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
|
; NOTE: Assertions have been autogenerated by utils/update_test_checks.py
; RUN: opt < %s -passes='default<O3>' -S | FileCheck %s
target datalayout = "e-m:e-p:32:32-Fi8-i64:64-v128:64:128-a:0:32-n32-S64"
target triple = "thumbv6m-none-none-eabi"
; Make sure we don't make a mess of vectorization/unrolling of the remainder loop.
define void @arm_mean_q7(ptr noundef %pSrc, i32 noundef %blockSize, ptr noundef %pResult) #0 {
; CHECK-LABEL: @arm_mean_q7(
; CHECK-NEXT: entry:
; CHECK-NEXT: [[CMP_NOT10:%.*]] = icmp ult i32 [[BLOCKSIZE:%.*]], 16
; CHECK-NEXT: br i1 [[CMP_NOT10]], label [[WHILE_END:%.*]], label [[WHILE_BODY_PREHEADER:%.*]]
; CHECK: while.body.preheader:
; CHECK-NEXT: [[SHR:%.*]] = lshr i32 [[BLOCKSIZE]], 4
; CHECK-NEXT: [[TMP0:%.*]] = and i32 [[BLOCKSIZE]], -16
; CHECK-NEXT: br label [[WHILE_BODY:%.*]]
; CHECK: while.body:
; CHECK-NEXT: [[SUM_013:%.*]] = phi i32 [ [[TMP3:%.*]], [[WHILE_BODY]] ], [ 0, [[WHILE_BODY_PREHEADER]] ]
; CHECK-NEXT: [[PSRC_ADDR_012:%.*]] = phi ptr [ [[ADD_PTR:%.*]], [[WHILE_BODY]] ], [ [[PSRC:%.*]], [[WHILE_BODY_PREHEADER]] ]
; CHECK-NEXT: [[BLKCNT_011:%.*]] = phi i32 [ [[DEC:%.*]], [[WHILE_BODY]] ], [ [[SHR]], [[WHILE_BODY_PREHEADER]] ]
; CHECK-NEXT: [[TMP1:%.*]] = load <16 x i8>, ptr [[PSRC_ADDR_012]], align 1
; CHECK-NEXT: [[TMP2:%.*]] = tail call i32 @llvm.arm.mve.addv.v16i8(<16 x i8> [[TMP1]], i32 0)
; CHECK-NEXT: [[TMP3]] = add i32 [[TMP2]], [[SUM_013]]
; CHECK-NEXT: [[DEC]] = add nsw i32 [[BLKCNT_011]], -1
; CHECK-NEXT: [[ADD_PTR]] = getelementptr inbounds i8, ptr [[PSRC_ADDR_012]], i32 16
; CHECK-NEXT: [[CMP_NOT:%.*]] = icmp eq i32 [[DEC]], 0
; CHECK-NEXT: br i1 [[CMP_NOT]], label [[WHILE_END_LOOPEXIT:%.*]], label [[WHILE_BODY]]
; CHECK: while.end.loopexit:
; CHECK-NEXT: [[SCEVGEP:%.*]] = getelementptr i8, ptr [[PSRC]], i32 [[TMP0]]
; CHECK-NEXT: br label [[WHILE_END]]
; CHECK: while.end:
; CHECK-NEXT: [[PSRC_ADDR_0_LCSSA:%.*]] = phi ptr [ [[PSRC]], [[ENTRY:%.*]] ], [ [[SCEVGEP]], [[WHILE_END_LOOPEXIT]] ]
; CHECK-NEXT: [[SUM_0_LCSSA:%.*]] = phi i32 [ 0, [[ENTRY]] ], [ [[TMP3]], [[WHILE_END_LOOPEXIT]] ]
; CHECK-NEXT: [[AND:%.*]] = and i32 [[BLOCKSIZE]], 15
; CHECK-NEXT: [[CMP2_NOT15:%.*]] = icmp eq i32 [[AND]], 0
; CHECK-NEXT: br i1 [[CMP2_NOT15]], label [[WHILE_END5:%.*]], label [[MIDDLE_BLOCK:%.*]]
; CHECK: middle.block:
; CHECK-NEXT: [[ACTIVE_LANE_MASK:%.*]] = tail call <16 x i1> @llvm.get.active.lane.mask.v16i1.i32(i32 0, i32 [[AND]])
; CHECK-NEXT: [[WIDE_MASKED_LOAD:%.*]] = tail call <16 x i8> @llvm.masked.load.v16i8.p0(ptr [[PSRC_ADDR_0_LCSSA]], i32 1, <16 x i1> [[ACTIVE_LANE_MASK]], <16 x i8> poison)
; CHECK-NEXT: [[TMP4:%.*]] = sext <16 x i8> [[WIDE_MASKED_LOAD]] to <16 x i32>
; CHECK-NEXT: [[TMP5:%.*]] = select <16 x i1> [[ACTIVE_LANE_MASK]], <16 x i32> [[TMP4]], <16 x i32> zeroinitializer
; CHECK-NEXT: [[TMP6:%.*]] = tail call i32 @llvm.vector.reduce.add.v16i32(<16 x i32> [[TMP5]])
; CHECK-NEXT: [[TMP7:%.*]] = add i32 [[TMP6]], [[SUM_0_LCSSA]]
; CHECK-NEXT: br label [[WHILE_END5]]
; CHECK: while.end5:
; CHECK-NEXT: [[SUM_1_LCSSA:%.*]] = phi i32 [ [[SUM_0_LCSSA]], [[WHILE_END]] ], [ [[TMP7]], [[MIDDLE_BLOCK]] ]
; CHECK-NEXT: [[DIV:%.*]] = sdiv i32 [[SUM_1_LCSSA]], [[BLOCKSIZE]]
; CHECK-NEXT: [[CONV6:%.*]] = trunc i32 [[DIV]] to i8
; CHECK-NEXT: store i8 [[CONV6]], ptr [[PRESULT:%.*]], align 1
; CHECK-NEXT: ret void
;
entry:
%pSrc.addr = alloca ptr, align 4
%blockSize.addr = alloca i32, align 4
%pResult.addr = alloca ptr, align 4
%blkCnt = alloca i32, align 4
