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
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
|
// REQUIRES: amdgpu-registered-target
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu tonga -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx900 -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1010 -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -triple amdgcn-unknown-unknown -target-cpu gfx1012 -emit-llvm -o - %s | FileCheck %s
// RUN: %clang_cc1 -triple spirv64-amd-amdhsa -emit-llvm -o - %s | FileCheck %s
#pragma OPENCL EXTENSION cl_khr_fp16 : enable
typedef unsigned long ulong;
typedef unsigned int uint;
// CHECK-LABEL: @test_div_fixup_f16
// CHECK: {{.*}}call{{.*}} half @llvm.amdgcn.div.fixup.f16
void test_div_fixup_f16(global half* out, half a, half b, half c)
{
*out = __builtin_amdgcn_div_fixuph(a, b, c);
}
// CHECK-LABEL: @test_rcp_f16
// CHECK: {{.*}}call{{.*}} half @llvm.amdgcn.rcp.f16
void test_rcp_f16(global half* out, half a)
{
*out = __builtin_amdgcn_rcph(a);
}
// CHECK-LABEL: @test_sqrt_f16
// CHECK: {{.*}}call{{.*}} half @llvm.{{((amdgcn.){0,1})}}sqrt.f16
void test_sqrt_f16(global half* out, half a)
{
*out = __builtin_amdgcn_sqrth(a);
}
// CHECK-LABEL: @test_rsq_f16
// CHECK: {{.*}}call{{.*}} half @llvm.amdgcn.rsq.f16
void test_rsq_f16(global half* out, half a)
{
*out = __builtin_amdgcn_rsqh(a);
}
// CHECK-LABEL: @test_sin_f16
// CHECK: {{.*}}call{{.*}} half @llvm.amdgcn.sin.f16
void test_sin_f16(global half* out, half a)
{
*out = __builtin_amdgcn_sinh(a);
}
// CHECK-LABEL: @test_cos_f16
// CHECK: {{.*}}call{{.*}} half @llvm.amdgcn.cos.f16
void test_cos_f16(global half* out, half a)
{
*out = __builtin_amdgcn_cosh(a);
}
// CHECK-LABEL: @test_ldexp_f16
// CHECK: [[TRUNC:%[0-9a-z]+]] = trunc i32
// CHECK: {{.*}}call{{.*}} half @llvm.ldexp.f16.i16(half %a, i16 [[TRUNC]])
void test_ldexp_f16(global half* out, half a, int b)
{
*out = __builtin_amdgcn_ldexph(a, b);
}
// CHECK-LABEL: @test_frexp_mant_f16
// CHECK: {{.*}}call{{.*}} half @llvm.amdgcn.frexp.mant.f16
void test_frexp_mant_f16(global half* out, half a)
{
*out = __builtin_amdgcn_frexp_manth(a);
}
// CHECK-LABEL: @test_frexp_exp_f16
// CHECK: {{.*}}call{{.*}} i16 @llvm.amdgcn.frexp.exp.i16.f16
void test_frexp_exp_f16(global short* out, half a)
{
*out = __builtin_amdgcn_frexp_exph(a);
}
// CHECK-LABEL: @test_fract_f16
// CHECK: {{.*}}call{{.*}} half @llvm.amdgcn.fract.f16
void test_fract_f16(global half* out, half a)
{
*out = __builtin_amdgcn_fracth(a);
}
// CHECK-LABEL: @test_class_f16
// CHECK: {{.*}}call{{.*}} i1 @llvm.amdgcn.class.f16
void test_class_f16(global half* out, half a, int b)
{
*out = __builtin_amdgcn_classh(a, b);
}
// CHECK-LABEL: @test_s_memrealtime
// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.s.memrealtime()
void test_s_memrealtime(global ulong* out)
{
*out = __builtin_amdgcn_s_memrealtime();
}
// CHECK-LABEL: @test_s_dcache_wb()
// CHECK: {{.*}}call{{.*}} void @llvm.amdgcn.s.dcache.wb()
void test_s_dcache_wb()
{
__builtin_amdgcn_s_dcache_wb();
}
// CHECK-LABEL: @test_mov_dpp_int
// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 poison, i32 %src, i32 0, i32 0, i32 0, i1 false)
void test_mov_dpp_int(global int* out, int src)
{
*out = __builtin_amdgcn_mov_dpp(src, 0, 0, 0, false);
}
// CHECK-LABEL: @test_mov_dpp_long
// CHECK: %0 = tail call{{.*}} i64 @llvm.amdgcn.update.dpp.i64(i64 poison, i64 %x, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: store i64 %0,
void test_mov_dpp_long(long x, global long *p) {
*p = __builtin_amdgcn_mov_dpp(x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_mov_dpp_float
// CHECK: %0 = bitcast float %x to i32
// CHECK-NEXT: %1 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 poison, i32 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: store i32 %1,
void test_mov_dpp_float(float x, global float *p) {
*p = __builtin_amdgcn_mov_dpp(x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_mov_dpp_double
// CHECK: %0 = bitcast double %x to i64
// CHECK-NEXT: %1 = tail call{{.*}} i64 @llvm.amdgcn.update.dpp.i64(i64 poison, i64 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: store i64 %1,
void test_mov_dpp_double(double x, global double *p) {
*p = __builtin_amdgcn_mov_dpp(x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_mov_dpp_short
// CHECK: %0 = zext i16 %x to i32
// CHECK-NEXT: %1 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 poison, i32 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: %2 = trunc i32 %1 to i16
// CHECK-NEXT: store i16 %2,
void test_mov_dpp_short(short x, global short *p) {
*p = __builtin_amdgcn_mov_dpp(x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_mov_dpp_char
// CHECK: %0 = zext i8 %x to i32
// CHECK-NEXT: %1 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 poison, i32 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: %2 = trunc i32 %1 to i8
// CHECK-NEXT: store i8 %2,
void test_mov_dpp_char(char x, global char *p) {
