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
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=i686-unknown -mattr=+sse2 | FileCheck %s --check-prefix=X86
; RUN: llc < %s -mtriple=x86_64-unknown -mattr=+avx | FileCheck %s --check-prefix=X64
; These tests just check that the plumbing is in place for @llvm.bitreverse. The
; actual output is massive at the moment as llvm.bitreverse is not yet legal.
declare i32 @llvm.bitreverse.i32(i32) readnone
declare i64 @llvm.bitreverse.i64(i64) readnone
declare <4 x i32> @llvm.bitreverse.v4i32(<4 x i32>) readnone
declare i32 @llvm.bswap.i32(i32) readnone
; fold (bitreverse undef) -> undef
define i32 @test_undef() nounwind {
; X86-LABEL: test_undef:
; X86: # %bb.0:
; X86-NEXT: retl
;
; X64-LABEL: test_undef:
; X64: # %bb.0:
; X64-NEXT: retq
%b = call i32 @llvm.bitreverse.i32(i32 undef)
ret i32 %b
}
; fold (bitreverse (bitreverse x)) -> x
define i32 @test_bitreverse_bitreverse(i32 %a0) nounwind {
; X86-LABEL: test_bitreverse_bitreverse:
; X86: # %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: retl
;
; X64-LABEL: test_bitreverse_bitreverse:
; X64: # %bb.0:
; X64-NEXT: movl %edi, %eax
; X64-NEXT: retq
%b = call i32 @llvm.bitreverse.i32(i32 %a0)
%c = call i32 @llvm.bitreverse.i32(i32 %b)
ret i32 %c
}
; fold (bitreverse(srl (bitreverse c), x)) -> (shl c, x)
define i32 @test_bitreverse_srli_bitreverse(i32 %a0) nounwind {
; X86-LABEL: test_bitreverse_srli_bitreverse:
; X86: # %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: shll $7, %eax
; X86-NEXT: retl
;
; X64-LABEL: test_bitreverse_srli_bitreverse:
; X64: # %bb.0:
; X64-NEXT: movl %edi, %eax
; X64-NEXT: shll $7, %eax
; X64-NEXT: retq
%b = call i32 @llvm.bitreverse.i32(i32 %a0)
%c = lshr i32 %b, 7
%d = call i32 @llvm.bitreverse.i32(i32 %c)
ret i32 %d
}
define i64 @test_bitreverse_srli_bitreverse_i64(i64 %a) nounwind {
; X86-LABEL: test_bitreverse_srli_bitreverse_i64:
; X86: # %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %edx
; X86-NEXT: addl %edx, %edx
; X86-NEXT: xorl %eax, %eax
; X86-NEXT: retl
;
; X64-LABEL: test_bitreverse_srli_bitreverse_i64:
; X64: # %bb.0:
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: shlq $33, %rax
; X64-NEXT: retq
%1 = call i64 @llvm.bitreverse.i64(i64 %a)
%2 = lshr i64 %1, 33
%3 = call i64 @llvm.bitreverse.i64(i64 %2)
ret i64 %3
}
; fold (bitreverse(shl (bitreverse c), x)) -> (srl c, x)
define i32 @test_bitreverse_shli_bitreverse(i32 %a0) nounwind {
; X86-LABEL: test_bitreverse_shli_bitreverse:
; X86: # %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: shrl $7, %eax
; X86-NEXT: retl
;
; X64-LABEL: test_bitreverse_shli_bitreverse:
; X64: # %bb.0:
; X64-NEXT: movl %edi, %eax
; X64-NEXT: shrl $7, %eax
; X64-NEXT: retq
%b = call i32 @llvm.bitreverse.i32(i32 %a0)
%c = shl i32 %b, 7
%d = call i32 @llvm.bitreverse.i32(i32 %c)
ret i32 %d
}
define i64 @test_bitreverse_shli_bitreverse_i64(i64 %a) nounwind {
; X86-LABEL: test_bitreverse_shli_bitreverse_i64:
; X86: # %bb.0:
