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
173
174
175
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py UTC_ARGS: --version 5
; RUN: opt -p loop-vectorize -S %s | FileCheck %s
target datalayout = "e-m:e-Fn32-i64:64-n32:64-S128-v256:256:256-v512:512:512"
target triple = "powerpc64le-unknown-linux-gnu"
define i1 @select_exit_cond(ptr %start, ptr %end, i64 %N) {
; CHECK-LABEL: define i1 @select_exit_cond(
; CHECK-SAME: ptr [[START:%.*]], ptr [[END:%.*]], i64 [[N:%.*]]) {
; CHECK-NEXT: [[ENTRY:.*]]:
; CHECK-NEXT: [[START2:%.*]] = ptrtoint ptr [[START]] to i64
; CHECK-NEXT: [[END1:%.*]] = ptrtoint ptr [[END]] to i64
; CHECK-NEXT: [[TMP0:%.*]] = freeze i64 [[N]]
; CHECK-NEXT: [[UMAX:%.*]] = call i64 @llvm.umax.i64(i64 [[END1]], i64 [[START2]])
; CHECK-NEXT: [[TMP1:%.*]] = sub i64 [[UMAX]], [[START2]]
; CHECK-NEXT: [[UMIN:%.*]] = call i64 @llvm.umin.i64(i64 [[TMP0]], i64 [[TMP1]])
; CHECK-NEXT: [[TMP2:%.*]] = add i64 [[UMIN]], 1
; CHECK-NEXT: [[MIN_ITERS_CHECK:%.*]] = icmp ult i64 [[TMP2]], 16
; CHECK-NEXT: br i1 [[MIN_ITERS_CHECK]], label %[[SCALAR_PH:.*]], label %[[VECTOR_PH:.*]]
; CHECK: [[VECTOR_PH]]:
; CHECK-NEXT: [[N_MOD_VF:%.*]] = urem i64 [[TMP2]], 16
; CHECK-NEXT: [[N_VEC:%.*]] = sub i64 [[TMP2]], [[N_MOD_VF]]
; CHECK-NEXT: [[IND_END:%.*]] = getelementptr i8, ptr [[START]], i64 [[N_VEC]]
; CHECK-NEXT: br label %[[VECTOR_BODY:.*]]
; CHECK: [[VECTOR_BODY]]:
; CHECK-NEXT: [[INDEX:%.*]] = phi i64 [ 0, %[[VECTOR_PH]] ], [ [[INDEX_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_IND:%.*]] = phi <2 x i64> [ <i64 0, i64 1>, %[[VECTOR_PH]] ], [ [[VEC_IND_NEXT:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP43:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI11:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP44:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI12:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP45:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI13:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP46:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI14:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP47:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI15:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP48:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI16:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP49:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[VEC_PHI17:%.*]] = phi <2 x i64> [ zeroinitializer, %[[VECTOR_PH]] ], [ [[TMP50:%.*]], %[[VECTOR_BODY]] ]
; CHECK-NEXT: [[STEP_ADD:%.*]] = add <2 x i64> [[VEC_IND]], <i64 2, i64 2>
; CHECK-NEXT: [[STEP_ADD4:%.*]] = add <2 x i64> [[STEP_ADD]], <i64 2, i64 2>
; CHECK-NEXT: [[STEP_ADD5:%.*]] = add <2 x i64> [[STEP_ADD4]], <i64 2, i64 2>
; CHECK-NEXT: [[STEP_ADD6:%.*]] = add <2 x i64> [[STEP_ADD5]], <i64 2, i64 2>
