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
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
|
// NOTE: Assertions have been autogenerated by utils/update_cc_test_checks.py
// RUN: %clang_cc1 -triple spirv64-amd-amdhsa -x hip \
// RUN: -aux-triple x86_64-unknown-linux-gnu -fcuda-is-device -emit-llvm %s \
// RUN: -o - | FileCheck %s
// RUN: %clang_cc1 -triple spirv64-amd-amdhsa -x hip \
// RUN: -aux-triple x86_64-pc-windows-msvc -fcuda-is-device -emit-llvm %s \
// RUN: -o - | FileCheck %s
#include "Inputs/cuda.h"
// CHECK-LABEL: @_Z16use_dispatch_ptrPi(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[OUT:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[DISPATCH_PTR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ASCAST:%.*]] = addrspacecast ptr [[OUT]] to ptr addrspace(4)
// CHECK-NEXT: [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr [[OUT_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[DISPATCH_PTR_ASCAST:%.*]] = addrspacecast ptr [[DISPATCH_PTR]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[OUT_COERCE:%.*]], ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: [[OUT1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: store ptr addrspace(4) [[OUT1]], ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = call align 4 dereferenceable(64) addrspace(4) ptr addrspace(4) @llvm.amdgcn.dispatch.ptr()
// CHECK-NEXT: store ptr addrspace(4) [[TMP0]], ptr addrspace(4) [[DISPATCH_PTR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[DISPATCH_PTR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4
// CHECK-NEXT: [[TMP3:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i32 [[TMP2]], ptr addrspace(4) [[TMP3]], align 4
// CHECK-NEXT: ret void
//
__global__ void use_dispatch_ptr(int* out) {
const int* dispatch_ptr = (const int*)__builtin_amdgcn_dispatch_ptr();
*out = *dispatch_ptr;
}
// CHECK-LABEL: @_Z13use_queue_ptrPi(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[OUT:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[QUEUE_PTR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ASCAST:%.*]] = addrspacecast ptr [[OUT]] to ptr addrspace(4)
// CHECK-NEXT: [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr [[OUT_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[QUEUE_PTR_ASCAST:%.*]] = addrspacecast ptr [[QUEUE_PTR]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[OUT_COERCE:%.*]], ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: [[OUT1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: store ptr addrspace(4) [[OUT1]], ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = call addrspace(4) ptr addrspace(4) @llvm.amdgcn.queue.ptr()
// CHECK-NEXT: store ptr addrspace(4) [[TMP0]], ptr addrspace(4) [[QUEUE_PTR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[QUEUE_PTR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4
// CHECK-NEXT: [[TMP3:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i32 [[TMP2]], ptr addrspace(4) [[TMP3]], align 4
// CHECK-NEXT: ret void
//
__global__ void use_queue_ptr(int* out) {
const int* queue_ptr = (const int*)__builtin_amdgcn_queue_ptr();
*out = *queue_ptr;
}
// CHECK-LABEL: @_Z19use_implicitarg_ptrPi(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[OUT:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[IMPLICITARG_PTR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ASCAST:%.*]] = addrspacecast ptr [[OUT]] to ptr addrspace(4)
// CHECK-NEXT: [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr [[OUT_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[IMPLICITARG_PTR_ASCAST:%.*]] = addrspacecast ptr [[IMPLICITARG_PTR]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[OUT_COERCE:%.*]], ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: [[OUT1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: store ptr addrspace(4) [[OUT1]], ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = call addrspace(4) ptr addrspace(4) @llvm.amdgcn.implicitarg.ptr()
// CHECK-NEXT: store ptr addrspace(4) [[TMP0]], ptr addrspace(4) [[IMPLICITARG_PTR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[IMPLICITARG_PTR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = load i32, ptr addrspace(4) [[TMP1]], align 4
