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
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -mtriple=x86_64-unknown-unknown -mattr=+avx < %s | FileCheck %s --check-prefix=AVX
; RUN: llc -mtriple=i386-unknown-linux-gnu -mcpu=knl < %s | FileCheck %s --check-prefix=KNL-32
; Verify that we don't crash during codegen due to a wrong lowering
; of a setcc node with illegal operand types and return type.
define <8 x i16> @pr25080(<8 x i32> %a) {
; AVX-LABEL: pr25080:
; AVX: # BB#0: # %entry
; AVX-NEXT: vandps {{.*}}(%rip), %ymm0, %ymm0
; AVX-NEXT: vextractf128 $1, %ymm0, %xmm1
; AVX-NEXT: vpxor %xmm2, %xmm2, %xmm2
; AVX-NEXT: vpcmpeqd %xmm2, %xmm1, %xmm1
; AVX-NEXT: vpcmpeqd %xmm2, %xmm0, %xmm0
; AVX-NEXT: vpacksswb %xmm1, %xmm0, %xmm0
; AVX-NEXT: vpor {{.*}}(%rip), %xmm0, %xmm0
; AVX-NEXT: vpsllw $15, %xmm0, %xmm0
; AVX-NEXT: vpsraw $15, %xmm0, %xmm0
; AVX-NEXT: vzeroupper
; AVX-NEXT: retq
;
; KNL-32-LABEL: pr25080:
; KNL-32: # BB#0: # %entry
; KNL-32-NEXT: vpbroadcastd {{\.LCPI.*}}, %ymm1
; KNL-32-NEXT: vpand %ymm1, %ymm0, %ymm0
; KNL-32-NEXT: vpxor %ymm1, %ymm1, %ymm1
; KNL-32-NEXT: vpcmpeqd %zmm1, %zmm0, %k0
; KNL-32-NEXT: movb $15, %al
; KNL-32-NEXT: kmovw %eax, %k1
; KNL-32-NEXT: korw %k1, %k0, %k1
; KNL-32-NEXT: vpternlogq $255, %zmm0, %zmm0, %zmm0 {%k1} {z}
; KNL-32-NEXT: vpmovqw %zmm0, %xmm0
; KNL-32-NEXT: retl
entry:
%0 = trunc <8 x i32> %a to <8 x i23>
%1 = icmp eq <8 x i23> %0, zeroinitializer
%2 = or <8 x i1> %1, <i1 true, i1 true, i1 true, i1 true, i1 false, i1 false, i1 false, i1 false>
%3 = sext <8 x i1> %2 to <8 x i16>
ret <8 x i16> %3
}
define void @pr26232(i64 %a, <16 x i1> %b) {
; AVX-LABEL: pr26232:
; AVX: # BB#0: # %for_loop599.preheader
; AVX-NEXT: vpxor %xmm1, %xmm1, %xmm1
; AVX-NEXT: .p2align 4, 0x90
; AVX-NEXT: .LBB1_1: # %for_loop599
; AVX-NEXT: # =>This Inner Loop Header: Depth=1
; AVX-NEXT: xorl %eax, %eax
; AVX-NEXT: cmpq $65536, %rdi # imm = 0x10000
; AVX-NEXT: setl %al
; AVX-NEXT: vmovd %eax, %xmm2
; AVX-NEXT: vpshufb %xmm1, %xmm2, %xmm2
; AVX-NEXT: vpand %xmm0, %xmm2, %xmm2
; AVX-NEXT: vpextrb $15, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $14, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $13, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $12, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $11, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $10, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $9, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $8, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $7, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $6, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $5, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $4, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $3, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $2, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $1, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: vpextrb $0, %xmm2, %eax
; AVX-NEXT: andb $1, %al
; AVX-NEXT: movb %al, -{{[0-9]+}}(%rsp)
; AVX-NEXT: cmpw $0, -{{[0-9]+}}(%rsp)
; AVX-NEXT: jne .LBB1_1
; AVX-NEXT: # BB#2: # %for_exit600
; AVX-NEXT: retq
;
; KNL-32-LABEL: pr26232:
; KNL-32: # BB#0: # %for_loop599.preheader
; KNL-32-NEXT: pushl %esi
; KNL-32-NEXT: .Lcfi0:
; KNL-32-NEXT: .cfi_def_cfa_offset 8
; KNL-32-NEXT: .Lcfi1:
; KNL-32-NEXT: .cfi_offset %esi, -8
; KNL-32-NEXT: vpmovsxbd %xmm0, %zmm0
; KNL-32-NEXT: vpslld $31, %zmm0, %zmm0
; KNL-32-NEXT: vptestmd %zmm0, %zmm0, %k0
; KNL-32-NEXT: movl {{[0-9]+}}(%esp), %eax
; KNL-32-NEXT: movl {{[0-9]+}}(%esp), %ecx
; KNL-32-NEXT: movw $-1, %dx
; KNL-32-NEXT: .p2align 4, 0x90
; KNL-32-NEXT: .LBB1_1: # %for_loop599
; KNL-32-NEXT: # =>This Inner Loop Header: Depth=1
; KNL-32-NEXT: cmpl $65536, %ecx # imm = 0x10000
; KNL-32-NEXT: movl %eax, %esi
; KNL-32-NEXT: sbbl $0, %esi
; KNL-32-NEXT: movl $0, %esi
; KNL-32-NEXT: cmovlw %dx, %si
; KNL-32-NEXT: kmovw %esi, %k1
; KNL-32-NEXT: kandw %k0, %k1, %k1
; KNL-32-NEXT: kmovw %k1, %esi
; KNL-32-NEXT: testw %si, %si
; KNL-32-NEXT: jne .LBB1_1
; KNL-32-NEXT: # BB#2: # %for_exit600
; KNL-32-NEXT: popl %esi
; KNL-32-NEXT: retl
allocas:
br label %for_test11.preheader
for_test11.preheader: ; preds = %for_test11.preheader, %allocas
br i1 undef, label %for_loop599, label %for_test11.preheader
for_loop599: ; preds = %for_loop599, %for_test11.preheader
%less_i_load605_ = icmp slt i64 %a, 65536
%less_i_load605__broadcast_init = insertelement <16 x i1> undef, i1 %less_i_load605_, i32 0
%less_i_load605__broadcast = shufflevector <16 x i1> %less_i_load605__broadcast_init, <16 x i1> undef, <16 x i32> zeroinitializer
%"oldMask&test607" = and <16 x i1> %less_i_load605__broadcast, %b
%intmask.i894 = bitcast <16 x i1> %"oldMask&test607" to i16
%res.i895 = icmp eq i16 %intmask.i894, 0
br i1 %res.i895, label %for_exit600, label %for_loop599
for_exit600: ; preds = %for_loop599
ret void
}
|
