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
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
|
; RUN: llc -mtriple=i686-pc-win32 -mattr=+sse2 < %s | FileCheck %s --check-prefix=CHECK --check-prefix=X86
; RUN: llc -mtriple=x86_64-pc-win32 < %s | FileCheck %s --check-prefix=CHECK --check-prefix=X64
; Test integer arguments.
define x86_vectorcallcc i32 @test_int_1() {
; CHECK-LABEL: {{^}}test_int_1@@0:
; CHECK: xorl %eax, %eax
ret i32 0
}
define x86_vectorcallcc i32 @test_int_2(i32 inreg %a) {
; X86-LABEL: {{^}}test_int_2@@4:
; X64-LABEL: {{^}}test_int_2@@8:
; CHECK: movl %ecx, %eax
ret i32 %a
}
define x86_vectorcallcc i32 @test_int_3(i64 inreg %a) {
; X86-LABEL: {{^}}test_int_3@@8:
; X64-LABEL: {{^}}test_int_3@@8:
; X86: movl %ecx, %eax
; X64: movq %rcx, %rax
%at = trunc i64 %a to i32
ret i32 %at
}
define x86_vectorcallcc i32 @test_int_4(i32 inreg %a, i32 inreg %b) {
; X86-LABEL: {{^}}test_int_4@@8:
; X86: leal (%ecx,%edx), %eax
; X64-LABEL: {{^}}test_int_4@@16:
; X64: leal (%rcx,%rdx), %eax
%s = add i32 %a, %b
ret i32 %s
}
define x86_vectorcallcc i32 @"\01test_int_5"(i32, i32) {
; CHECK-LABEL: {{^}}test_int_5:
ret i32 0
}
define x86_vectorcallcc double @test_fp_1(double %a, double %b) {
; CHECK-LABEL: {{^}}test_fp_1@@16:
; CHECK: movaps %xmm1, %xmm0
ret double %b
}
define x86_vectorcallcc double @test_fp_2(double, double, double, double, double, double, double %r) {
; CHECK-LABEL: {{^}}test_fp_2@@56:
; CHECK: movsd {{[0-9]+\(%[re]sp\)}}, %xmm0
ret double %r
}
define x86_vectorcallcc {double, double, double, double} @test_fp_3() {
; CHECK-LABEL: {{^}}test_fp_3@@0:
; CHECK: xorps %xmm0
; CHECK: xorps %xmm1
; CHECK: xorps %xmm2
; CHECK: xorps %xmm3
ret {double, double, double, double}
{ double 0.0, double 0.0, double 0.0, double 0.0 }
}
; FIXME: Returning via x87 isn't compatible, but its hard to structure the
; tablegen any other way.
define x86_vectorcallcc {double, double, double, double, double} @test_fp_4() {
; CHECK-LABEL: {{^}}test_fp_4@@0:
; CHECK: fldz
; CHECK: xorps %xmm0
; CHECK: xorps %xmm1
; CHECK: xorps %xmm2
; CHECK: xorps %xmm3
ret {double, double, double, double, double}
{ double 0.0, double 0.0, double 0.0, double 0.0, double 0.0 }
}
define x86_vectorcallcc <16 x i8> @test_vec_1(<16 x i8> %a, <16 x i8> %b) {
; CHECK-LABEL: {{^}}test_vec_1@@32:
; CHECK: movaps %xmm1, %xmm0
ret <16 x i8> %b
}
define x86_vectorcallcc <16 x i8> @test_vec_2(double, <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8>, <16 x i8> %r) {
; CHECK-LABEL: {{^}}test_vec_2@@104:
; X64: movq {{[0-9]*}}(%rsp), %rax
; CHECK: movaps (%{{rax|ecx}}), %xmm0
ret <16 x i8> %r
}
%struct.HVA5 = type { <4 x float>, <4 x float>, <4 x float>, <4 x float>, <4 x float> }
%struct.HVA4 = type { <4 x float>, <4 x float>, <4 x float>, <4 x float> }
%struct.HVA3 = type { <4 x float>, <4 x float>, <4 x float> }
%struct.HVA2 = type { <4 x float>, <4 x float> }
define x86_vectorcallcc <4 x float> @test_mixed_1(i32 %a, %struct.HVA4 inreg %bb, i32 %c) {
; CHECK-LABEL: test_mixed_1
; CHECK: movaps %xmm1, 16(%{{(e|r)}}sp)
; CHECK: movaps %xmm1, %xmm0
; CHECK: ret{{q|l}}
entry:
%b = alloca %struct.HVA4, align 16
store %struct.HVA4 %bb, ptr %b, align 16
%w1 = getelementptr inbounds %struct.HVA4, ptr %b, i32 0, i32 1
%0 = load <4 x float>, ptr %w1, align 16
ret <4 x float> %0
}
define x86_vectorcallcc <4 x float> @test_mixed_2(%struct.HVA4 inreg %a, ptr %b, <4 x float> %c) {
; CHECK-LABEL: test_mixed_2
; X86: movaps %xmm0, (%esp)
; X64: movaps %xmm2, %xmm0
; CHECK: ret{{[ql]}}
entry:
%c.addr = alloca <4 x float>, align 16
store <4 x float> %c, ptr %c.addr, align 16