; CHECK-NEXT: [[STEP_ADD7:%.*]] = add <2 x i64> [[STEP_ADD6]], <i64 2, i64 2>
; CHECK-NEXT: [[STEP_ADD8:%.*]] = add <2 x i64> [[STEP_ADD7]], <i64 2, i64 2>
; CHECK-NEXT: [[STEP_ADD9:%.*]] = add <2 x i64> [[STEP_ADD8]], <i64 2, i64 2>
; CHECK-NEXT: [[TMP3:%.*]] = add i64 [[INDEX]], 0
; CHECK-NEXT: [[TMP4:%.*]] = add i64 [[INDEX]], 2
; CHECK-NEXT: [[TMP5:%.*]] = add i64 [[INDEX]], 4
; CHECK-NEXT: [[TMP6:%.*]] = add i64 [[INDEX]], 6
; CHECK-NEXT: [[TMP7:%.*]] = add i64 [[INDEX]], 8
; CHECK-NEXT: [[TMP8:%.*]] = add i64 [[INDEX]], 10
; CHECK-NEXT: [[TMP9:%.*]] = add i64 [[INDEX]], 12
; CHECK-NEXT: [[TMP10:%.*]] = add i64 [[INDEX]], 14
; CHECK-NEXT: [[NEXT_GEP:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP3]]
; CHECK-NEXT: [[NEXT_GEP18:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP4]]
; CHECK-NEXT: [[NEXT_GEP19:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP5]]
; CHECK-NEXT: [[NEXT_GEP20:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP6]]
; CHECK-NEXT: [[NEXT_GEP21:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP7]]
; CHECK-NEXT: [[NEXT_GEP22:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP8]]
; CHECK-NEXT: [[NEXT_GEP23:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP9]]
; CHECK-NEXT: [[NEXT_GEP24:%.*]] = getelementptr i8, ptr [[START]], i64 [[TMP10]]
; CHECK-NEXT: [[TMP11:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 0
; CHECK-NEXT: [[TMP12:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 2
; CHECK-NEXT: [[TMP13:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 4
; CHECK-NEXT: [[TMP14:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 6
; CHECK-NEXT: [[TMP15:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 8
; CHECK-NEXT: [[TMP16:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 10
; CHECK-NEXT: [[TMP17:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 12
; CHECK-NEXT: [[TMP18:%.*]] = getelementptr i8, ptr [[NEXT_GEP]], i32 14
; CHECK-NEXT: [[WIDE_LOAD:%.*]] = load <2 x i8>, ptr [[TMP11]], align 1
; CHECK-NEXT: [[WIDE_LOAD25:%.*]] = load <2 x i8>, ptr [[TMP12]], align 1
; CHECK-NEXT: [[WIDE_LOAD26:%.*]] = load <2 x i8>, ptr [[TMP13]], align 1
; CHECK-NEXT: [[WIDE_LOAD27:%.*]] = load <2 x i8>, ptr [[TMP14]], align 1
; CHECK-NEXT: [[WIDE_LOAD28:%.*]] = load <2 x i8>, ptr [[TMP15]], align 1
; CHECK-NEXT: [[WIDE_LOAD29:%.*]] = load <2 x i8>, ptr [[TMP16]], align 1
; CHECK-NEXT: [[WIDE_LOAD30:%.*]] = load <2 x i8>, ptr [[TMP17]], align 1
; CHECK-NEXT: [[WIDE_LOAD31:%.*]] = load <2 x i8>, ptr [[TMP18]], align 1
; CHECK-NEXT: [[TMP19:%.*]] = zext <2 x i8> [[WIDE_LOAD]] to <2 x i64>
; CHECK-NEXT: [[TMP20:%.*]] = zext <2 x i8> [[WIDE_LOAD25]] to <2 x i64>
; CHECK-NEXT: [[TMP21:%.*]] = zext <2 x i8> [[WIDE_LOAD26]] to <2 x i64>
; CHECK-NEXT: [[TMP22:%.*]] = zext <2 x i8> [[WIDE_LOAD27]] to <2 x i64>
