aarch64.neon.addp.v8i8(<8 x i8> %lhs, <8 x i8> %rhs)
13 declare <16 x i8> @llvm.aarch64.neon.addp.v16i8(<16 x i8>, <16 x i8>)
17 %tmp1 = call <16 x i8> @llvm.aarch64.neon.addp.v16i8(<16 x i8> %lhs, <16 x i8> %rhs)
22 declare <4 x i16> @llvm.aarch64.neon.addp.v4i16(<4 x i16>, <4 x i16>)
26 %tmp1 = call <4 x i16> @llvm.aarch64.neon.addp.v4i16(<4 x i16> %lhs, <4 x i16> %rhs)
31 declare <8 x i16> @llvm.aarch64.neon.addp.v8i16(<8 x i16>, <8 x i16>)
35 %tmp1 = call <8 x i16> @llvm.aarch64.neon.addp.v8i16(<8 x i16> %lhs, <8 x i16> %rhs)
40 declare <2 x i32> @llvm.aarch64.neon.addp.v2i32(<2 x i32>, <2 x i32>)
[all …]
|
| D | arm64-vhadd.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
8 %tmp3 = call <8 x i8> @llvm.aarch64.neon.shadd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 %tmp3 = call <16 x i8> @llvm.aarch64.neon.shadd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
26 %tmp3 = call <4 x i16> @llvm.aarch64.neon.shadd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
35 %tmp3 = call <8 x i16> @llvm.aarch64.neon.shadd.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
44 %tmp3 = call <2 x i32> @llvm.aarch64.neon.shadd.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
53 %tmp3 = call <4 x i32> @llvm.aarch64.neon.shadd.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
62 %tmp3 = call <8 x i8> @llvm.aarch64.neon.uhadd.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
71 %tmp3 = call <16 x i8> @llvm.aarch64.neon.uhadd.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
80 %tmp3 = call <4 x i16> @llvm.aarch64.neon.uhadd.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
[all …]
|
| D | arm64-neon-aba-abd.ll | 1 ; RUN: llc -mtriple=arm64-none-linux-gnu -mattr=+neon < %s | FileCheck %s
3 declare <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8>, <8 x i8>)
4 declare <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8>, <8 x i8>)
8 %abd = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %lhs, <8 x i8> %rhs)
15 %abd = call <8 x i8> @llvm.aarch64.neon.uabd.v8i8(<8 x i8> %lhs, <8 x i8> %rhs)
23 %abd = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %lhs, <8 x i8> %rhs)
30 %abd = call <8 x i8> @llvm.aarch64.neon.sabd.v8i8(<8 x i8> %lhs, <8 x i8> %rhs)
36 declare <16 x i8> @llvm.aarch64.neon.uabd.v16i8(<16 x i8>, <16 x i8>)
37 declare <16 x i8> @llvm.aarch64.neon.sabd.v16i8(<16 x i8>, <16 x i8>)
41 %abd = call <16 x i8> @llvm.aarch64.neon.uabd.v16i8(<16 x i8> %lhs, <16 x i8> %rhs)
[all …]
|
| D | arm64-vminmaxnm.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
6 …%vmaxnm2.i = tail call <2 x float> @llvm.aarch64.neon.fmaxnm.v2f32(<2 x float> %a, <2 x float> %b)…
13 …%vmaxnm2.i = tail call <4 x float> @llvm.aarch64.neon.fmaxnm.v4f32(<4 x float> %a, <4 x float> %b)…
20 …%vmaxnm2.i = tail call <2 x double> @llvm.aarch64.neon.fmaxnm.v2f64(<2 x double> %a, <2 x double> …