*p = __builtin_amdgcn_mov_dpp(x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_mov_dpp_half
// CHECK: %0 = load i16,
// CHECK: %1 = zext i16 %0 to i32
// CHECK-NEXT: %2 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 poison, i32 %1, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: %3 = trunc i32 %2 to i16
// CHECK-NEXT: store i16 %3,
void test_mov_dpp_half(half *x, global half *p) {
*p = __builtin_amdgcn_mov_dpp(*x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_update_dpp_int
// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 %arg1, i32 %arg2, i32 0, i32 0, i32 0, i1 false)
void test_update_dpp_int(global int* out, int arg1, int arg2)
{
*out = __builtin_amdgcn_update_dpp(arg1, arg2, 0, 0, 0, false);
}
// CHECK-LABEL: @test_update_dpp_long
// CHECK: %0 = tail call{{.*}} i64 @llvm.amdgcn.update.dpp.i64(i64 %x, i64 %x, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: store i64 %0,
void test_update_dpp_long(long x, global long *p) {
*p = __builtin_amdgcn_update_dpp(x, x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_update_dpp_float
// CHECK: %0 = bitcast float %x to i32
// CHECK-NEXT: %1 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 %0, i32 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: store i32 %1,
void test_update_dpp_float(float x, global float *p) {
*p = __builtin_amdgcn_update_dpp(x, x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_update_dpp_double
// CHECK: %0 = bitcast double %x to i64
// CHECK-NEXT: %1 = tail call{{.*}} i64 @llvm.amdgcn.update.dpp.i64(i64 %0, i64 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: store i64 %1,
void test_update_dpp_double(double x, global double *p) {
*p = __builtin_amdgcn_update_dpp(x, x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_update_dpp_short
// CHECK: %0 = zext i16 %x to i32
// CHECK-NEXT: %1 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 %0, i32 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: %2 = trunc i32 %1 to i16
// CHECK-NEXT: store i16 %2,
void test_update_dpp_short(short x, global short *p) {
*p = __builtin_amdgcn_update_dpp(x, x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_update_dpp_char
// CHECK: %0 = zext i8 %x to i32
// CHECK-NEXT: %1 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 %0, i32 %0, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: %2 = trunc i32 %1 to i8
// CHECK-NEXT: store i8 %2,
void test_update_dpp_char(char x, global char *p) {
*p = __builtin_amdgcn_update_dpp(x, x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_update_dpp_half
// CHECK: %0 = load i16,
// CHECK: %1 = zext i16 %0 to i32
// CHECK-NEXT: %2 = tail call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 %1, i32 %1, i32 257, i32 15, i32 15, i1 false)
// CHECK-NEXT: %3 = trunc i32 %2 to i16
// CHECK-NEXT: store i16 %3,
void test_update_dpp_half(half *x, global half *p) {
*p = __builtin_amdgcn_update_dpp(*x, *x, 0x101, 0xf, 0xf, 0);
}
// CHECK-LABEL: @test_update_dpp_int_uint
// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 %arg1, i32 %arg2, i32 0, i32 0, i32 0, i1 false)
void test_update_dpp_int_uint(global int* out, int arg1, unsigned int arg2)
{
*out = __builtin_amdgcn_update_dpp(arg1, arg2, 0, 0, 0, false);
}
// CHECK-LABEL: @test_update_dpp_lit_int
// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 5, i32 %arg1, i32 0, i32 0, i32 0, i1 false)
void test_update_dpp_lit_int(global int* out, int arg1)
{
*out = __builtin_amdgcn_update_dpp(5, arg1, 0, 0, 0, false);
}
__constant int gi = 5;
// CHECK-LABEL: @test_update_dpp_const_int
// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.update.dpp.i32(i32 5, i32 %arg1, i32 0, i32 0, i32 0, i1 false)
void test_update_dpp_const_int(global int* out, int arg1)
{
*out = __builtin_amdgcn_update_dpp(gi, arg1, 0, 0, 0, false);
}
// CHECK-LABEL: @test_ds_fadd
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
// CHECK: atomicrmw volatile fadd ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src acquire, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src acquire, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src release, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src acq_rel, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src seq_cst, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src seq_cst, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src syncscope("agent") monotonic, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src syncscope("workgroup") monotonic, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src syncscope("wavefront") monotonic, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src syncscope("singlethread") monotonic, align 4{{$}}
// CHECK: atomicrmw fadd ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
#if !defined(__SPIRV__)
void test_ds_faddf(local float *out, float src) {
#else
void test_ds_faddf(__attribute__((address_space(3))) float *out, float src) {
#endif
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_SYSTEM, true);
// Test all orders.
