1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=i386-- -mattr=sse2 | FileCheck %s --check-prefixes=X86,X86-SSE
; RUN: llc < %s -mtriple=i386-- -mattr=sse4.1 | FileCheck %s --check-prefixes=X86,X86-SSE
; RUN: llc < %s -mtriple=i386-- -mattr=avx2 | FileCheck %s --check-prefixes=X86,X86-AVX2
; RUN: llc < %s -mtriple=i386-- -mattr=avx512f | FileCheck %s --check-prefixes=X86,X86-AVX512F
; RUN: llc < %s -mtriple=x86_64-- -mattr=sse2 | FileCheck %s --check-prefix=X64-SSE
; RUN: llc < %s -mtriple=x86_64-- -mattr=sse4.1 | FileCheck %s --check-prefix=X64-SSE
; RUN: llc < %s -mtriple=x86_64-- -mattr=avx2 | FileCheck %s --check-prefixes=X64-AVX,X64-AVX2
; RUN: llc < %s -mtriple=x86_64-- -mattr=avx512f | FileCheck %s --check-prefixes=X64-AVX,X64-AVX512F
; This should do a single load into the fp stack for the return, not diddle with xmm registers.
define float @icmp_select_fp_constants(i32 %x) nounwind readnone {
; X86-LABEL: icmp_select_fp_constants:
; X86: # %bb.0:
; X86-NEXT: xorl %eax, %eax
; X86-NEXT: cmpl $0, {{[0-9]+}}(%esp)
; X86-NEXT: sete %al
; X86-NEXT: flds {{\.LCPI[0-9]+_[0-9]+}}(,%eax,4)
; X86-NEXT: retl
;
; X64-SSE-LABEL: icmp_select_fp_constants:
; X64-SSE: # %bb.0:
; X64-SSE-NEXT: xorl %eax, %eax
; X64-SSE-NEXT: testl %edi, %edi
; X64-SSE-NEXT: sete %al
; X64-SSE-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X64-SSE-NEXT: retq
;
; X64-AVX-LABEL: icmp_select_fp_constants:
; X64-AVX: # %bb.0:
; X64-AVX-NEXT: xorl %eax, %eax
; X64-AVX-NEXT: testl %edi, %edi
; X64-AVX-NEXT: sete %al
; X64-AVX-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X64-AVX-NEXT: retq
%c = icmp eq i32 %x, 0
%r = select i1 %c, float 42.0, float 23.0
ret float %r
}
define float @fcmp_select_fp_constants(float %x) nounwind readnone {
; X86-SSE-LABEL: fcmp_select_fp_constants:
; X86-SSE: # %bb.0:
; X86-SSE-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-SSE-NEXT: cmpneqss {{[0-9]+}}(%esp), %xmm0
; X86-SSE-NEXT: movd %xmm0, %eax
; X86-SSE-NEXT: andl $1, %eax
; X86-SSE-NEXT: flds {{\.LCPI[0-9]+_[0-9]+}}(,%eax,4)
; X86-SSE-NEXT: retl
;
; X86-AVX2-LABEL: fcmp_select_fp_constants:
; X86-AVX2: # %bb.0:
; X86-AVX2-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-AVX2-NEXT: vcmpneqss {{[0-9]+}}(%esp), %xmm0, %xmm0
; X86-AVX2-NEXT: vmovd %xmm0, %eax
; X86-AVX2-NEXT: andl $1, %eax
; X86-AVX2-NEXT: flds {{\.LCPI[0-9]+_[0-9]+}}(,%eax,4)
; X86-AVX2-NEXT: retl
;
; X86-AVX512F-LABEL: fcmp_select_fp_constants:
; X86-AVX512F: # %bb.0:
; X86-AVX512F-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X86-AVX512F-NEXT: vcmpneqss {{\.LCPI[0-9]+_[0-9]+}}, %xmm0, %k0
; X86-AVX512F-NEXT: kmovw %k0, %eax
; X86-AVX512F-NEXT: flds {{\.LCPI[0-9]+_[0-9]+}}(,%eax,4)
; X86-AVX512F-NEXT: retl
;
; X64-SSE-LABEL: fcmp_select_fp_constants:
; X64-SSE: # %bb.0:
; X64-SSE-NEXT: cmpneqss {{.*}}(%rip), %xmm0
; X64-SSE-NEXT: movd %xmm0, %eax
; X64-SSE-NEXT: andl $1, %eax
; X64-SSE-NEXT: movss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X64-SSE-NEXT: retq
;
; X64-AVX2-LABEL: fcmp_select_fp_constants:
; X64-AVX2: # %bb.0:
; X64-AVX2-NEXT: vcmpneqss {{.*}}(%rip), %xmm0, %xmm0
; X64-AVX2-NEXT: vmovss {{.*#+}} xmm1 = mem[0],zero,zero,zero
; X64-AVX2-NEXT: vmovss {{.*#+}} xmm2 = mem[0],zero,zero,zero
; X64-AVX2-NEXT: vblendvps %xmm0, %xmm1, %xmm2, %xmm0
; X64-AVX2-NEXT: retq
;
; X64-AVX512F-LABEL: fcmp_select_fp_constants:
; X64-AVX512F: # %bb.0:
; X64-AVX512F-NEXT: vcmpneqss {{.*}}(%rip), %xmm0, %k1
; X64-AVX512F-NEXT: vmovss {{.*#+}} xmm0 = mem[0],zero,zero,zero
; X64-AVX512F-NEXT: vmovss {{.*}}(%rip), %xmm0 {%k1}
; X64-AVX512F-NEXT: retq
%c = fcmp une float %x, -4.0
%r = select i1 %c, float 42.0, float 23.0
ret float %r
}
|
GitWeb