27 …%vminnm2.i = tail call <2 x float> @llvm.aarch64.neon.fminnm.v2f32(<2 x float> %a, <2 x float> %b)…
34 …%vminnm2.i = tail call <4 x float> @llvm.aarch64.neon.fminnm.v4f32(<4 x float> %a, <4 x float> %b)…
41 …%vminnm2.i = tail call <2 x double> @llvm.aarch64.neon.fminnm.v2f64(<2 x double> %a, <2 x double> …
48 %vmaxnm2.i = tail call float @llvm.aarch64.neon.fmaxnm.f32(float %a, float %b) nounwind
55 %vmaxnm2.i = tail call double @llvm.aarch64.neon.fminnm.f64(double %a, double %b) nounwind
59 declare <2 x double> @llvm.aarch64.neon.fminnm.v2f64(<2 x double>, <2 x double>) nounwind readnone
[all …]
|
| D | arm64-arith-saturating.ll | 8 %vqadd.i = tail call i32 @llvm.aarch64.neon.sqadd.i32(i32 %vecext, i32 %vecext1) nounwind
17 %vqadd.i = tail call i64 @llvm.aarch64.neon.sqadd.i64(i64 %vecext, i64 %vecext1) nounwind
26 %vqadd.i = tail call i32 @llvm.aarch64.neon.uqadd.i32(i32 %vecext, i32 %vecext1) nounwind
35 %vqadd.i = tail call i64 @llvm.aarch64.neon.uqadd.i64(i64 %vecext, i64 %vecext1) nounwind
39 declare i64 @llvm.aarch64.neon.uqadd.i64(i64, i64) nounwind readnone
40 declare i32 @llvm.aarch64.neon.uqadd.i32(i32, i32) nounwind readnone
41 declare i64 @llvm.aarch64.neon.sqadd.i64(i64, i64) nounwind readnone
42 declare i32 @llvm.aarch64.neon.sqadd.i32(i32, i32) nounwind readnone
49 %vqsub.i = tail call i32 @llvm.aarch64.neon.sqsub.i32(i32 %vecext, i32 %vecext1) nounwind
58 %vqsub.i = tail call i64 @llvm.aarch64.neon.sqsub.i64(i64 %vecext, i64 %vecext1) nounwind
[all …]
|
| D | arm64-st1.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple -verify-machineinstrs | FileCheck %s
278 call void @llvm.aarch64.neon.st2lane.v16i8.p0i8(<16 x i8> %A, <16 x i8> %B, i64 1, i8* %D)
285 call void @llvm.aarch64.neon.st2lane.v8i16.p0i16(<8 x i16> %A, <8 x i16> %B, i64 1, i16* %D)
292 call void @llvm.aarch64.neon.st2lane.v4i32.p0i32(<4 x i32> %A, <4 x i32> %B, i64 1, i32* %D)
299 call void @llvm.aarch64.neon.st2lane.v2i64.p0i64(<2 x i64> %A, <2 x i64> %B, i64 1, i64* %D)
303 declare void @llvm.aarch64.neon.st2lane.v16i8.p0i8(<16 x i8>, <16 x i8>, i64, i8*) nounwind readnone
304 declare void @llvm.aarch64.neon.st2lane.v8i16.p0i16(<8 x i16>, <8 x i16>, i64, i16*) nounwind readn…
305 declare void @llvm.aarch64.neon.st2lane.v4i32.p0i32(<4 x i32>, <4 x i32>, i64, i32*) nounwind readn…
306 declare void @llvm.aarch64.neon.st2lane.v2i64.p0i64(<2 x i64>, <2 x i64>, i64, i64*) nounwind readn…
311 …call void @llvm.aarch64.neon.st3lane.v16i8.p0i8(<16 x i8> %A, <16 x i8> %B, <16 x i8> %C, i64 1, i…
[all …]
|
| D | arm64-tbl.ll | 1 ; RUN: llc < %s -march=arm64 -aarch64-neon-syntax=apple | FileCheck %s
6 %tmp3 = call <8 x i8> @llvm.aarch64.neon.tbl1.v8i8(<16 x i8> %A, <8 x i8> %B)
13 %tmp3 = call <16 x i8> @llvm.aarch64.neon.tbl1.v16i8(<16 x i8> %A, <16 x i8> %B)