// CHECK-NEXT: [[TMP3:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i32 [[TMP2]], ptr addrspace(4) [[TMP3]], align 4
// CHECK-NEXT: ret void
//
__global__ void use_implicitarg_ptr(int* out) {
const int* implicitarg_ptr = (const int*)__builtin_amdgcn_implicitarg_ptr();
*out = *implicitarg_ptr;
}
__global__
//
void
// CHECK-LABEL: @_Z12test_ds_fmaxf(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SRC_ADDR:%.*]] = alloca float, align 4
// CHECK-NEXT: [[X:%.*]] = alloca float, align 4
// CHECK-NEXT: [[SRC_ADDR_ASCAST:%.*]] = addrspacecast ptr [[SRC_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[X_ASCAST:%.*]] = addrspacecast ptr [[X]] to ptr addrspace(4)
// CHECK-NEXT: store float [[SRC:%.*]], ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: [[TMP0:%.*]] = load float, ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: [[TMP1:%.*]] = atomicrmw fmax ptr addrspace(3) @_ZZ12test_ds_fmaxfE6shared, float [[TMP0]] monotonic, align 4
// CHECK-NEXT: store volatile float [[TMP1]], ptr addrspace(4) [[X_ASCAST]], align 4
// CHECK-NEXT: ret void
//
test_ds_fmax(float src) {
__shared__ float shared;
volatile float x = __builtin_amdgcn_ds_fmaxf(&shared, src, 0, 0, false);
}
// CHECK-LABEL: @_Z12test_ds_faddf(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SRC_ADDR:%.*]] = alloca float, align 4
// CHECK-NEXT: [[X:%.*]] = alloca float, align 4
// CHECK-NEXT: [[SRC_ADDR_ASCAST:%.*]] = addrspacecast ptr [[SRC_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[X_ASCAST:%.*]] = addrspacecast ptr [[X]] to ptr addrspace(4)
// CHECK-NEXT: store float [[SRC:%.*]], ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: [[TMP0:%.*]] = load float, ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: [[TMP1:%.*]] = atomicrmw fadd ptr addrspace(3) @_ZZ12test_ds_faddfE6shared, float [[TMP0]] monotonic, align 4
// CHECK-NEXT: store volatile float [[TMP1]], ptr addrspace(4) [[X_ASCAST]], align 4
// CHECK-NEXT: ret void
//
__global__ void test_ds_fadd(float src) {
__shared__ float shared;
volatile float x = __builtin_amdgcn_ds_faddf(&shared, src, 0, 0, false);
}
// CHECK-LABEL: @_Z12test_ds_fminfPf(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SHARED:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[SRC_ADDR:%.*]] = alloca float, align 4
// CHECK-NEXT: [[SHARED_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[X:%.*]] = alloca float, align 4
// CHECK-NEXT: [[SHARED_ASCAST:%.*]] = addrspacecast ptr [[SHARED]] to ptr addrspace(4)
// CHECK-NEXT: [[SRC_ADDR_ASCAST:%.*]] = addrspacecast ptr [[SRC_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[SHARED_ADDR_ASCAST:%.*]] = addrspacecast ptr [[SHARED_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[X_ASCAST:%.*]] = addrspacecast ptr [[X]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[SHARED_COERCE:%.*]], ptr addrspace(4) [[SHARED_ASCAST]], align 8
// CHECK-NEXT: [[SHARED1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[SHARED_ASCAST]], align 8
// CHECK-NEXT: store float [[SRC:%.*]], ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: store ptr addrspace(4) [[SHARED1]], ptr addrspace(4) [[SHARED_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[SHARED_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = addrspacecast ptr addrspace(4) [[TMP0]] to ptr addrspace(3)
// CHECK-NEXT: [[TMP2:%.*]] = load float, ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: [[TMP3:%.*]] = atomicrmw fmin ptr addrspace(3) [[TMP1]], float [[TMP2]] monotonic, align 4
// CHECK-NEXT: store volatile float [[TMP3]], ptr addrspace(4) [[X_ASCAST]], align 4
// CHECK-NEXT: ret void
//
__global__ void test_ds_fmin(float src, float *shared) {
volatile float x = __builtin_amdgcn_ds_fminf(shared, src, 0, 0, false);
}
#if 0 // FIXME: returning a pointer to AS4 explicitly is wrong for AMDGPU SPIRV
//
__device__ void test_ret_builtin_nondef_addrspace() {
void *x = __builtin_amdgcn_dispatch_ptr();
}
#endif
// CHECK-LABEL: @_Z6endpgmv(
// CHECK-NEXT: entry:
// CHECK-NEXT: call addrspace(4) void @llvm.amdgcn.endpgm()
// CHECK-NEXT: ret void
//
__global__ void endpgm() {
__builtin_amdgcn_endpgm();
}
// Check the 64 bit argument is correctly passed to the intrinsic without truncation or assertion.
