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
176
177
178
179
180
181
182
183
|
; NOTE: Assertions have been autogenerated by utils/update_llc_test_checks.py
; RUN: llc -verify-machineinstrs -disable-ppc-instr-form-prep=true -mcpu=pwr9 < %s \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names | FileCheck %s -check-prefix=CHECK-P9
; RUN: llc -verify-machineinstrs -disable-ppc-instr-form-prep=true -mcpu=pwr10 < %s \
; RUN: -ppc-vsr-nums-as-vr -ppc-asm-full-reg-names | FileCheck %s -check-prefix=CHECK-P10
target triple = "powerpc64le-unknown-linux-gnu"
%_elem_type_of_a = type <{ double }>
%_elem_type_of_x = type <{ double }>
%_elem_type_of_y = type <{ double }>
define void @test(i32* dereferenceable(4) %.ial, i32* noalias dereferenceable(4) %.m, i32* noalias dereferenceable(4) %.n, [0 x %_elem_type_of_a]* %.a, i32* noalias dereferenceable(4) %.lda, [0 x %_elem_type_of_x]* noalias %.x, [0 x %_elem_type_of_y]* noalias %.y) {
; CHECK-P9-LABEL: test:
; CHECK-P9: # %bb.0: # %test_entry
; CHECK-P9-NEXT: andi. r3, r6, 15
; CHECK-P9-NEXT: lwz r4, 0(r4)
; CHECK-P9-NEXT: lwz r5, 0(r5)
; CHECK-P9-NEXT: li r11, 1
; CHECK-P9-NEXT: addic r3, r3, -1
; CHECK-P9-NEXT: subfe r10, r3, r3
; CHECK-P9-NEXT: li r3, 2
; CHECK-P9-NEXT: not r10, r10
; CHECK-P9-NEXT: iseleq r3, r11, r3
; CHECK-P9-NEXT: add r4, r10, r4
; CHECK-P9-NEXT: srawi r4, r4, 4
; CHECK-P9-NEXT: addze r4, r4
; CHECK-P9-NEXT: srawi r5, r5, 1
; CHECK-P9-NEXT: slwi r4, r4, 4
; CHECK-P9-NEXT: addze r5, r5
; CHECK-P9-NEXT: sub r4, r4, r10
; CHECK-P9-NEXT: cmpw r3, r4
; CHECK-P9-NEXT: bgtlr cr0
; CHECK-P9-NEXT: # %bb.1: # %_loop_2_do_.lr.ph
; CHECK-P9-NEXT: extswsli r5, r5, 3
; CHECK-P9-NEXT: extsw r10, r4
; CHECK-P9-NEXT: add r5, r8, r5
; CHECK-P9-NEXT: clrldi r8, r3, 32
; CHECK-P9-NEXT: lwa r3, 0(r7)
; CHECK-P9-NEXT: addi r4, r8, 1
; CHECK-P9-NEXT: addi r5, r5, -8
; CHECK-P9-NEXT: lxvdsx vs0, 0, r5
; CHECK-P9-NEXT: sub r3, r4, r3
; CHECK-P9-NEXT: sldi r4, r4, 3
; CHECK-P9-NEXT: sldi r3, r3, 3
; CHECK-P9-NEXT: add r4, r9, r4
; CHECK-P9-NEXT: add r3, r6, r3
; CHECK-P9-NEXT: sub r6, r10, r8
; CHECK-P9-NEXT: rldicl r6, r6, 60, 4
; CHECK-P9-NEXT: addi r6, r6, 1
; CHECK-P9-NEXT: mtctr r6
; CHECK-P9-NEXT: .p2align 4
; CHECK-P9-NEXT: .LBB0_2: # %_loop_2_do_
; CHECK-P9-NEXT: #
; CHECK-P9-NEXT: lxv vs1, -16(r4)
; CHECK-P9-NEXT: lxv vs2, 0(r4)
; CHECK-P9-NEXT: lxv vs3, -16(r3)
; CHECK-P9-NEXT: lxv vs4, 0(r3)