20 %tmp3 = call <8 x i8> @llvm.aarch64.neon.tbl2.v8i8(<16 x i8> %A, <16 x i8> %B, <8 x i8> %C)
27 %tmp3 = call <16 x i8> @llvm.aarch64.neon.tbl2.v16i8(<16 x i8> %A, <16 x i8> %B, <16 x i8> %C)
34 …%tmp3 = call <8 x i8> @llvm.aarch64.neon.tbl3.v8i8(<16 x i8> %A, <16 x i8> %B, <16 x i8> %C, <8 x …
41 …%tmp3 = call <16 x i8> @llvm.aarch64.neon.tbl3.v16i8(<16 x i8> %A, <16 x i8> %B, <16 x i8> %C, <16…
48 …%tmp3 = call <8 x i8> @llvm.aarch64.neon.tbl4.v8i8(<16 x i8> %A, <16 x i8> %B, <16 x i8> %C, <16 x…
55 …%tmp3 = call <16 x i8> @llvm.aarch64.neon.tbl4.v16i8(<16 x i8> %A, <16 x i8> %B, <16 x i8> %C, <16…
59 declare <8 x i8> @llvm.aarch64.neon.tbl1.v8i8(<16 x i8>, <8 x i8>) nounwind readnone
[all …]
|
| /external/llvm/test/CodeGen/ARM/ |
| D | intrinsics-crypto.ll | 6 %tmp3 = call <16 x i8> @llvm.arm.neon.aesd(<16 x i8> %tmp, <16 x i8> %tmp2)
8 %tmp4 = call <16 x i8> @llvm.arm.neon.aese(<16 x i8> %tmp3, <16 x i8> %tmp2)
10 %tmp5 = call <16 x i8> @llvm.arm.neon.aesimc(<16 x i8> %tmp4)
12 %tmp6 = call <16 x i8> @llvm.arm.neon.aesmc(<16 x i8> %tmp5)
22 %resscalar = call i32 @llvm.arm.neon.sha1h(i32 %scalar)
25 %res2 = call <4 x i32> @llvm.arm.neon.sha1c(<4 x i32> %tmp2, i32 %scalar, <4 x i32> %res1)
27 %res3 = call <4 x i32> @llvm.arm.neon.sha1m(<4 x i32> %res2, i32 %scalar, <4 x i32> %res1)
29 %res4 = call <4 x i32> @llvm.arm.neon.sha1p(<4 x i32> %res3, i32 %scalar, <4 x i32> %res1)
31 %res5 = call <4 x i32> @llvm.arm.neon.sha1su0(<4 x i32> %res4, <4 x i32> %tmp3, <4 x i32> %res1)
33 %res6 = call <4 x i32> @llvm.arm.neon.sha1su1(<4 x i32> %res5, <4 x i32> %res1)
[all …]
|
| D | vcvt-v8.ll | 1 ; RUN: llc < %s -mtriple=armv8 -mattr=+neon | FileCheck %s
6 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtas.v4i32.v4f32(<4 x float> %tmp1)
14 %tmp2 = call <2 x i32> @llvm.arm.neon.vcvtas.v2i32.v2f32(<2 x float> %tmp1)
22 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtns.v4i32.v4f32(<4 x float> %tmp1)
30 %tmp2 = call <2 x i32> @llvm.arm.neon.vcvtns.v2i32.v2f32(<2 x float> %tmp1)
38 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtps.v4i32.v4f32(<4 x float> %tmp1)
46 %tmp2 = call <2 x i32> @llvm.arm.neon.vcvtps.v2i32.v2f32(<2 x float> %tmp1)
54 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtms.v4i32.v4f32(<4 x float> %tmp1)
62 %tmp2 = call <2 x i32> @llvm.arm.neon.vcvtms.v2i32.v2f32(<2 x float> %tmp1)
70 %tmp2 = call <4 x i32> @llvm.arm.neon.vcvtau.v4i32.v4f32(<4 x float> %tmp1)
[all …]
|
| D | neon-v8.1a.ll | 6 declare <4 x i16> @llvm.arm.neon.vqrdmulh.v4i16(<4 x i16>, <4 x i16>)
7 declare <8 x i16> @llvm.arm.neon.vqrdmulh.v8i16(<8 x i16>, <8 x i16>)
8 declare <2 x i32> @llvm.arm.neon.vqrdmulh.v2i32(<2 x i32>, <2 x i32>)
9 declare <4 x i32> @llvm.arm.neon.vqrdmulh.v4i32(<4 x i32>, <4 x i32>)
