Blame view
tools/cub-1.8.0/cub/device/device_radix_sort.cuh
41.4 KB
8dcb6dfcb
first commit
|
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 |
/******************************************************************************
* Copyright (c) 2011, Duane Merrill. All rights reserved.
* Copyright (c) 2011-2018, NVIDIA CORPORATION. All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions are met:
* * Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* * Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* * Neither the name of the NVIDIA CORPORATION nor the
* names of its contributors may be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED
* WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
* DISCLAIMED. IN NO EVENT SHALL NVIDIA CORPORATION BE LIABLE FOR ANY
* DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
* (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
* LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND
* ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS
* SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
******************************************************************************/
/**
* \file
* cub::DeviceRadixSort provides device-wide, parallel operations for computing a radix sort across a sequence of data items residing within device-accessible memory.
*/
#pragma once
#include <stdio.h>
#include <iterator>
#include "dispatch/dispatch_radix_sort.cuh"
#include "../util_arch.cuh"
#include "../util_namespace.cuh"
/// Optional outer namespace(s)
CUB_NS_PREFIX
/// CUB namespace
namespace cub {
/**
* \brief DeviceRadixSort provides device-wide, parallel operations for computing a radix sort across a sequence of data items residing within device-accessible memory. 
* \ingroup SingleModule
*
* \par Overview
* The [<em>radix sorting method</em>](http://en.wikipedia.org/wiki/Radix_sort) arranges
* items into ascending (or descending) order. The algorithm relies upon a positional representation for
* keys, i.e., each key is comprised of an ordered sequence of symbols (e.g., digits,
* characters, etc.) specified from least-significant to most-significant. For a
* given input sequence of keys and a set of rules specifying a total ordering
* of the symbolic alphabet, the radix sorting method produces a lexicographic
* ordering of those keys.
*
* \par
* DeviceRadixSort can sort all of the built-in C++ numeric primitive types
* (<tt>unsigned char</tt>, \p int, \p double, etc.) as well as CUDA's \p __half
* half-precision floating-point type. Although the direct radix sorting
* method can only be applied to unsigned integral types, DeviceRadixSort
* is able to sort signed and floating-point types via simple bit-wise transformations
* that ensure lexicographic key ordering.
*
* \par Usage Considerations
* \cdp_class{DeviceRadixSort}
*
* \par Performance
* \linear_performance{radix sort} The following chart illustrates DeviceRadixSort::SortKeys
* performance across different CUDA architectures for uniform-random \p uint32 keys.
* \plots_below
*
* \image html lsb_radix_sort_int32_keys.png
*
*/
struct DeviceRadixSort
{
/******************************************************************//**
*
ame KeyT-value pairs
*********************************************************************/
//@{
/**
* \brief Sorts key-value pairs into ascending order. (~<em>2N </em>auxiliary storage required)
*
* \par
* - The contents of the input data are not altered by the sorting operation
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageNP For sorting using only <em>O</em>(<tt>P</tt>) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
* - \devicestorage
*
* \par Performance
* The following charts illustrate saturated sorting performance across different
* CUDA architectures for uniform-random <tt>uint32,uint32</tt> and
* <tt>uint64,uint64</tt> pairs, respectively.
*
* \image html lsb_radix_sort_int32_pairs.png
* \image html lsb_radix_sort_int64_pairs.png
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys
* with associated vector of \p int values.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_keys_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_keys_out; // e.g., [ ... ]
* int *d_values_in; // e.g., [0, 1, 2, 3, 4, 5, 6]
* int *d_values_out; // e.g., [ ... ]
* ...
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortPairs(d_temp_storage, temp_storage_bytes,
* d_keys_in, d_keys_out, d_values_in, d_values_out, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortPairs(d_temp_storage, temp_storage_bytes,
* d_keys_in, d_keys_out, d_values_in, d_values_out, num_items);
*
* // d_keys_out <-- [0, 3, 5, 6, 7, 8, 9]
* // d_values_out <-- [5, 4, 3, 1, 2, 0, 6]
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
* \tparam ValueT <b>[inferred]</b> ValueT type
*/
template <
typename KeyT,
typename ValueT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortPairs(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
const KeyT *d_keys_in, ///< [in] Pointer to the input data of key data to sort
KeyT *d_keys_out, ///< [out] Pointer to the sorted output sequence of key data
const ValueT *d_values_in, ///< [in] Pointer to the corresponding input sequence of associated value items
ValueT *d_values_out, ///< [out] Pointer to the correspondingly-reordered output sequence of associated value items
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
DoubleBuffer<KeyT> d_keys(const_cast<KeyT*>(d_keys_in), d_keys_out);
DoubleBuffer<ValueT> d_values(const_cast<ValueT*>(d_values_in), d_values_out);
return DispatchRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
false,
stream,
debug_synchronous);
}
/**
* \brief Sorts key-value pairs into ascending order. (~<em>N </em>auxiliary storage required)
*
* \par
* - The sorting operation is given a pair of key buffers and a corresponding
* pair of associated value buffers. Each pair is managed by a DoubleBuffer
* structure that indicates which of the two buffers is "current" (and thus
* contains the input data to be sorted).