%0 = load <4 x float>, ptr %c.addr, align 16
ret <4 x float> %0
}
define x86_vectorcallcc <4 x float> @test_mixed_3(<4 x float> %a, <4 x float> %b, <4 x float> %c, <4 x float> %d, <4 x float> %e, ptr %f) {
; CHECK-LABEL: test_mixed_3
; CHECK: movaps (%{{[re][ac]}}x), %xmm0
; CHECK: ret{{[ql]}}
entry:
%0 = load <4 x float>, ptr %f, align 16
ret <4 x float> %0
}
define x86_vectorcallcc <4 x float> @test_mixed_4(%struct.HVA4 inreg %a, ptr %bb, <4 x float> %c) {
; CHECK-LABEL: test_mixed_4
; X86: movaps 16(%eax), %xmm0
; X64: movaps 16(%rdx), %xmm0
; CHECK: ret{{[ql]}}
entry:
%y4 = getelementptr inbounds %struct.HVA2, ptr %bb, i32 0, i32 1
%0 = load <4 x float>, ptr %y4, align 16
ret <4 x float> %0
}
define x86_vectorcallcc <4 x float> @test_mixed_5(%struct.HVA3 inreg %a, ptr %b, <4 x float> %c, %struct.HVA2 inreg %dd) {
; CHECK-LABEL: test_mixed_5
; CHECK-DAG: movaps %xmm{{[0,5]}}, 16(%{{(e|r)}}sp)
; CHECK-DAG: movaps %xmm5, %xmm0
; CHECK: ret{{[ql]}}
entry:
%d = alloca %struct.HVA2, align 16
store %struct.HVA2 %dd, ptr %d, align 16
%y5 = getelementptr inbounds %struct.HVA2, ptr %d, i32 0, i32 1
%0 = load <4 x float>, ptr %y5, align 16
ret <4 x float> %0
}
define x86_vectorcallcc %struct.HVA4 @test_mixed_6(%struct.HVA4 inreg %a, ptr %b) {
; CHECK-LABEL: test_mixed_6
; CHECK: movaps (%{{[re]}}sp), %xmm0
; CHECK: movaps 16(%{{[re]}}sp), %xmm1
; CHECK: movaps 32(%{{[re]}}sp), %xmm2
; CHECK: movaps 48(%{{[re]}}sp), %xmm3
; CHECK: ret{{[ql]}}
entry:
%retval = alloca %struct.HVA4, align 16
call void @llvm.memcpy.p0.p0.i32(ptr align 16 %retval, ptr align 16 %b, i32 64, i1 false)
%0 = load %struct.HVA4, ptr %retval, align 16
ret %struct.HVA4 %0
}
declare void @llvm.memset.p0.i64(ptr nocapture writeonly, i8, i64, i1)
declare void @llvm.memcpy.p0.p0.i64(ptr nocapture writeonly, ptr nocapture readonly, i64, i1)
declare void @llvm.memcpy.p0.p0.i32(ptr nocapture writeonly, ptr nocapture readonly, i32, i1)
define x86_vectorcallcc void @test_mixed_7(ptr noalias sret(%struct.HVA5) %agg.result) {
; CHECK-LABEL: test_mixed_7@@0
; X64: mov{{[ql]}} %rcx, %rax
; CHECK: movaps %xmm{{[0-9]}}, 64(%{{rcx|eax}})
; CHECK: movaps %xmm{{[0-9]}}, 48(%{{rcx|eax}})
; CHECK: movaps %xmm{{[0-9]}}, 32(%{{rcx|eax}})
; CHECK: movaps %xmm{{[0-9]}}, 16(%{{rcx|eax}})
; CHECK: movaps %xmm{{[0-9]}}, (%{{rcx|eax}})
; CHECK: ret{{[ql]}}
entry:
%a = alloca %struct.HVA5, align 16
call void @llvm.memset.p0.i64(ptr align 16 %a, i8 0, i64 80, i1 false)
call void @llvm.memcpy.p0.p0.i64(ptr align 16 %agg.result, ptr align 16 %a, i64 80, i1 false)
ret void
}
define x86_vectorcallcc <4 x float> @test_mixed_8(<4 x float> %a, <4 x float> %b, <4 x float> %c, <4 x float> %d, i32 %e, <4 x float> %f) {
; CHECK-LABEL: test_mixed_8
; X86: movaps %xmm4, %xmm0
; X64: movaps %xmm5, %xmm0
; CHECK: ret{{[ql]}}
entry:
%f.addr = alloca <4 x float>, align 16
store <4 x float> %f, ptr %f.addr, align 16
%0 = load <4 x float>, ptr %f.addr, align 16
ret <4 x float> %0
}
%struct.HFA4 = type { double, double, double, double }
declare x86_vectorcallcc double @test_mixed_9_callee(%struct.HFA4 %x, double %y)
define x86_vectorcallcc double @test_mixed_9_caller(%struct.HFA4 inreg %b) {
; CHECK-LABEL: test_mixed_9_caller
; CHECK: movaps %xmm3, %xmm4
; CHECK: movaps %xmm2, %xmm3
; CHECK: movaps %xmm1, %xmm2
; X32: movasd %xmm0, %xmm1
; X64: movap{{d|s}} %xmm5, %xmm1
; CHECK: call{{l|q}} test_mixed_9_callee@@40
; CHECK: addsd {{.*}}, %xmm0
; CHECK: ret{{l|q}}
entry:
%call = call x86_vectorcallcc double @test_mixed_9_callee(%struct.HFA4 inreg %b, double 3.000000e+00)
%add = fadd double 1.000000e+00, %call
ret double %add
}
|