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_CONSUME, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_ACQUIRE, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELEASE, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_ACQ_REL, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_SEQ_CST, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_SEQ_CST, __MEMORY_SCOPE_SYSTEM, false); // invalid
// Test all syncscopes.
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_DEVICE, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_WRKGRP, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_WVFRNT, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_SINGLE, false);
*out = __builtin_amdgcn_ds_faddf(out, src, __ATOMIC_RELAXED, 5, false); // invalid
}
// CHECK-LABEL: @test_ds_fmin
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
// CHECK: atomicrmw volatile fmin ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src acquire, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src acquire, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src release, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src acq_rel, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src seq_cst, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src seq_cst, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src syncscope("agent") monotonic, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src syncscope("workgroup") monotonic, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src syncscope("wavefront") monotonic, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src syncscope("singlethread") monotonic, align 4{{$}}
// CHECK: atomicrmw fmin ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
#if !defined(__SPIRV__)
void test_ds_fminf(local float *out, float src) {
#else
void test_ds_fminf(__attribute__((address_space(3))) float *out, float src) {
#endif
*out = __builtin_amdgcn_ds_fminf(out, src, 0, 0, false);
*out = __builtin_amdgcn_ds_fminf(out, src, 0, 0, true);
// Test all orders.
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_CONSUME, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_ACQUIRE, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_RELEASE, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_ACQ_REL, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_SEQ_CST, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_SEQ_CST, __MEMORY_SCOPE_SYSTEM, false); // invalid
// Test all syncscopes.
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_DEVICE, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_WRKGRP, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_WVFRNT, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_SINGLE, false);
*out = __builtin_amdgcn_ds_fminf(out, src, __ATOMIC_RELAXED, 5, false); // invalid
}
// CHECK-LABEL: @test_ds_fmax
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
// CHECK: atomicrmw volatile fmax ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src acquire, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src acquire, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src release, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src acq_rel, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src seq_cst, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src seq_cst, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src syncscope("agent") monotonic, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src syncscope("workgroup") monotonic, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src syncscope("wavefront") monotonic, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src syncscope("singlethread") monotonic, align 4{{$}}
// CHECK: atomicrmw fmax ptr addrspace(3) %out, float %src monotonic, align 4{{$}}
#if !defined(__SPIRV__)
void test_ds_fmaxf(local float *out, float src) {
#else
void test_ds_fmaxf(__attribute__((address_space(3))) float *out, float src) {
#endif
*out = __builtin_amdgcn_ds_fmaxf(out, src, 0, 0, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, 0, 0, true);
// Test all orders.
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_CONSUME, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_ACQUIRE, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_RELEASE, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_ACQ_REL, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_SEQ_CST, __MEMORY_SCOPE_SYSTEM, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_SEQ_CST, __MEMORY_SCOPE_SYSTEM, false); // invalid
// Test all syncscopes.
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_DEVICE, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_WRKGRP, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_WVFRNT, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_RELAXED, __MEMORY_SCOPE_SINGLE, false);
*out = __builtin_amdgcn_ds_fmaxf(out, src, __ATOMIC_RELAXED, 5, false); // invalid
}
// CHECK-LABEL: @test_s_memtime
// CHECK: {{.*}}call{{.*}} i64 @llvm.amdgcn.s.memtime()
void test_s_memtime(global ulong* out)
{
*out = __builtin_amdgcn_s_memtime();
}
// CHECK-LABEL: @test_perm
// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.perm(i32 %a, i32 %b, i32 %s)
void test_perm(global uint* out, uint a, uint b, uint s)
{
*out = __builtin_amdgcn_perm(a, b, s);
}
// CHECK-LABEL: @test_groupstaticsize
// CHECK: {{.*}}call{{.*}} i32 @llvm.amdgcn.groupstaticsize()
void test_groupstaticsize(global uint* out)
{
*out = __builtin_amdgcn_groupstaticsize();
}
|