* - The contents of both buffers within each pair may be altered by the sorting
* operation.
* - Upon completion, the sorting operation will update the "current" indicator
* within each DoubleBuffer wrapper to reference which of the two buffers
* now contains the sorted output sequence (a function of the number of key bits
* specified and the targeted device architecture).
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageP
* - \devicestorage
*
* \par Performance
* The following charts illustrate saturated sorting performance across different
* CUDA architectures for uniform-random <tt>uint32,uint32</tt> and
* <tt>uint64,uint64</tt> pairs, respectively.
*
* \image html lsb_radix_sort_int32_pairs.png
* \image html lsb_radix_sort_int64_pairs.png
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys
* with associated vector of \p int values.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_key_buf; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_key_alt_buf; // e.g., [ ... ]
* int *d_value_buf; // e.g., [0, 1, 2, 3, 4, 5, 6]
* int *d_value_alt_buf; // e.g., [ ... ]
* ...
*
* // Create a set of DoubleBuffers to wrap pairs of device pointers
* cub::DoubleBuffer<int> d_keys(d_key_buf, d_key_alt_buf);
* cub::DoubleBuffer<int> d_values(d_value_buf, d_value_alt_buf);
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortPairs(d_temp_storage, temp_storage_bytes, d_keys, d_values, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortPairs(d_temp_storage, temp_storage_bytes, d_keys, d_values, num_items);
*
* // d_keys.Current() <-- [0, 3, 5, 6, 7, 8, 9]
* // d_values.Current() <-- [5, 4, 3, 1, 2, 0, 6]
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
* \tparam ValueT <b>[inferred]</b> ValueT type
*/
template <
typename KeyT,
typename ValueT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortPairs(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
DoubleBuffer<KeyT> &d_keys, ///< [in,out] Reference to the double-buffer of keys whose "current" device-accessible buffer contains the unsorted input keys and, upon return, is updated to point to the sorted output keys
DoubleBuffer<ValueT> &d_values, ///< [in,out] Double-buffer of values whose "current" device-accessible buffer contains the unsorted input values and, upon return, is updated to point to the sorted output values
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
return DispatchRadixSort<false, KeyT, ValueT, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
true,
stream,
debug_synchronous);
}
/**
* \brief Sorts key-value pairs into descending order. (~<em>2N</em> auxiliary storage required).
*
* \par
* - The contents of the input data are not altered by the sorting operation
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageNP For sorting using only <em>O</em>(<tt>P</tt>) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
* - \devicestorage
*
* \par Performance
* Performance is similar to DeviceRadixSort::SortPairs.
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys
* with associated vector of \p int values.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_keys_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_keys_out; // e.g., [ ... ]
* int *d_values_in; // e.g., [0, 1, 2, 3, 4, 5, 6]
* int *d_values_out; // e.g., [ ... ]
* ...
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortPairsDescending(d_temp_storage, temp_storage_bytes,
* d_keys_in, d_keys_out, d_values_in, d_values_out, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortPairsDescending(d_temp_storage, temp_storage_bytes,
* d_keys_in, d_keys_out, d_values_in, d_values_out, num_items);
*
* // d_keys_out <-- [9, 8, 7, 6, 5, 3, 0]
* // d_values_out <-- [6, 0, 2, 1, 3, 4, 5]
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
* \tparam ValueT <b>[inferred]</b> ValueT type
*/
template <
typename KeyT,
typename ValueT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortPairsDescending(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
const KeyT *d_keys_in, ///< [in] Pointer to the input data of key data to sort
KeyT *d_keys_out, ///< [out] Pointer to the sorted output sequence of key data
const ValueT *d_values_in, ///< [in] Pointer to the corresponding input sequence of associated value items
ValueT *d_values_out, ///< [out] Pointer to the correspondingly-reordered output sequence of associated value items
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
DoubleBuffer<KeyT> d_keys(const_cast<KeyT*>(d_keys_in), d_keys_out);
DoubleBuffer<ValueT> d_values(const_cast<ValueT*>(d_values_in), d_values_out);
return DispatchRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
false,
stream,
debug_synchronous);
}
/**
* \brief Sorts key-value pairs into descending order. (~<em>N </em>auxiliary storage required).