11 declare <4 x i16> @llvm.arm.neon.vqadds.v4i16(<4 x i16>, <4 x i16>)
12 declare <8 x i16> @llvm.arm.neon.vqadds.v8i16(<8 x i16>, <8 x i16>)
13 declare <2 x i32> @llvm.arm.neon.vqadds.v2i32(<2 x i32>, <2 x i32>)
14 declare <4 x i32> @llvm.arm.neon.vqadds.v4i32(<4 x i32>, <4 x i32>)
16 declare <4 x i16> @llvm.arm.neon.vqsubs.v4i16(<4 x i16>, <4 x i16>)
17 declare <8 x i16> @llvm.arm.neon.vqsubs.v8i16(<8 x i16>, <8 x i16>)
[all …]
|
| D | vminmax.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
8 %tmp3 = call <8 x i8> @llvm.arm.neon.vmins.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 %tmp3 = call <4 x i16> @llvm.arm.neon.vmins.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
26 %tmp3 = call <2 x i32> @llvm.arm.neon.vmins.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
35 %tmp3 = call <8 x i8> @llvm.arm.neon.vminu.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
44 %tmp3 = call <4 x i16> @llvm.arm.neon.vminu.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
53 %tmp3 = call <2 x i32> @llvm.arm.neon.vminu.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
62 %tmp3 = call <2 x float> @llvm.arm.neon.vmins.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
71 %tmp3 = call <16 x i8> @llvm.arm.neon.vmins.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
80 %tmp3 = call <8 x i16> @llvm.arm.neon.vmins.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
[all …]
|
| D | vrec.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
7 %tmp2 = call <2 x i32> @llvm.arm.neon.vrecpe.v2i32(<2 x i32> %tmp1)
15 %tmp2 = call <4 x i32> @llvm.arm.neon.vrecpe.v4i32(<4 x i32> %tmp1)
23 %tmp2 = call <2 x float> @llvm.arm.neon.vrecpe.v2f32(<2 x float> %tmp1)
31 %tmp2 = call <4 x float> @llvm.arm.neon.vrecpe.v4f32(<4 x float> %tmp1)
35 declare <2 x i32> @llvm.arm.neon.vrecpe.v2i32(<2 x i32>) nounwind readnone
36 declare <4 x i32> @llvm.arm.neon.vrecpe.v4i32(<4 x i32>) nounwind readnone
38 declare <2 x float> @llvm.arm.neon.vrecpe.v2f32(<2 x float>) nounwind readnone
39 declare <4 x float> @llvm.arm.neon.vrecpe.v4f32(<4 x float>) nounwind readnone
46 %tmp3 = call <2 x float> @llvm.arm.neon.vrecps.v2f32(<2 x float> %tmp1, <2 x float> %tmp2)
[all …]
|
| D | vhadd.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
8 %tmp3 = call <8 x i8> @llvm.arm.neon.vhadds.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 %tmp3 = call <4 x i16> @llvm.arm.neon.vhadds.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
26 %tmp3 = call <2 x i32> @llvm.arm.neon.vhadds.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
35 %tmp3 = call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
44 %tmp3 = call <4 x i16> @llvm.arm.neon.vhaddu.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
53 %tmp3 = call <2 x i32> @llvm.arm.neon.vhaddu.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
