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
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
| /*===--- __clang_cuda_intrinsics.h - Device-side CUDA intrinsic wrappers ---===
*
* Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
* See https://llvm.org/LICENSE.txt for license information.
* SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
*
*===-----------------------------------------------------------------------===
*/
#ifndef __CLANG_CUDA_INTRINSICS_H__
#define __CLANG_CUDA_INTRINSICS_H__
#ifndef __CUDA__
#error "This file is for CUDA compilation only."
#endif
// sm_30 intrinsics: __shfl_{up,down,xor}.
#define __SM_30_INTRINSICS_H__
#define __SM_30_INTRINSICS_HPP__
#if !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300
#pragma push_macro("__MAKE_SHUFFLES")
#define __MAKE_SHUFFLES(__FnName, __IntIntrinsic, __FloatIntrinsic, __Mask, \
__Type) \
inline __device__ int __FnName(int __val, __Type __offset, \
int __width = warpSize) { \
return __IntIntrinsic(__val, __offset, \
((warpSize - __width) << 8) | (__Mask)); \
} \
inline __device__ float __FnName(float __val, __Type __offset, \
int __width = warpSize) { \
return __FloatIntrinsic(__val, __offset, \
((warpSize - __width) << 8) | (__Mask)); \
} \
inline __device__ unsigned int __FnName(unsigned int __val, __Type __offset, \
int __width = warpSize) { \
return static_cast<unsigned int>( \
::__FnName(static_cast<int>(__val), __offset, __width)); \
} \
inline __device__ long long __FnName(long long __val, __Type __offset, \
int __width = warpSize) { \
struct __Bits { \
int __a, __b; \
}; \
_Static_assert(sizeof(__val) == sizeof(__Bits)); \
_Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \
__Bits __tmp; \
memcpy(&__val, &__tmp, sizeof(__val)); \
__tmp.__a = ::__FnName(__tmp.__a, __offset, __width); \
__tmp.__b = ::__FnName(__tmp.__b, __offset, __width); \
long long __ret; \
memcpy(&__ret, &__tmp, sizeof(__tmp)); \
return __ret; \
} \
inline __device__ long __FnName(long __val, __Type __offset, \
int __width = warpSize) { \
_Static_assert(sizeof(long) == sizeof(long long) || \
sizeof(long) == sizeof(int)); \
if (sizeof(long) == sizeof(long long)) { \
return static_cast<long>( \
::__FnName(static_cast<long long>(__val), __offset, __width)); \
} else if (sizeof(long) == sizeof(int)) { \
return static_cast<long>( \
::__FnName(static_cast<int>(__val), __offset, __width)); \
} \
} \
inline __device__ unsigned long __FnName( \
unsigned long __val, __Type __offset, int __width = warpSize) { \
return static_cast<unsigned long>( \
::__FnName(static_cast<long>(__val), __offset, __width)); \
} \
inline __device__ unsigned long long __FnName( \
unsigned long long __val, __Type __offset, int __width = warpSize) { \
return static_cast<unsigned long long>(::__FnName( \
static_cast<unsigned long long>(__val), __offset, __width)); \
} \
inline __device__ double __FnName(double __val, __Type __offset, \
int __width = warpSize) { \
long long __tmp; \
_Static_assert(sizeof(__tmp) == sizeof(__val)); \
memcpy(&__tmp, &__val, sizeof(__val)); \
__tmp = ::__FnName(__tmp, __offset, __width); \
double __ret; \
memcpy(&__ret, &__tmp, sizeof(__ret)); \
return __ret; \
}
__MAKE_SHUFFLES(__shfl, __nvvm_shfl_idx_i32, __nvvm_shfl_idx_f32, 0x1f, int);
// We use 0 rather than 31 as our mask, because shfl.up applies to lanes >=
// maxLane.
