1 ; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
2 ; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=SSE
3 ; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=SSE
4 ; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX
5 ; RUN: llc < %s -mtriple=i686-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefix=AVX
6 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse2 | FileCheck %s --check-prefix=SSE
7 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+sse4.1 | FileCheck %s --check-prefix=SSE
8 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx | FileCheck %s --check-prefix=AVX
9 ; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mattr=+avx512f | FileCheck %s --check-prefix=AVX
11 ; PR21507 - https://llvm.org/bugs/show_bug.cgi?id=21507
12 ; Each function should be a single math op; no extra moves.
15 define <4 x float> @recip(<4 x float> %x) {
18 ; SSE-NEXT: rcpss %xmm0, %xmm0
19 ; SSE-NEXT: ret{{[l|q]}}
23 ; AVX-NEXT: vrcpss %xmm0, %xmm0, %xmm0
24 ; AVX-NEXT: ret{{[l|q]}}
25 %y = tail call <4 x float> @llvm.x86.sse.rcp.ss(<4 x float> %x)
26 %shuf = shufflevector <4 x float> %y, <4 x float> %x, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
30 define <4 x float> @recip_square_root(<4 x float> %x) {
31 ; SSE-LABEL: recip_square_root:
33 ; SSE-NEXT: rsqrtss %xmm0, %xmm0
34 ; SSE-NEXT: ret{{[l|q]}}
36 ; AVX-LABEL: recip_square_root:
38 ; AVX-NEXT: vrsqrtss %xmm0, %xmm0, %xmm0
39 ; AVX-NEXT: ret{{[l|q]}}
40 %y = tail call <4 x float> @llvm.x86.sse.rsqrt.ss(<4 x float> %x)
41 %shuf = shufflevector <4 x float> %y, <4 x float> %x, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
45 define <4 x float> @square_root(<4 x float> %x) {
46 ; SSE-LABEL: square_root:
48 ; SSE-NEXT: sqrtss %xmm0, %xmm0
49 ; SSE-NEXT: ret{{[l|q]}}
51 ; AVX-LABEL: square_root:
53 ; AVX-NEXT: vsqrtss %xmm0, %xmm0, %xmm0
54 ; AVX-NEXT: ret{{[l|q]}}
55 %y = tail call <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float> %x)
56 %shuf = shufflevector <4 x float> %y, <4 x float> %x, <4 x i32> <i32 0, i32 5, i32 6, i32 7>
60 define <2 x double> @square_root_double(<2 x double> %x) {
61 ; SSE-LABEL: square_root_double:
63 ; SSE-NEXT: sqrtsd %xmm0, %xmm0
64 ; SSE-NEXT: ret{{[l|q]}}
66 ; AVX-LABEL: square_root_double:
68 ; AVX-NEXT: vsqrtsd %xmm0, %xmm0, %xmm0
69 ; AVX-NEXT: ret{{[l|q]}}
70 %y = tail call <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double> %x)
71 %shuf = shufflevector <2 x double> %y, <2 x double> %x, <2 x i32> <i32 0, i32 3>
72 ret <2 x double> %shuf
75 declare <4 x float> @llvm.x86.sse.rcp.ss(<4 x float>)
76 declare <4 x float> @llvm.x86.sse.rsqrt.ss(<4 x float>)
77 declare <4 x float> @llvm.x86.sse.sqrt.ss(<4 x float>)
78 declare <2 x double> @llvm.x86.sse2.sqrt.sd(<2 x double>)