; CHECK-P9-NEXT: addi r3, r3, 128
; CHECK-P9-NEXT: xvmaddadp vs1, vs3, vs1
; CHECK-P9-NEXT: stxv vs1, -16(r4)
; CHECK-P9-NEXT: xvmaddadp vs2, vs4, vs0
; CHECK-P9-NEXT: stxv vs2, 0(r4)
; CHECK-P9-NEXT: addi r4, r4, 128
; CHECK-P9-NEXT: bdnz .LBB0_2
; CHECK-P9-NEXT: # %bb.3: # %_return_bb
; CHECK-P9-NEXT: blr
;
; CHECK-P10-LABEL: test:
; CHECK-P10: # %bb.0: # %test_entry
; CHECK-P10-NEXT: lwz r4, 0(r4)
; CHECK-P10-NEXT: andi. r3, r6, 15
; CHECK-P10-NEXT: li r3, 2
; CHECK-P10-NEXT: li r10, 1
; CHECK-P10-NEXT: lwz r5, 0(r5)
; CHECK-P10-NEXT: iseleq r3, r10, r3
; CHECK-P10-NEXT: setnbc r10, eq
; CHECK-P10-NEXT: not r10, r10
; CHECK-P10-NEXT: add r4, r10, r4
; CHECK-P10-NEXT: srawi r4, r4, 4
; CHECK-P10-NEXT: addze r4, r4
; CHECK-P10-NEXT: srawi r5, r5, 1
; CHECK-P10-NEXT: slwi r4, r4, 4
; CHECK-P10-NEXT: addze r5, r5
; CHECK-P10-NEXT: sub r4, r4, r10
; CHECK-P10-NEXT: cmpw r3, r4
; CHECK-P10-NEXT: bgtlr cr0
; CHECK-P10-NEXT: # %bb.1: # %_loop_2_do_.lr.ph
; CHECK-P10-NEXT: extswsli r5, r5, 3
; CHECK-P10-NEXT: extsw r10, r4
; CHECK-P10-NEXT: add r5, r8, r5
; CHECK-P10-NEXT: clrldi r8, r3, 32
; CHECK-P10-NEXT: lwa r3, 0(r7)
; CHECK-P10-NEXT: addi r4, r8, 1
; CHECK-P10-NEXT: addi r5, r5, -8
; CHECK-P10-NEXT: lxvdsx vs0, 0, r5
; CHECK-P10-NEXT: sub r3, r4, r3
; CHECK-P10-NEXT: sldi r4, r4, 3
; CHECK-P10-NEXT: add r4, r9, r4
; CHECK-P10-NEXT: sldi r3, r3, 3
; CHECK-P10-NEXT: add r3, r6, r3
; CHECK-P10-NEXT: sub r6, r10, r8
; CHECK-P10-NEXT: rldicl r6, r6, 60, 4
; CHECK-P10-NEXT: addi r6, r6, 1
; CHECK-P10-NEXT: mtctr r6
; CHECK-P10-NEXT: .p2align 4
; CHECK-P10-NEXT: .LBB0_2: # %_loop_2_do_
; CHECK-P10-NEXT: #
; CHECK-P10-NEXT: lxv vs1, -16(r4)
; CHECK-P10-NEXT: lxv vs2, 0(r4)
; CHECK-P10-NEXT: lxv vs3, -16(r3)
; CHECK-P10-NEXT: xvmaddadp vs1, vs3, vs1
; CHECK-P10-NEXT: lxv vs4, 0(r3)
; CHECK-P10-NEXT: xvmaddadp vs2, vs4, vs0
; CHECK-P10-NEXT: addi r3, r3, 128
; CHECK-P10-NEXT: stxv vs1, -16(r4)
; CHECK-P10-NEXT: stxv vs2, 0(r4)
; CHECK-P10-NEXT: addi r4, r4, 128
; CHECK-P10-NEXT: bdnz .LBB0_2
; CHECK-P10-NEXT: # %bb.3: # %_return_bb
; CHECK-P10-NEXT: blr
; FIXME: use pair load/store instructions lxvp/stxvp
test_entry:
%_conv5 = ptrtoint [0 x %_elem_type_of_a]* %.a to i64
%_andi_tmp = and i64 %_conv5, 15
%_equ_tmp = icmp eq i64 %_andi_tmp, 0
%. = select i1 %_equ_tmp, i32 1, i32 2
%_val_m_ = load i32, i32* %.m, align 4
%_sub_tmp9 = sub nsw i32 1, %.