__MAKE_SHUFFLES(__shfl_up, __nvvm_shfl_up_i32, __nvvm_shfl_up_f32, 0,
unsigned int);
__MAKE_SHUFFLES(__shfl_down, __nvvm_shfl_down_i32, __nvvm_shfl_down_f32, 0x1f,
unsigned int);
__MAKE_SHUFFLES(__shfl_xor, __nvvm_shfl_bfly_i32, __nvvm_shfl_bfly_f32, 0x1f,
int);
#pragma pop_macro("__MAKE_SHUFFLES")
#endif // !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300
#if CUDA_VERSION >= 9000
#if (!defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300)
// __shfl_sync_* variants available in CUDA-9
#pragma push_macro("__MAKE_SYNC_SHUFFLES")
#define __MAKE_SYNC_SHUFFLES(__FnName, __IntIntrinsic, __FloatIntrinsic, \
__Mask, __Type) \
inline __device__ int __FnName(unsigned int __mask, int __val, \
__Type __offset, int __width = warpSize) { \
return __IntIntrinsic(__mask, __val, __offset, \
((warpSize - __width) << 8) | (__Mask)); \
} \
inline __device__ float __FnName(unsigned int __mask, float __val, \
__Type __offset, int __width = warpSize) { \
return __FloatIntrinsic(__mask, __val, __offset, \
((warpSize - __width) << 8) | (__Mask)); \
} \
inline __device__ unsigned int __FnName(unsigned int __mask, \
unsigned int __val, __Type __offset, \
int __width = warpSize) { \
return static_cast<unsigned int>( \
::__FnName(__mask, static_cast<int>(__val), __offset, __width)); \
} \
inline __device__ long long __FnName(unsigned int __mask, long long __val, \
__Type __offset, \
int __width = warpSize) { \
struct __Bits { \
int __a, __b; \
}; \
_Static_assert(sizeof(__val) == sizeof(__Bits)); \
_Static_assert(sizeof(__Bits) == 2 * sizeof(int)); \
__Bits __tmp; \
memcpy(&__val, &__tmp, sizeof(__val)); \
__tmp.__a = ::__FnName(__mask, __tmp.__a, __offset, __width); \
__tmp.__b = ::__FnName(__mask, __tmp.__b, __offset, __width); \
long long __ret; \
memcpy(&__ret, &__tmp, sizeof(__tmp)); \
return __ret; \
} \
inline __device__ unsigned long long __FnName( \
unsigned int __mask, unsigned long long __val, __Type __offset, \
int __width = warpSize) { \
return static_cast<unsigned long long>(::__FnName( \
__mask, static_cast<unsigned long long>(__val), __offset, __width)); \
} \
inline __device__ long __FnName(unsigned int __mask, long __val, \
__Type __offset, int __width = warpSize) { \
_Static_assert(sizeof(long) == sizeof(long long) || \
sizeof(long) == sizeof(int)); \
if (sizeof(long) == sizeof(long long)) { \
return static_cast<long>(::__FnName( \
__mask, static_cast<long long>(__val), __offset, __width)); \
} else if (sizeof(long) == sizeof(int)) { \
return static_cast<long>( \
::__FnName(__mask, static_cast<int>(__val), __offset, __width)); \
} \
} \
inline __device__ unsigned long __FnName( \
unsigned int __mask, unsigned long __val, __Type __offset, \
int __width = warpSize) { \
return static_cast<unsigned long>( \
::__FnName(__mask, static_cast<long>(__val), __offset, __width)); \
} \
inline __device__ double __FnName(unsigned int __mask, double __val, \
__Type __offset, int __width = warpSize) { \
long long __tmp; \
_Static_assert(sizeof(__tmp) == sizeof(__val)); \
memcpy(&__tmp, &__val, sizeof(__val)); \
__tmp = ::__FnName(__mask, __tmp, __offset, __width); \
double __ret; \
memcpy(&__ret, &__tmp, sizeof(__ret)); \
return __ret; \
}
__MAKE_SYNC_SHUFFLES(__shfl_sync, __nvvm_shfl_sync_idx_i32,
__nvvm_shfl_sync_idx_f32, 0x1f, int);
// We use 0 rather than 31 as our mask, because shfl.up applies to lanes >=
// maxLane.
