fp128-select.ll
4.7 KB
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
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -O2 -mtriple=x86_64-linux-android -mattr=+sse \
; RUN: -enable-legalize-types-checking | FileCheck %s --check-prefix=SSE
; RUN: llc < %s -O2 -mtriple=x86_64-linux-gnu -mattr=+sse \
; RUN: -enable-legalize-types-checking | FileCheck %s --check-prefix=SSE
; RUN: llc < %s -O2 -mtriple=x86_64-linux-android -mattr=-sse \
; RUN: -enable-legalize-types-checking | FileCheck %s --check-prefix=NOSSE
; RUN: llc < %s -O2 -mtriple=x86_64-linux-gnu -mattr=-sse \
; RUN: -enable-legalize-types-checking | FileCheck %s --check-prefix=NOSSE
define void @test_select(fp128* %p, fp128* %q, i1 zeroext %c) {
; SSE-LABEL: test_select:
; SSE: # %bb.0:
; SSE-NEXT: testl %edx, %edx
; SSE-NEXT: jne .LBB0_1
; SSE-NEXT: # %bb.2:
; SSE-NEXT: movaps {{.*}}(%rip), %xmm0
; SSE-NEXT: movaps %xmm0, (%rsi)
; SSE-NEXT: retq
; SSE-NEXT: .LBB0_1:
; SSE-NEXT: movups (%rdi), %xmm0
; SSE-NEXT: movaps %xmm0, (%rsi)
; SSE-NEXT: retq
;
; NOSSE-LABEL: test_select:
; NOSSE: # %bb.0:
; NOSSE-NEXT: xorl %eax, %eax
; NOSSE-NEXT: testl %edx, %edx
; NOSSE-NEXT: cmovneq (%rdi), %rax
; NOSSE-NEXT: movabsq $9223231299366420480, %rcx # imm = 0x7FFF800000000000
; NOSSE-NEXT: cmovneq 8(%rdi), %rcx
; NOSSE-NEXT: movq %rcx, 8(%rsi)
; NOSSE-NEXT: movq %rax, (%rsi)
; NOSSE-NEXT: retq
%a = load fp128, fp128* %p, align 2
%r = select i1 %c, fp128 %a, fp128 0xL00000000000000007FFF800000000000
store fp128 %r, fp128* %q
ret void
}
; The uitofp will become a select_cc. This used to crash during type
; legalization because we didn't expect the operands to need to be softened.
define fp128 @test_select_cc(fp128, fp128) {
; SSE-LABEL: test_select_cc:
; SSE: # %bb.0: # %BB0
; SSE-NEXT: pushq %rbx
; SSE-NEXT: .cfi_def_cfa_offset 16
; SSE-NEXT: subq $32, %rsp
; SSE-NEXT: .cfi_def_cfa_offset 48
; SSE-NEXT: .cfi_offset %rbx, -16
; SSE-NEXT: movaps %xmm1, {{[-0-9]+}}(%r{{[sb]}}p) # 16-byte Spill
; SSE-NEXT: movaps %xmm0, (%rsp) # 16-byte Spill
; SSE-NEXT: callq __netf2
; SSE-NEXT: movl %eax, %ebx
; SSE-NEXT: movaps (%rsp), %xmm0 # 16-byte Reload
; SSE-NEXT: movaps {{[-0-9]+}}(%r{{[sb]}}p), %xmm1 # 16-byte Reload
; SSE-NEXT: callq __eqtf2
; SSE-NEXT: testl %eax, %eax