// CHECK-LABEL: @_Z14test_uicmp_i64Pyyy(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[OUT:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[A_ADDR:%.*]] = alloca i64, align 8
// CHECK-NEXT: [[B_ADDR:%.*]] = alloca i64, align 8
// CHECK-NEXT: [[OUT_ASCAST:%.*]] = addrspacecast ptr [[OUT]] to ptr addrspace(4)
// CHECK-NEXT: [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr [[OUT_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[A_ADDR_ASCAST:%.*]] = addrspacecast ptr [[A_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[B_ADDR_ASCAST:%.*]] = addrspacecast ptr [[B_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[OUT_COERCE:%.*]], ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: [[OUT1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: store ptr addrspace(4) [[OUT1]], ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[A:%.*]], ptr addrspace(4) [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[B:%.*]], ptr addrspace(4) [[B_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load i64, ptr addrspace(4) [[A_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = load i64, ptr addrspace(4) [[B_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP2:%.*]] = call addrspace(4) i64 @llvm.amdgcn.icmp.i64.i64(i64 [[TMP0]], i64 [[TMP1]], i32 35)
// CHECK-NEXT: [[TMP3:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[TMP2]], ptr addrspace(4) [[TMP3]], align 8
// CHECK-NEXT: ret void
//
__global__ void test_uicmp_i64(unsigned long long *out, unsigned long long a, unsigned long long b)
{
*out = __builtin_amdgcn_uicmpl(a, b, 30+5);
}
// Check the 64 bit return value is correctly returned without truncation or assertion.
// CHECK-LABEL: @_Z14test_s_memtimePy(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[OUT:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[OUT_ASCAST:%.*]] = addrspacecast ptr [[OUT]] to ptr addrspace(4)
// CHECK-NEXT: [[OUT_ADDR_ASCAST:%.*]] = addrspacecast ptr [[OUT_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[OUT_COERCE:%.*]], ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: [[OUT1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ASCAST]], align 8
// CHECK-NEXT: store ptr addrspace(4) [[OUT1]], ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = call addrspace(4) i64 @llvm.amdgcn.s.memtime()
// CHECK-NEXT: [[TMP1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[OUT_ADDR_ASCAST]], align 8
// CHECK-NEXT: store i64 [[TMP0]], ptr addrspace(4) [[TMP1]], align 8
// CHECK-NEXT: ret void
//
__global__ void test_s_memtime(unsigned long long* out)
{
*out = __builtin_amdgcn_s_memtime();
}
// Check a generic pointer can be passed as a shared pointer and a generic pointer.
__device__ void func(float *x);
// CHECK-LABEL: @_Z17test_ds_fmin_funcfPf(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[SHARED:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[SRC_ADDR:%.*]] = alloca float, align 4
// CHECK-NEXT: [[SHARED_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[X:%.*]] = alloca float, align 4
// CHECK-NEXT: [[SHARED_ASCAST:%.*]] = addrspacecast ptr [[SHARED]] to ptr addrspace(4)
// CHECK-NEXT: [[SRC_ADDR_ASCAST:%.*]] = addrspacecast ptr [[SRC_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[SHARED_ADDR_ASCAST:%.*]] = addrspacecast ptr [[SHARED_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[X_ASCAST:%.*]] = addrspacecast ptr [[X]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[SHARED_COERCE:%.*]], ptr addrspace(4) [[SHARED_ASCAST]], align 8
// CHECK-NEXT: [[SHARED1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[SHARED_ASCAST]], align 8