__MAKE_SYNC_SHUFFLES(__shfl_up_sync, __nvvm_shfl_sync_up_i32,
__nvvm_shfl_sync_up_f32, 0, unsigned int);
__MAKE_SYNC_SHUFFLES(__shfl_down_sync, __nvvm_shfl_sync_down_i32,
__nvvm_shfl_sync_down_f32, 0x1f, unsigned int);
__MAKE_SYNC_SHUFFLES(__shfl_xor_sync, __nvvm_shfl_sync_bfly_i32,
__nvvm_shfl_sync_bfly_f32, 0x1f, int);
#pragma pop_macro("__MAKE_SYNC_SHUFFLES")
inline __device__ void __syncwarp(unsigned int mask = 0xffffffff) {
return __nvvm_bar_warp_sync(mask);
}
inline __device__ void __barrier_sync(unsigned int id) {
__nvvm_barrier_sync(id);
}
inline __device__ void __barrier_sync_count(unsigned int id,
unsigned int count) {
__nvvm_barrier_sync_cnt(id, count);
}
inline __device__ int __all_sync(unsigned int mask, int pred) {
return __nvvm_vote_all_sync(mask, pred);
}
inline __device__ int __any_sync(unsigned int mask, int pred) {
return __nvvm_vote_any_sync(mask, pred);
}
inline __device__ int __uni_sync(unsigned int mask, int pred) {
return __nvvm_vote_uni_sync(mask, pred);
}
inline __device__ unsigned int __ballot_sync(unsigned int mask, int pred) {
return __nvvm_vote_ballot_sync(mask, pred);
}
inline __device__ unsigned int __activemask() {
#if CUDA_VERSION < 9020
return __nvvm_vote_ballot(1);
#else
unsigned int mask;
asm volatile("activemask.b32 %0;" : "=r"(mask));
return mask;
#endif
}
inline __device__ unsigned int __fns(unsigned mask, unsigned base, int offset) {
return __nvvm_fns(mask, base, offset);
}
#endif // !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 300
// Define __match* builtins CUDA-9 headers expect to see.
#if !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 700
inline __device__ unsigned int __match32_any_sync(unsigned int mask,
unsigned int value) {
return __nvvm_match_any_sync_i32(mask, value);
}
inline __device__ unsigned long long
__match64_any_sync(unsigned int mask, unsigned long long value) {
return __nvvm_match_any_sync_i64(mask, value);
}
inline __device__ unsigned int
__match32_all_sync(unsigned int mask, unsigned int value, int *pred) {
return __nvvm_match_all_sync_i32p(mask, value, pred);
}
inline __device__ unsigned long long
__match64_all_sync(unsigned int mask, unsigned long long value, int *pred) {
return __nvvm_match_all_sync_i64p(mask, value, pred);
}
#include "crt/sm_70_rt.hpp"
#endif // !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 700
#endif // __CUDA_VERSION >= 9000
// sm_32 intrinsics: __ldg and __funnelshift_{l,lc,r,rc}.
// Prevent the vanilla sm_32 intrinsics header from being included.