%vecSrc = alloca <16 x i8>, align 1
%sum = alloca i32, align 4
store ptr %pSrc, ptr %pSrc.addr, align 4
store i32 %blockSize, ptr %blockSize.addr, align 4
store ptr %pResult, ptr %pResult.addr, align 4
call void @llvm.lifetime.start.p0(i64 4, ptr %blkCnt) #3
call void @llvm.lifetime.start.p0(i64 16, ptr %vecSrc) #3
call void @llvm.lifetime.start.p0(i64 4, ptr %sum) #3
store i32 0, ptr %sum, align 4
%0 = load i32, ptr %blockSize.addr, align 4
%shr = lshr i32 %0, 4
store i32 %shr, ptr %blkCnt, align 4
br label %while.cond
while.cond: ; preds = %while.body, %entry
%1 = load i32, ptr %blkCnt, align 4
%cmp = icmp ugt i32 %1, 0
br i1 %cmp, label %while.body, label %while.end
while.body: ; preds = %while.cond
%2 = load ptr, ptr %pSrc.addr, align 4
%3 = load <16 x i8>, ptr %2, align 1
store <16 x i8> %3, ptr %vecSrc, align 1
%4 = load <16 x i8>, ptr %vecSrc, align 1
%5 = call i32 @llvm.arm.mve.addv.v16i8(<16 x i8> %4, i32 0)
%6 = load i32, ptr %sum, align 4
%7 = add i32 %5, %6
store i32 %7, ptr %sum, align 4
%8 = load i32, ptr %blkCnt, align 4
%dec = add i32 %8, -1
store i32 %dec, ptr %blkCnt, align 4
%9 = load ptr, ptr %pSrc.addr, align 4
%add.ptr = getelementptr inbounds i8, ptr %9, i32 16
store ptr %add.ptr, ptr %pSrc.addr, align 4
br label %while.cond
while.end: ; preds = %while.cond
%10 = load i32, ptr %blockSize.addr, align 4
%and = and i32 %10, 15
store i32 %and, ptr %blkCnt, align 4
br label %while.cond1
while.cond1: ; preds = %while.body3, %while.end
%11 = load i32, ptr %blkCnt, align 4
%cmp2 = icmp ugt i32 %11, 0
br i1 %cmp2, label %while.body3, label %while.end5
while.body3: ; preds = %while.cond1
%12 = load ptr, ptr %pSrc.addr, align 4
%incdec.ptr = getelementptr inbounds i8, ptr %12, i32 1
store ptr %incdec.ptr, ptr %pSrc.addr, align 4
%13 = load i8, ptr %12, align 1
%conv = sext i8 %13 to i32
%14 = load i32, ptr %sum, align 4
%add = add nsw i32 %14, %conv
store i32 %add, ptr %sum, align 4
%15 = load i32, ptr %blkCnt, align 4
%dec4 = add i32 %15, -1
store i32 %dec4, ptr %blkCnt, align 4
br label %while.cond1
while.end5: ; preds = %while.cond1
%16 = load i32, ptr %sum, align 4
%17 = load i32, ptr %blockSize.addr, align 4
%div = sdiv i32 %16, %17
%conv6 = trunc i32 %div to i8
%18 = load ptr, ptr %pResult.addr, align 4
store i8 %conv6, ptr %18, align 1
call void @llvm.lifetime.end.p0(i64 4, ptr %sum) #3
call void @llvm.lifetime.end.p0(i64 16, ptr %vecSrc) #3
call void @llvm.lifetime.end.p0(i64 4, ptr %blkCnt) #3
ret void
}
declare void @llvm.lifetime.start.p0(i64 immarg, ptr nocapture) #1
declare i32 @llvm.arm.mve.addv.v16i8(<16 x i8>, i32) #2
declare void @llvm.lifetime.end.p0(i64 immarg, ptr nocapture) #1
attributes #0 = { nounwind "approx-func-fp-math"="true" "frame-pointer"="all" "min-legal-vector-width"="0" "no-infs-fp-math"="true" "no-nans-fp-math"="true" "no-signed-zeros-fp-math"="true" "no-trapping-math"="true" "stack-protector-buffer-size"="8" "target-cpu"="cortex-m55" "target-features"="+armv8.1-m.main,+dsp,+fp-armv8d16,+fp-armv8d16sp,+fp16,+fp64,+fullfp16,+hwdiv,+lob,+mve,+mve.fp,+ras,+strict-align,+thumb-mode,+vfp2,+vfp2sp,+vfp3d16,+vfp3d16sp,+vfp4d16,+vfp4d16sp,-aes,-bf16,-cdecp0,-cdecp1,-cdecp2,-cdecp3,-cdecp4,-cdecp5,-cdecp6,-cdecp7,-crc,-crypto,-d32,-dotprod,-fp-armv8,-fp-armv8sp,-fp16fml,-hwdiv-arm,-i8mm,-neon,-pacbti,-sb,-sha2,-vfp3,-vfp3sp,-vfp4,-vfp4sp" "unsafe-fp-math"="true" }
attributes #1 = { argmemonly nocallback nofree nosync nounwind willreturn }
attributes #2 = { nounwind readnone }
attributes #3 = { nounwind }
|