; X86-NEXT: movl {{[0-9]+}}(%esp), %eax
; X86-NEXT: shrl %eax
; X86-NEXT: xorl %edx, %edx
; X86-NEXT: retl
;
; X64-LABEL: test_bitreverse_shli_bitreverse_i64:
; X64: # %bb.0:
; X64-NEXT: movq %rdi, %rax
; X64-NEXT: shrq $33, %rax
; X64-NEXT: retq
%1 = call i64 @llvm.bitreverse.i64(i64 %a)
%2 = shl i64 %1, 33
%3 = call i64 @llvm.bitreverse.i64(i64 %2)
ret i64 %3
}
define <4 x i32> @test_demandedbits_bitreverse(<4 x i32> %a0) nounwind {
; X86-LABEL: test_demandedbits_bitreverse:
; X86: # %bb.0:
; X86-NEXT: pxor %xmm1, %xmm1
; X86-NEXT: movdqa %xmm0, %xmm2
; X86-NEXT: punpckhbw {{.*#+}} xmm2 = xmm2[8],xmm1[8],xmm2[9],xmm1[9],xmm2[10],xmm1[10],xmm2[11],xmm1[11],xmm2[12],xmm1[12],xmm2[13],xmm1[13],xmm2[14],xmm1[14],xmm2[15],xmm1[15]
; X86-NEXT: pshuflw {{.*#+}} xmm2 = xmm2[3,2,1,0,4,5,6,7]
; X86-NEXT: pshufhw {{.*#+}} xmm2 = xmm2[0,1,2,3,7,6,5,4]
; X86-NEXT: punpcklbw {{.*#+}} xmm0 = xmm0[0],xmm1[0],xmm0[1],xmm1[1],xmm0[2],xmm1[2],xmm0[3],xmm1[3],xmm0[4],xmm1[4],xmm0[5],xmm1[5],xmm0[6],xmm1[6],xmm0[7],xmm1[7]
; X86-NEXT: pshuflw {{.*#+}} xmm0 = xmm0[3,2,1,0,4,5,6,7]
; X86-NEXT: pshufhw {{.*#+}} xmm0 = xmm0[0,1,2,3,7,6,5,4]
; X86-NEXT: packuswb %xmm2, %xmm0
; X86-NEXT: movdqa %xmm0, %xmm1
; X86-NEXT: psrlw $4, %xmm1
; X86-NEXT: movdqa {{.*#+}} xmm2 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
; X86-NEXT: pand %xmm2, %xmm1
; X86-NEXT: pand %xmm2, %xmm0
; X86-NEXT: psllw $4, %xmm0
; X86-NEXT: por %xmm1, %xmm0
; X86-NEXT: movdqa %xmm0, %xmm1
; X86-NEXT: psrlw $2, %xmm1
; X86-NEXT: movdqa {{.*#+}} xmm2 = [51,51,51,51,51,51,51,51,51,51,51,51,51,51,51,51]
; X86-NEXT: pand %xmm2, %xmm1
; X86-NEXT: pand %xmm2, %xmm0
; X86-NEXT: psllw $2, %xmm0
; X86-NEXT: por %xmm1, %xmm0
; X86-NEXT: movdqa %xmm0, %xmm1
; X86-NEXT: psrlw $1, %xmm1
; X86-NEXT: movdqa {{.*#+}} xmm2 = [85,85,85,85,85,85,85,85,85,85,85,85,85,85,85,85]
; X86-NEXT: pand %xmm2, %xmm1
; X86-NEXT: pand %xmm2, %xmm0
; X86-NEXT: paddb %xmm0, %xmm0
; X86-NEXT: por %xmm1, %xmm0
; X86-NEXT: pand {{\.?LCPI[0-9]+_[0-9]+}}, %xmm0
; X86-NEXT: retl
;
; X64-LABEL: test_demandedbits_bitreverse:
; X64: # %bb.0:
; X64-NEXT: vpshufb {{.*#+}} xmm0 = xmm0[3,2,1,0,7,6,5,4,11,10,9,8,15,14,13,12]
; X64-NEXT: vbroadcastss {{.*#+}} xmm1 = [15,15,15,15,15,15,15,15,15,15,15,15,15,15,15,15]
; X64-NEXT: vpand %xmm1, %xmm0, %xmm2
; X64-NEXT: vmovdqa {{.*#+}} xmm3 = [0,128,64,192,32,160,96,224,16,144,80,208,48,176,112,240]
; X64-NEXT: vpshufb %xmm2, %xmm3, %xmm2
; X64-NEXT: vpsrlw $4, %xmm0, %xmm0
; X64-NEXT: vpand %xmm1, %xmm0, %xmm0
; X64-NEXT: vmovdqa {{.*#+}} xmm1 = [0,8,4,12,2,10,6,14,1,9,5,13,3,11,7,15]
; X64-NEXT: vpshufb %xmm0, %xmm1, %xmm0
; X64-NEXT: vpor %xmm0, %xmm2, %xmm0
; X64-NEXT: vpand {{\.?LCPI[0-9]+_[0-9]+}}(%rip), %xmm0, %xmm0
; X64-NEXT: retq
%b = or <4 x i32> %a0, <i32 2147483648, i32 2147483648, i32 2147483648, i32 2147483648>
%c = call <4 x i32> @llvm.bitreverse.v4i32(<4 x i32> %b)
%d = and <4 x i32> %c, <i32 -2, i32 -2, i32 -2, i32 -2>
ret <4 x i32> %d
}
|