; SSE-NEXT: je .LBB1_1
; SSE-NEXT: # %bb.2: # %BB0
; SSE-NEXT: xorps %xmm1, %xmm1
; SSE-NEXT: jmp .LBB1_3
; SSE-NEXT: .LBB1_1:
; SSE-NEXT: movaps {{.*}}(%rip), %xmm1
; SSE-NEXT: .LBB1_3: # %BB0
; SSE-NEXT: testl %ebx, %ebx
; SSE-NEXT: movaps (%rsp), %xmm0 # 16-byte Reload
; SSE-NEXT: jne .LBB1_5
; SSE-NEXT: # %bb.4: # %BB1
; SSE-NEXT: movaps %xmm1, %xmm0
; SSE-NEXT: .LBB1_5: # %BB2
; SSE-NEXT: addq $32, %rsp
; SSE-NEXT: .cfi_def_cfa_offset 16
; SSE-NEXT: popq %rbx
; SSE-NEXT: .cfi_def_cfa_offset 8
; SSE-NEXT: retq
;
; NOSSE-LABEL: test_select_cc:
; NOSSE: # %bb.0: # %BB0
; NOSSE-NEXT: pushq %rbp
; NOSSE-NEXT: .cfi_def_cfa_offset 16
; NOSSE-NEXT: pushq %r15
; NOSSE-NEXT: .cfi_def_cfa_offset 24
; NOSSE-NEXT: pushq %r14
; NOSSE-NEXT: .cfi_def_cfa_offset 32
; NOSSE-NEXT: pushq %r12
; NOSSE-NEXT: .cfi_def_cfa_offset 40
; NOSSE-NEXT: pushq %rbx
; NOSSE-NEXT: .cfi_def_cfa_offset 48
; NOSSE-NEXT: .cfi_offset %rbx, -48
; NOSSE-NEXT: .cfi_offset %r12, -40
; NOSSE-NEXT: .cfi_offset %r14, -32
; NOSSE-NEXT: .cfi_offset %r15, -24
; NOSSE-NEXT: .cfi_offset %rbp, -16
; NOSSE-NEXT: movq %rcx, %r12
; NOSSE-NEXT: movq %rdx, %rbx
; NOSSE-NEXT: movq %rsi, %r14
; NOSSE-NEXT: movq %rdi, %r15
; NOSSE-NEXT: callq __netf2
; NOSSE-NEXT: movl %eax, %ebp
; NOSSE-NEXT: movq %r15, %rdi
; NOSSE-NEXT: movq %r14, %rsi
; NOSSE-NEXT: movq %rbx, %rdx
; NOSSE-NEXT: movq %r12, %rcx
; NOSSE-NEXT: callq __eqtf2
; NOSSE-NEXT: movl %eax, %ecx
; NOSSE-NEXT: xorl %eax, %eax
; NOSSE-NEXT: testl %ecx, %ecx
; NOSSE-NEXT: movabsq $4611404543450677248, %rdx # imm = 0x3FFF000000000000
; NOSSE-NEXT: cmovneq %rax, %rdx
; NOSSE-NEXT: testl %ebp, %ebp
; NOSSE-NEXT: je .LBB1_2
; NOSSE-NEXT: # %bb.1:
; NOSSE-NEXT: movq %r15, %rax
; NOSSE-NEXT: movq %r14, %rdx
; NOSSE-NEXT: .LBB1_2: # %BB2
; NOSSE-NEXT: popq %rbx
; NOSSE-NEXT: .cfi_def_cfa_offset 40
; NOSSE-NEXT: popq %r12
; NOSSE-NEXT: .cfi_def_cfa_offset 32
; NOSSE-NEXT: popq %r14
; NOSSE-NEXT: .cfi_def_cfa_offset 24
; NOSSE-NEXT: popq %r15
; NOSSE-NEXT: .cfi_def_cfa_offset 16
; NOSSE-NEXT: popq %rbp
; NOSSE-NEXT: .cfi_def_cfa_offset 8
; NOSSE-NEXT: retq
BB0:
%a = fcmp oeq fp128 %0, %1
%b = uitofp i1 %a to fp128
br i1 %a, label %BB1, label %BB2
BB1:
br label %BB2
BB2:
%c = phi fp128 [ %0, %BB0 ], [ %b, %BB1 ]
ret fp128 %c
}