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
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc < %s -mtriple=riscv32 -mattr=+m | FileCheck %s --check-prefixes=RISCV32
define { i128, i8 } @muloti_test(i128 %l, i128 %r) #0 {
; RISCV32-LABEL: muloti_test:
; RISCV32: # %bb.0: # %start
; RISCV32-NEXT: addi sp, sp, -32
; RISCV32-NEXT: sw s0, 28(sp) # 4-byte Folded Spill
; RISCV32-NEXT: sw s1, 24(sp) # 4-byte Folded Spill
; RISCV32-NEXT: sw s2, 20(sp) # 4-byte Folded Spill
; RISCV32-NEXT: sw s3, 16(sp) # 4-byte Folded Spill
; RISCV32-NEXT: sw s4, 12(sp) # 4-byte Folded Spill
; RISCV32-NEXT: lw a3, 12(a1)
; RISCV32-NEXT: lw a7, 12(a2)
; RISCV32-NEXT: lw a6, 8(a1)
; RISCV32-NEXT: lw a4, 0(a2)
; RISCV32-NEXT: lw a5, 0(a1)
; RISCV32-NEXT: lw t3, 4(a1)
; RISCV32-NEXT: lw t0, 8(a2)
; RISCV32-NEXT: lw a2, 4(a2)
; RISCV32-NEXT: mulhu a1, a5, a4
; RISCV32-NEXT: mul t1, t3, a4
; RISCV32-NEXT: add a1, t1, a1
; RISCV32-NEXT: sltu t1, a1, t1
; RISCV32-NEXT: mulhu t2, t3, a4
; RISCV32-NEXT: add t4, t2, t1
; RISCV32-NEXT: mul t1, a5, a2
; RISCV32-NEXT: add a1, t1, a1
; RISCV32-NEXT: sltu t1, a1, t1
; RISCV32-NEXT: mulhu t2, a5, a2
; RISCV32-NEXT: add t1, t2, t1
; RISCV32-NEXT: add t5, t4, t1
; RISCV32-NEXT: mul t6, t3, a2
; RISCV32-NEXT: add s0, t6, t5
; RISCV32-NEXT: mul t1, t0, a5
; RISCV32-NEXT: mul s3, a6, a4
; RISCV32-NEXT: add s4, s3, t1
; RISCV32-NEXT: add t1, s0, s4
; RISCV32-NEXT: sltu t2, t1, s0
; RISCV32-NEXT: sltu t6, s0, t6
; RISCV32-NEXT: sltu t4, t5, t4
; RISCV32-NEXT: mulhu t5, t3, a2
; RISCV32-NEXT: add t4, t5, t4
; RISCV32-NEXT: add s0, t4, t6
; RISCV32-NEXT: mul t4, t3, t0
; RISCV32-NEXT: mul t5, a7, a5
; RISCV32-NEXT: add t4, t5, t4
; RISCV32-NEXT: mulhu s1, t0, a5
; RISCV32-NEXT: add s2, s1, t4
; RISCV32-NEXT: mul t4, a2, a6
; RISCV32-NEXT: mul t5, a3, a4
; RISCV32-NEXT: add t4, t5, t4
; RISCV32-NEXT: mulhu t5, a6, a4
; RISCV32-NEXT: add t6, t5, t4
; RISCV32-NEXT: add t4, t6, s2
; RISCV32-NEXT: sltu s3, s4, s3
; RISCV32-NEXT: add t4, t4, s3
; RISCV32-NEXT: add t4, s0, t4
; RISCV32-NEXT: add t4, t4, t2
; RISCV32-NEXT: beq t4, s0, .LBB0_2
; RISCV32-NEXT: # %bb.1: # %start
; RISCV32-NEXT: sltu t2, t4, s0
; RISCV32-NEXT: .LBB0_2: # %start
; RISCV32-NEXT: sltu s0, s2, s1
; RISCV32-NEXT: snez s1, t3
; RISCV32-NEXT: snez s2, a7
; RISCV32-NEXT: and s1, s2, s1
; RISCV32-NEXT: mulhu s2, a7, a5
; RISCV32-NEXT: snez s2, s2
; RISCV32-NEXT: or s1, s1, s2
; RISCV32-NEXT: mulhu t3, t3, t0
; RISCV32-NEXT: snez t3, t3
; RISCV32-NEXT: or t3, s1, t3
; RISCV32-NEXT: or t3, t3, s0
; RISCV32-NEXT: sltu t5, t6, t5
; RISCV32-NEXT: snez t6, a2
; RISCV32-NEXT: snez s0, a3
; RISCV32-NEXT: and t6, s0, t6
; RISCV32-NEXT: mulhu s0, a3, a4
; RISCV32-NEXT: snez s0, s0
; RISCV32-NEXT: or t6, t6, s0
; RISCV32-NEXT: mulhu a2, a2, a6
; RISCV32-NEXT: snez a2, a2
; RISCV32-NEXT: or a2, t6, a2
; RISCV32-NEXT: or a2, a2, t5
; RISCV32-NEXT: or a7, t0, a7
; RISCV32-NEXT: snez a7, a7
; RISCV32-NEXT: or a3, a6, a3
; RISCV32-NEXT: snez a3, a3
; RISCV32-NEXT: and a3, a3, a7
; RISCV32-NEXT: or a2, a3, a2
; RISCV32-NEXT: or a2, a2, t3
; RISCV32-NEXT: or a2, a2, t2
; RISCV32-NEXT: mul a3, a5, a4
; RISCV32-NEXT: andi a2, a2, 1
; RISCV32-NEXT: sw a3, 0(a0)
; RISCV32-NEXT: sw a1, 4(a0)
; RISCV32-NEXT: sw t1, 8(a0)
; RISCV32-NEXT: sw t4, 12(a0)
; RISCV32-NEXT: sb a2, 16(a0)
; RISCV32-NEXT: lw s0, 28(sp) # 4-byte Folded Reload
; RISCV32-NEXT: lw s1, 24(sp) # 4-byte Folded Reload
; RISCV32-NEXT: lw s2, 20(sp) # 4-byte Folded Reload
; RISCV32-NEXT: lw s3, 16(sp) # 4-byte Folded Reload
; RISCV32-NEXT: lw s4, 12(sp) # 4-byte Folded Reload
; RISCV32-NEXT: addi sp, sp, 32
; RISCV32-NEXT: ret
start:
%0 = tail call { i128, i1 } @llvm.umul.with.overflow.i128(i128 %l, i128 %r) #2
%1 = extractvalue { i128, i1 } %0, 0
%2 = extractvalue { i128, i1 } %0, 1
%3 = zext i1 %2 to i8
%4 = insertvalue { i128, i8 } undef, i128 %1, 0
%5 = insertvalue { i128, i8 } %4, i8 %3, 1
ret { i128, i8 } %5
}
; Function Attrs: nounwind readnone speculatable
declare { i128, i1 } @llvm.umul.with.overflow.i128(i128, i128) #1
attributes #0 = { nounwind readnone }
attributes #1 = { nounwind readnone speculatable }
attributes #2 = { nounwind }
|