; CHECK-NEXT: [[TMP23:%.*]] = zext <2 x i8> [[WIDE_LOAD28]] to <2 x i64>
; CHECK-NEXT: [[TMP24:%.*]] = zext <2 x i8> [[WIDE_LOAD29]] to <2 x i64>
; CHECK-NEXT: [[TMP25:%.*]] = zext <2 x i8> [[WIDE_LOAD30]] to <2 x i64>
; CHECK-NEXT: [[TMP26:%.*]] = zext <2 x i8> [[WIDE_LOAD31]] to <2 x i64>
; CHECK-NEXT: [[TMP27:%.*]] = shl <2 x i64> [[VEC_IND]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP28:%.*]] = shl <2 x i64> [[STEP_ADD]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP29:%.*]] = shl <2 x i64> [[STEP_ADD4]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP30:%.*]] = shl <2 x i64> [[STEP_ADD5]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP31:%.*]] = shl <2 x i64> [[STEP_ADD6]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP32:%.*]] = shl <2 x i64> [[STEP_ADD7]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP33:%.*]] = shl <2 x i64> [[STEP_ADD8]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP34:%.*]] = shl <2 x i64> [[STEP_ADD9]], <i64 1, i64 1>
; CHECK-NEXT: [[TMP35:%.*]] = shl <2 x i64> [[TMP19]], [[TMP27]]
; CHECK-NEXT: [[TMP36:%.*]] = shl <2 x i64> [[TMP20]], [[TMP28]]
; CHECK-NEXT: [[TMP37:%.*]] = shl <2 x i64> [[TMP21]], [[TMP29]]
; CHECK-NEXT: [[TMP38:%.*]] = shl <2 x i64> [[TMP22]], [[TMP30]]
; CHECK-NEXT: [[TMP39:%.*]] = shl <2 x i64> [[TMP23]], [[TMP31]]
; CHECK-NEXT: [[TMP40:%.*]] = shl <2 x i64> [[TMP24]], [[TMP32]]
; CHECK-NEXT: [[TMP41:%.*]] = shl <2 x i64> [[TMP25]], [[TMP33]]
; CHECK-NEXT: [[TMP42:%.*]] = shl <2 x i64> [[TMP26]], [[TMP34]]
; CHECK-NEXT: [[TMP43]] = or <2 x i64> [[TMP35]], [[VEC_PHI]]
; CHECK-NEXT: [[TMP44]] = or <2 x i64> [[TMP36]], [[VEC_PHI11]]
; CHECK-NEXT: [[TMP45]] = or <2 x i64> [[TMP37]], [[VEC_PHI12]]
; CHECK-NEXT: [[TMP46]] = or <2 x i64> [[TMP38]], [[VEC_PHI13]]
; CHECK-NEXT: [[TMP47]] = or <2 x i64> [[TMP39]], [[VEC_PHI14]]
; CHECK-NEXT: [[TMP48]] = or <2 x i64> [[TMP40]], [[VEC_PHI15]]
; CHECK-NEXT: [[TMP49]] = or <2 x i64> [[TMP41]], [[VEC_PHI16]]
; CHECK-NEXT: [[TMP50]] = or <2 x i64> [[TMP42]], [[VEC_PHI17]]
; CHECK-NEXT: [[INDEX_NEXT]] = add nuw i64 [[INDEX]], 16
; CHECK-NEXT: [[VEC_IND_NEXT]] = add <2 x i64> [[STEP_ADD9]], <i64 2, i64 2>
; CHECK-NEXT: [[TMP51:%.*]] = icmp eq i64 [[INDEX_NEXT]], [[N_VEC]]
; CHECK-NEXT: br i1 [[TMP51]], label %[[MIDDLE_BLOCK:.*]], label %[[VECTOR_BODY]], !llvm.loop [[LOOP0:![0-9]+]]
; CHECK: [[MIDDLE_BLOCK]]:
; CHECK-NEXT: [[BIN_RDX:%.*]] = or <2 x i64> [[TMP44]], [[TMP43]]
; CHECK-NEXT: [[BIN_RDX32:%.*]] = or <2 x i64> [[TMP45]], [[BIN_RDX]]
; CHECK-NEXT: [[BIN_RDX33:%.*]] = or <2 x i64> [[TMP46]], [[BIN_RDX32]]
; CHECK-NEXT: [[BIN_RDX34:%.*]] = or <2 x i64> [[TMP47]], [[BIN_RDX33]]
; CHECK-NEXT: [[BIN_RDX35:%.*]] = or <2 x i64> [[TMP48]], [[BIN_RDX34]]
; CHECK-NEXT: [[BIN_RDX36:%.*]] = or <2 x i64> [[TMP49]], [[BIN_RDX35]]