#define __SM_32_INTRINSICS_H__
#define __SM_32_INTRINSICS_HPP__
#if !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 320
inline __device__ char __ldg(const char *ptr) { return __nvvm_ldg_c(ptr); }
inline __device__ short __ldg(const short *ptr) { return __nvvm_ldg_s(ptr); }
inline __device__ int __ldg(const int *ptr) { return __nvvm_ldg_i(ptr); }
inline __device__ long __ldg(const long *ptr) { return __nvvm_ldg_l(ptr); }
inline __device__ long long __ldg(const long long *ptr) {
return __nvvm_ldg_ll(ptr);
}
inline __device__ unsigned char __ldg(const unsigned char *ptr) {
return __nvvm_ldg_uc(ptr);
}
inline __device__ signed char __ldg(const signed char *ptr) {
return __nvvm_ldg_uc((const unsigned char *)ptr);
}
inline __device__ unsigned short __ldg(const unsigned short *ptr) {
return __nvvm_ldg_us(ptr);
}
inline __device__ unsigned int __ldg(const unsigned int *ptr) {
return __nvvm_ldg_ui(ptr);
}
inline __device__ unsigned long __ldg(const unsigned long *ptr) {
return __nvvm_ldg_ul(ptr);
}
inline __device__ unsigned long long __ldg(const unsigned long long *ptr) {
return __nvvm_ldg_ull(ptr);
}
inline __device__ float __ldg(const float *ptr) { return __nvvm_ldg_f(ptr); }
inline __device__ double __ldg(const double *ptr) { return __nvvm_ldg_d(ptr); }
inline __device__ char2 __ldg(const char2 *ptr) {
typedef char c2 __attribute__((ext_vector_type(2)));
// We can assume that ptr is aligned at least to char2's alignment, but the
// load will assume that ptr is aligned to char2's alignment. This is only
// safe if alignof(c2) <= alignof(char2).
c2 rv = __nvvm_ldg_c2(reinterpret_cast<const c2 *>(ptr));
char2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ char4 __ldg(const char4 *ptr) {
typedef char c4 __attribute__((ext_vector_type(4)));
c4 rv = __nvvm_ldg_c4(reinterpret_cast<const c4 *>(ptr));
char4 ret;
ret.x = rv[0];
ret.y = rv[1];
ret.z = rv[2];
ret.w = rv[3];
return ret;
}
inline __device__ short2 __ldg(const short2 *ptr) {
typedef short s2 __attribute__((ext_vector_type(2)));
s2 rv = __nvvm_ldg_s2(reinterpret_cast<const s2 *>(ptr));
short2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ short4 __ldg(const short4 *ptr) {
typedef short s4 __attribute__((ext_vector_type(4)));
s4 rv = __nvvm_ldg_s4(reinterpret_cast<const s4 *>(ptr));
short4 ret;
ret.x = rv[0];
ret.y = rv[1];
ret.z = rv[2];
ret.w = rv[3];
return ret;
}
inline __device__ int2 __ldg(const int2 *ptr) {
typedef int i2 __attribute__((ext_vector_type(2)));
i2 rv = __nvvm_ldg_i2(reinterpret_cast<const i2 *>(ptr));
int2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ int4 __ldg(const int4 *ptr) {
typedef int i4 __attribute__((ext_vector_type(4)));
i4 rv = __nvvm_ldg_i4(reinterpret_cast<const i4 *>(ptr));
int4 ret;
ret.x = rv[0];
ret.y = rv[1];
ret.z = rv[2];
ret.w = rv[3];
return ret;
}
inline __device__ longlong2 __ldg(const longlong2 *ptr) {
typedef long long ll2 __attribute__((ext_vector_type(2)));
ll2 rv = __nvvm_ldg_ll2(reinterpret_cast<const ll2 *>(ptr));
longlong2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ uchar2 __ldg(const uchar2 *ptr) {
typedef unsigned char uc2 __attribute__((ext_vector_type(2)));
uc2 rv = __nvvm_ldg_uc2(reinterpret_cast<const uc2 *>(ptr));
uchar2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ uchar4 __ldg(const uchar4 *ptr) {