%_add_tmp10 = add i32 %_sub_tmp9, %_val_m_
%_mod_tmp = srem i32 %_add_tmp10, 16
%_sub_tmp11 = sub i32 %_val_m_, %_mod_tmp
%_val_n_ = load i32, i32* %.n, align 4
%x_rvo_based_addr_17 = getelementptr inbounds [0 x %_elem_type_of_x], [0 x %_elem_type_of_x]* %.x, i64 0, i64 -1
%_div_tmp = sdiv i32 %_val_n_, 2
%_conv16 = sext i32 %_div_tmp to i64
%_ind_cast = getelementptr inbounds %_elem_type_of_x, %_elem_type_of_x* %x_rvo_based_addr_17, i64 %_conv16, i32 0
%_val_x_ = load double, double* %_ind_cast, align 8
%.splatinsert = insertelement <2 x double> undef, double %_val_x_, i32 0
%.splat = shufflevector <2 x double> %.splatinsert, <2 x double> undef, <2 x i32> zeroinitializer
%_grt_tmp21 = icmp sgt i32 %., %_sub_tmp11
br i1 %_grt_tmp21, label %_return_bb, label %_loop_2_do_.lr.ph
_loop_2_do_.lr.ph: ; preds = %test_entry
%_val_lda_ = load i32, i32* %.lda, align 4
%_conv = sext i32 %_val_lda_ to i64
%_mult_tmp = shl nsw i64 %_conv, 3
%_sub_tmp4 = sub nuw nsw i64 -8, %_mult_tmp
%y_rvo_based_addr_19 = getelementptr inbounds [0 x %_elem_type_of_y], [0 x %_elem_type_of_y]* %.y, i64 0, i64 -1
%a_byte_ptr_ = bitcast [0 x %_elem_type_of_a]* %.a to i8*
%a_rvo_based_addr_ = getelementptr inbounds i8, i8* %a_byte_ptr_, i64 %_sub_tmp4
%0 = zext i32 %. to i64
%1 = sext i32 %_sub_tmp11 to i64
br label %_loop_2_do_
_loop_2_do_: ; preds = %_loop_2_do_.lr.ph, %_loop_2_do_
%indvars.iv = phi i64 [ %0, %_loop_2_do_.lr.ph ], [ %indvars.iv.next, %_loop_2_do_ ]
%_ix_x_len19 = shl nuw nsw i64 %indvars.iv, 3
%y_ix_dim_0_20 = getelementptr inbounds %_elem_type_of_y, %_elem_type_of_y* %y_rvo_based_addr_19, i64 %indvars.iv
%2 = bitcast %_elem_type_of_y* %y_ix_dim_0_20 to <2 x double>*
%3 = load <2 x double>, <2 x double>* %2, align 1
%4 = getelementptr %_elem_type_of_y, %_elem_type_of_y* %y_ix_dim_0_20, i64 2
%5 = bitcast %_elem_type_of_y* %4 to <2 x double>*
%6 = load <2 x double>, <2 x double>* %5, align 1
%a_ix_dim_1_ = getelementptr inbounds i8, i8* %a_rvo_based_addr_, i64 %_ix_x_len19
%7 = bitcast i8* %a_ix_dim_1_ to <2 x double>*
%8 = load <2 x double>, <2 x double>* %7, align 1
%9 = getelementptr i8, i8* %a_ix_dim_1_, i64 16
%10 = bitcast i8* %9 to <2 x double>*
%11 = load <2 x double>, <2 x double>* %10, align 1
%12 = tail call nsz contract <2 x double> @llvm.fma.v2f64(<2 x double> %8, <2 x double> %3, <2 x double> %3)
%13 = tail call nsz contract <2 x double> @llvm.fma.v2f64(<2 x double> %11, <2 x double> %.splat, <2 x double> %6)
store <2 x double> %12, <2 x double>* %2, align 1
store <2 x double> %13, <2 x double>* %5, align 1
%indvars.iv.next = add nuw nsw i64 %indvars.iv, 16
%_grt_tmp = icmp sgt i64 %indvars.iv.next, %1
br i1 %_grt_tmp, label %_return_bb, label %_loop_2_do_
_return_bb: ; preds = %_loop_2_do_, %test_entry
ret void
}
declare <2 x double> @llvm.fma.v2f64(<2 x double>, <2 x double>, <2 x double>)
|