; CHECK-NEXT: [[BIN_RDX37:%.*]] = or <2 x i64> [[TMP50]], [[BIN_RDX36]]
; CHECK-NEXT: [[TMP52:%.*]] = call i64 @llvm.vector.reduce.or.v2i64(<2 x i64> [[BIN_RDX37]])
; CHECK-NEXT: [[CMP_N:%.*]] = icmp eq i64 [[TMP2]], [[N_VEC]]
; CHECK-NEXT: br i1 [[CMP_N]], label %[[EXIT:.*]], label %[[SCALAR_PH]]
; CHECK: [[SCALAR_PH]]:
; CHECK-NEXT: [[BC_RESUME_VAL:%.*]] = phi i64 [ [[N_VEC]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT: [[BC_RESUME_VAL3:%.*]] = phi ptr [ [[IND_END]], %[[MIDDLE_BLOCK]] ], [ [[START]], %[[ENTRY]] ]
; CHECK-NEXT: [[BC_MERGE_RDX:%.*]] = phi i64 [ [[TMP52]], %[[MIDDLE_BLOCK]] ], [ 0, %[[ENTRY]] ]
; CHECK-NEXT: br label %[[LOOP:.*]]
; CHECK: [[LOOP]]:
; CHECK-NEXT: [[IV:%.*]] = phi i64 [ [[BC_RESUME_VAL]], %[[SCALAR_PH]] ], [ [[IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT: [[RED:%.*]] = phi i64 [ [[BC_MERGE_RDX]], %[[SCALAR_PH]] ], [ [[RED_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT: [[PTR_IV:%.*]] = phi ptr [ [[BC_RESUME_VAL3]], %[[SCALAR_PH]] ], [ [[PTR_IV_NEXT:%.*]], %[[LOOP]] ]
; CHECK-NEXT: [[TMP53:%.*]] = load i8, ptr [[PTR_IV]], align 1
; CHECK-NEXT: [[CONV3:%.*]] = zext i8 [[TMP53]] to i64
; CHECK-NEXT: [[MUL:%.*]] = shl i64 [[IV]], 1
; CHECK-NEXT: [[SHL:%.*]] = shl i64 [[CONV3]], [[MUL]]
; CHECK-NEXT: [[RED_NEXT]] = or i64 [[SHL]], [[RED]]
; CHECK-NEXT: [[PTR_IV_NEXT]] = getelementptr i8, ptr [[PTR_IV]], i64 1
; CHECK-NEXT: [[IV_NEXT]] = add i64 [[IV]], 1
; CHECK-NEXT: [[CMP_I166_I:%.*]] = icmp ult ptr [[PTR_IV]], [[END]]
; CHECK-NEXT: [[CMP2:%.*]] = icmp ne i64 [[IV]], [[N]]
; CHECK-NEXT: [[AND:%.*]] = select i1 [[CMP_I166_I]], i1 [[CMP2]], i1 false
; CHECK-NEXT: br i1 [[AND]], label %[[LOOP]], label %[[EXIT]], !llvm.loop [[LOOP3:![0-9]+]]
; CHECK: [[EXIT]]:
; CHECK-NEXT: [[RED_NEXT_LCSSA:%.*]] = phi i64 [ [[RED_NEXT]], %[[LOOP]] ], [ [[TMP52]], %[[MIDDLE_BLOCK]] ]
; CHECK-NEXT: [[RES:%.*]] = icmp eq i64 [[RED_NEXT_LCSSA]], 0
; CHECK-NEXT: ret i1 [[RES]]
;
entry:
br label %loop
loop:
%iv = phi i64 [ 0, %entry ], [ %iv.next, %loop ]
%red = phi i64 [ 0, %entry ], [ %red.next, %loop ]
%ptr.iv = phi ptr [ %start, %entry ], [ %ptr.iv.next, %loop ]
%1 = load i8, ptr %ptr.iv, align 1
%conv3 = zext i8 %1 to i64
%mul = shl i64 %iv, 1
%shl = shl i64 %conv3, %mul
%red.next = or i64 %shl, %red
%ptr.iv.next = getelementptr i8, ptr %ptr.iv, i64 1
%iv.next = add i64 %iv, 1
%cmp.i166.i = icmp ult ptr %ptr.iv, %end
%cmp2 = icmp ne i64 %iv, %N
%and = select i1 %cmp.i166.i, i1 %cmp2, i1 false
br i1 %and, label %loop, label %exit
exit:
%res = icmp eq i64 %red.next, 0
ret i1 %res
}
;.
; CHECK: [[LOOP0]] = distinct !{[[LOOP0]], [[META1:![0-9]+]], [[META2:![0-9]+]]}
; CHECK: [[META1]] = !{!"llvm.loop.isvectorized", i32 1}
; CHECK: [[META2]] = !{!"llvm.loop.unroll.runtime.disable"}
; CHECK: [[LOOP3]] = distinct !{[[LOOP3]], [[META2]], [[META1]]}
;.
|