// CHECK-NEXT: store float [[SRC:%.*]], ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: store ptr addrspace(4) [[SHARED1]], ptr addrspace(4) [[SHARED_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[SHARED_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = addrspacecast ptr addrspace(4) [[TMP0]] to ptr addrspace(3)
// CHECK-NEXT: [[TMP2:%.*]] = load float, ptr addrspace(4) [[SRC_ADDR_ASCAST]], align 4
// CHECK-NEXT: [[TMP3:%.*]] = atomicrmw fmin ptr addrspace(3) [[TMP1]], float [[TMP2]] monotonic, align 4
// CHECK-NEXT: store volatile float [[TMP3]], ptr addrspace(4) [[X_ASCAST]], align 4
// CHECK-NEXT: [[TMP4:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[SHARED_ADDR_ASCAST]], align 8
// CHECK-NEXT: call spir_func addrspace(4) void @_Z4funcPf(ptr addrspace(4) noundef [[TMP4]]) #[[ATTR6:[0-9]+]]
// CHECK-NEXT: ret void
//
__global__ void test_ds_fmin_func(float src, float *__restrict shared) {
volatile float x = __builtin_amdgcn_ds_fminf(shared, src, 0, 0, false);
func(shared);
}
// CHECK-LABEL: @_Z14test_is_sharedPf(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[X:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[X_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[RET:%.*]] = alloca i8, align 1
// CHECK-NEXT: [[X_ASCAST:%.*]] = addrspacecast ptr [[X]] to ptr addrspace(4)
// CHECK-NEXT: [[X_ADDR_ASCAST:%.*]] = addrspacecast ptr [[X_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[RET_ASCAST:%.*]] = addrspacecast ptr [[RET]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[X_COERCE:%.*]], ptr addrspace(4) [[X_ASCAST]], align 8
// CHECK-NEXT: [[X1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[X_ASCAST]], align 8
// CHECK-NEXT: store ptr addrspace(4) [[X1]], ptr addrspace(4) [[X_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[X_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = addrspacecast ptr addrspace(4) [[TMP0]] to ptr
// CHECK-NEXT: [[TMP2:%.*]] = call addrspace(4) i1 @llvm.amdgcn.is.shared(ptr [[TMP1]])
// CHECK-NEXT: [[STOREDV:%.*]] = zext i1 [[TMP2]] to i8
// CHECK-NEXT: store i8 [[STOREDV]], ptr addrspace(4) [[RET_ASCAST]], align 1
// CHECK-NEXT: ret void
//
__global__ void test_is_shared(float *x){
bool ret = __builtin_amdgcn_is_shared(x);
}
// CHECK-LABEL: @_Z15test_is_privatePi(
// CHECK-NEXT: entry:
// CHECK-NEXT: [[X:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[X_ADDR:%.*]] = alloca ptr addrspace(4), align 8
// CHECK-NEXT: [[RET:%.*]] = alloca i8, align 1
// CHECK-NEXT: [[X_ASCAST:%.*]] = addrspacecast ptr [[X]] to ptr addrspace(4)
// CHECK-NEXT: [[X_ADDR_ASCAST:%.*]] = addrspacecast ptr [[X_ADDR]] to ptr addrspace(4)
// CHECK-NEXT: [[RET_ASCAST:%.*]] = addrspacecast ptr [[RET]] to ptr addrspace(4)
// CHECK-NEXT: store ptr addrspace(1) [[X_COERCE:%.*]], ptr addrspace(4) [[X_ASCAST]], align 8
// CHECK-NEXT: [[X1:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[X_ASCAST]], align 8
// CHECK-NEXT: store ptr addrspace(4) [[X1]], ptr addrspace(4) [[X_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP0:%.*]] = load ptr addrspace(4), ptr addrspace(4) [[X_ADDR_ASCAST]], align 8
// CHECK-NEXT: [[TMP1:%.*]] = addrspacecast ptr addrspace(4) [[TMP0]] to ptr
// CHECK-NEXT: [[TMP2:%.*]] = call addrspace(4) i1 @llvm.amdgcn.is.private(ptr [[TMP1]])
// CHECK-NEXT: [[STOREDV:%.*]] = zext i1 [[TMP2]] to i8
// CHECK-NEXT: store i8 [[STOREDV]], ptr addrspace(4) [[RET_ASCAST]], align 1
// CHECK-NEXT: ret void
//
__global__ void test_is_private(int *x){
bool ret = __builtin_amdgcn_is_private(x);
}
|