62 %tmp3 = call <16 x i8> @llvm.arm.neon.vhadds.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
71 %tmp3 = call <8 x i16> @llvm.arm.neon.vhadds.v8i16(<8 x i16> %tmp1, <8 x i16> %tmp2)
80 %tmp3 = call <4 x i32> @llvm.arm.neon.vhadds.v4i32(<4 x i32> %tmp1, <4 x i32> %tmp2)
[all …]
|
| D | vqshl.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
8 %tmp3 = call <8 x i8> @llvm.arm.neon.vqshifts.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 %tmp3 = call <4 x i16> @llvm.arm.neon.vqshifts.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
26 %tmp3 = call <2 x i32> @llvm.arm.neon.vqshifts.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
35 %tmp3 = call <1 x i64> @llvm.arm.neon.vqshifts.v1i64(<1 x i64> %tmp1, <1 x i64> %tmp2)
44 %tmp3 = call <8 x i8> @llvm.arm.neon.vqshiftu.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
53 %tmp3 = call <4 x i16> @llvm.arm.neon.vqshiftu.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
62 %tmp3 = call <2 x i32> @llvm.arm.neon.vqshiftu.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
71 %tmp3 = call <1 x i64> @llvm.arm.neon.vqshiftu.v1i64(<1 x i64> %tmp1, <1 x i64> %tmp2)
80 %tmp3 = call <16 x i8> @llvm.arm.neon.vqshifts.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
[all …]
|
| D | vqshrn.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
7 …%tmp2 = call <8 x i8> @llvm.arm.neon.vqshiftns.v8i8(<8 x i16> %tmp1, <8 x i16> < i16 -8, i16 -8, i…
15 …%tmp2 = call <4 x i16> @llvm.arm.neon.vqshiftns.v4i16(<4 x i32> %tmp1, <4 x i32> < i32 -16, i32 -1…
23 …%tmp2 = call <2 x i32> @llvm.arm.neon.vqshiftns.v2i32(<2 x i64> %tmp1, <2 x i64> < i64 -32, i64 -3…
31 …%tmp2 = call <8 x i8> @llvm.arm.neon.vqshiftnu.v8i8(<8 x i16> %tmp1, <8 x i16> < i16 -8, i16 -8, i…
39 …%tmp2 = call <4 x i16> @llvm.arm.neon.vqshiftnu.v4i16(<4 x i32> %tmp1, <4 x i32> < i32 -16, i32 -1…
47 …%tmp2 = call <2 x i32> @llvm.arm.neon.vqshiftnu.v2i32(<2 x i64> %tmp1, <2 x i64> < i64 -32, i64 -3…
55 …%tmp2 = call <8 x i8> @llvm.arm.neon.vqshiftnsu.v8i8(<8 x i16> %tmp1, <8 x i16> < i16 -8, i16 -8, …
63 …%tmp2 = call <4 x i16> @llvm.arm.neon.vqshiftnsu.v4i16(<4 x i32> %tmp1, <4 x i32> < i32 -16, i32 -…
71 …%tmp2 = call <2 x i32> @llvm.arm.neon.vqshiftnsu.v2i32(<2 x i64> %tmp1, <2 x i64> < i64 -32, i64 -…
[all …]
|
| D | vld-vst-upgrade.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon < %s | FileCheck %s
12 %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32(i8* %ptr, i32 1)
16 declare <2 x i32> @llvm.arm.neon.vld1.v2i32(i8*, i32) nounwind readonly
21 %tmp1 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32(i8* %ptr, i32 1)
25 declare %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2.v2i32(i8*, i32) nounwind readonly
30 %tmp1 = call %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32(i8* %ptr, i32 1)