typedef unsigned char uc4 __attribute__((ext_vector_type(4)));
uc4 rv = __nvvm_ldg_uc4(reinterpret_cast<const uc4 *>(ptr));
uchar4 ret;
ret.x = rv[0];
ret.y = rv[1];
ret.z = rv[2];
ret.w = rv[3];
return ret;
}
inline __device__ ushort2 __ldg(const ushort2 *ptr) {
typedef unsigned short us2 __attribute__((ext_vector_type(2)));
us2 rv = __nvvm_ldg_us2(reinterpret_cast<const us2 *>(ptr));
ushort2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ ushort4 __ldg(const ushort4 *ptr) {
typedef unsigned short us4 __attribute__((ext_vector_type(4)));
us4 rv = __nvvm_ldg_us4(reinterpret_cast<const us4 *>(ptr));
ushort4 ret;
ret.x = rv[0];
ret.y = rv[1];
ret.z = rv[2];
ret.w = rv[3];
return ret;
}
inline __device__ uint2 __ldg(const uint2 *ptr) {
typedef unsigned int ui2 __attribute__((ext_vector_type(2)));
ui2 rv = __nvvm_ldg_ui2(reinterpret_cast<const ui2 *>(ptr));
uint2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ uint4 __ldg(const uint4 *ptr) {
typedef unsigned int ui4 __attribute__((ext_vector_type(4)));
ui4 rv = __nvvm_ldg_ui4(reinterpret_cast<const ui4 *>(ptr));
uint4 ret;
ret.x = rv[0];
ret.y = rv[1];
ret.z = rv[2];
ret.w = rv[3];
return ret;
}
inline __device__ ulonglong2 __ldg(const ulonglong2 *ptr) {
typedef unsigned long long ull2 __attribute__((ext_vector_type(2)));
ull2 rv = __nvvm_ldg_ull2(reinterpret_cast<const ull2 *>(ptr));
ulonglong2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ float2 __ldg(const float2 *ptr) {
typedef float f2 __attribute__((ext_vector_type(2)));
f2 rv = __nvvm_ldg_f2(reinterpret_cast<const f2 *>(ptr));
float2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
inline __device__ float4 __ldg(const float4 *ptr) {
typedef float f4 __attribute__((ext_vector_type(4)));
f4 rv = __nvvm_ldg_f4(reinterpret_cast<const f4 *>(ptr));
float4 ret;
ret.x = rv[0];
ret.y = rv[1];
ret.z = rv[2];
ret.w = rv[3];
return ret;
}
inline __device__ double2 __ldg(const double2 *ptr) {
typedef double d2 __attribute__((ext_vector_type(2)));
d2 rv = __nvvm_ldg_d2(reinterpret_cast<const d2 *>(ptr));
double2 ret;
ret.x = rv[0];
ret.y = rv[1];
return ret;
}
// TODO: Implement these as intrinsics, so the backend can work its magic on
// these. Alternatively, we could implement these as plain C and try to get
// llvm to recognize the relevant patterns.
inline __device__ unsigned __funnelshift_l(unsigned low32, unsigned high32,
unsigned shiftWidth) {
unsigned result;
asm("shf.l.wrap.b32 %0, %1, %2, %3;"
: "=r"(result)
: "r"(low32), "r"(high32), "r"(shiftWidth));
return result;
}
inline __device__ unsigned __funnelshift_lc(unsigned low32, unsigned high32,
unsigned shiftWidth) {
unsigned result;
asm("shf.l.clamp.b32 %0, %1, %2, %3;"
: "=r"(result)
: "r"(low32), "r"(high32), "r"(shiftWidth));
return result;
}
inline __device__ unsigned __funnelshift_r(unsigned low32, unsigned high32,
unsigned shiftWidth) {
unsigned result;
asm("shf.r.wrap.b32 %0, %1, %2, %3;"
: "=r"(result)
: "r"(low32), "r"(high32), "r"(shiftWidth));
return result;
}
inline __device__ unsigned __funnelshift_rc(unsigned low32, unsigned high32,
unsigned shiftWidth) {
unsigned ret;
asm("shf.r.clamp.b32 %0, %1, %2, %3;"
: "=r"(ret)
: "r"(low32), "r"(high32), "r"(shiftWidth));
return ret;
}
#endif // !defined(__CUDA_ARCH__) || __CUDA_ARCH__ >= 320
#endif // defined(__CLANG_CUDA_INTRINSICS_H__)
|