34 declare %struct.__neon_int32x2x3_t @llvm.arm.neon.vld3.v2i32(i8*, i32) nounwind readonly
39 %tmp1 = call %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8* %ptr, i32 1)
43 declare %struct.__neon_int32x2x4_t @llvm.arm.neon.vld4.v2i32(i8*, i32) nounwind readonly
50 …%tmp1 = call %struct.__neon_int32x2x2_t @llvm.arm.neon.vld2lane.v2i32(i8* %ptr, <2 x i32> %A, <2 x…
[all …]
|
| D | vabs.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
7 %tmp2 = call <8 x i8> @llvm.arm.neon.vabs.v8i8(<8 x i8> %tmp1)
15 %tmp2 = call <4 x i16> @llvm.arm.neon.vabs.v4i16(<4 x i16> %tmp1)
23 %tmp2 = call <2 x i32> @llvm.arm.neon.vabs.v2i32(<2 x i32> %tmp1)
39 %tmp2 = call <16 x i8> @llvm.arm.neon.vabs.v16i8(<16 x i8> %tmp1)
47 %tmp2 = call <8 x i16> @llvm.arm.neon.vabs.v8i16(<8 x i16> %tmp1)
55 %tmp2 = call <4 x i32> @llvm.arm.neon.vabs.v4i32(<4 x i32> %tmp1)
67 declare <8 x i8> @llvm.arm.neon.vabs.v8i8(<8 x i8>) nounwind readnone
68 declare <4 x i16> @llvm.arm.neon.vabs.v4i16(<4 x i16>) nounwind readnone
69 declare <2 x i32> @llvm.arm.neon.vabs.v2i32(<2 x i32>) nounwind readnone
[all …]
|
| D | vld1.ll | 1 ; RUN: llc -mtriple=arm-eabi -float-abi=soft -mattr=+neon %s -o - | FileCheck %s
3 ; RUN: llc -mtriple=arm-eabi -float-abi=soft -mattr=+neon -regalloc=basic %s -o - \
10 %tmp1 = call <8 x i8> @llvm.arm.neon.vld1.v8i8.p0i8(i8* %A, i32 16)
18 %tmp1 = call <4 x i16> @llvm.arm.neon.vld1.v4i16.p0i8(i8* %tmp0, i32 1)
28 %tmp1 = call <4 x i16> @llvm.arm.neon.vld1.v4i16.p0i8(i8* %tmp0, i32 1)
38 %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32.p0i8(i8* %tmp0, i32 1)
48 %tmp1 = call <2 x i32> @llvm.arm.neon.vld1.v2i32.p0i8(i8* %tmp0, i32 1)
58 %tmp1 = call <2 x float> @llvm.arm.neon.vld1.v2f32.p0i8(i8* %tmp0, i32 1)
66 %tmp1 = call <1 x i64> @llvm.arm.neon.vld1.v1i64.p0i8(i8* %tmp0, i32 1)
74 %tmp1 = call <16 x i8> @llvm.arm.neon.vld1.v16i8.p0i8(i8* %A, i32 8)
[all …]
|
| D | vqsub.ll | 1 ; RUN: llc -mtriple=arm-eabi -mattr=+neon %s -o - | FileCheck %s
8 %tmp3 = call <8 x i8> @llvm.arm.neon.vqsubs.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
17 %tmp3 = call <4 x i16> @llvm.arm.neon.vqsubs.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
26 %tmp3 = call <2 x i32> @llvm.arm.neon.vqsubs.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
35 %tmp3 = call <1 x i64> @llvm.arm.neon.vqsubs.v1i64(<1 x i64> %tmp1, <1 x i64> %tmp2)
44 %tmp3 = call <8 x i8> @llvm.arm.neon.vqsubu.v8i8(<8 x i8> %tmp1, <8 x i8> %tmp2)
53 %tmp3 = call <4 x i16> @llvm.arm.neon.vqsubu.v4i16(<4 x i16> %tmp1, <4 x i16> %tmp2)
62 %tmp3 = call <2 x i32> @llvm.arm.neon.vqsubu.v2i32(<2 x i32> %tmp1, <2 x i32> %tmp2)
71 %tmp3 = call <1 x i64> @llvm.arm.neon.vqsubu.v1i64(<1 x i64> %tmp1, <1 x i64> %tmp2)
80 %tmp3 = call <16 x i8> @llvm.arm.neon.vqsubs.v16i8(<16 x i8> %tmp1, <16 x i8> %tmp2)
[all …]
|