*
* \par
* - The sorting operation is given a pair of key buffers and a corresponding
* pair of associated value buffers. Each pair is managed by a DoubleBuffer
* structure that indicates which of the two buffers is "current" (and thus
* contains the input data to be sorted).
* - The contents of both buffers within each pair may be altered by the sorting
* operation.
* - Upon completion, the sorting operation will update the "current" indicator
* within each DoubleBuffer wrapper to reference which of the two buffers
* now contains the sorted output sequence (a function of the number of key bits
* specified and the targeted device architecture).
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageP
* - \devicestorage
*
* \par Performance
* Performance is similar to DeviceRadixSort::SortPairs.
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys
* with associated vector of \p int values.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_key_buf; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_key_alt_buf; // e.g., [ ... ]
* int *d_value_buf; // e.g., [0, 1, 2, 3, 4, 5, 6]
* int *d_value_alt_buf; // e.g., [ ... ]
* ...
*
* // Create a set of DoubleBuffers to wrap pairs of device pointers
* cub::DoubleBuffer<int> d_keys(d_key_buf, d_key_alt_buf);
* cub::DoubleBuffer<int> d_values(d_value_buf, d_value_alt_buf);
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortPairsDescending(d_temp_storage, temp_storage_bytes, d_keys, d_values, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortPairsDescending(d_temp_storage, temp_storage_bytes, d_keys, d_values, num_items);
*
* // d_keys.Current() <-- [9, 8, 7, 6, 5, 3, 0]
* // d_values.Current() <-- [6, 0, 2, 1, 3, 4, 5]
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
* \tparam ValueT <b>[inferred]</b> ValueT type
*/
template <
typename KeyT,
typename ValueT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortPairsDescending(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
DoubleBuffer<KeyT> &d_keys, ///< [in,out] Reference to the double-buffer of keys whose "current" device-accessible buffer contains the unsorted input keys and, upon return, is updated to point to the sorted output keys
DoubleBuffer<ValueT> &d_values, ///< [in,out] Double-buffer of values whose "current" device-accessible buffer contains the unsorted input values and, upon return, is updated to point to the sorted output values
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
return DispatchRadixSort<true, KeyT, ValueT, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
true,
stream,
debug_synchronous);
}
//@} end member group
/******************************************************************//**
*
ame Keys-only
*********************************************************************/
//@{
/**
* \brief Sorts keys into ascending order. (~<em>2N </em>auxiliary storage required)
*
* \par
* - The contents of the input data are not altered by the sorting operation
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageNP For sorting using only <em>O</em>(<tt>P</tt>) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
* - \devicestorage
*
* \par Performance
* The following charts illustrate saturated sorting performance across different
* CUDA architectures for uniform-random \p uint32 and \p uint64 keys, respectively.
*
* \image html lsb_radix_sort_int32_keys.png
* \image html lsb_radix_sort_int64_keys.png
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_keys_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_keys_out; // e.g., [ ... ]
* ...
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortKeys(d_temp_storage, temp_storage_bytes, d_keys_in, d_keys_out, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortKeys(d_temp_storage, temp_storage_bytes, d_keys_in, d_keys_out, num_items);
*
* // d_keys_out <-- [0, 3, 5, 6, 7, 8, 9]
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
*/
template <typename KeyT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortKeys(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
const KeyT *d_keys_in, ///< [in] Pointer to the input data of key data to sort
KeyT *d_keys_out, ///< [out] Pointer to the sorted output sequence of key data
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
// Null value type
DoubleBuffer<KeyT> d_keys(const_cast<KeyT*>(d_keys_in), d_keys_out);
DoubleBuffer<NullType> d_values;
return DispatchRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
false,
stream,
debug_synchronous);
}
/**
* \brief Sorts keys into ascending order. (~<em>N </em>auxiliary storage required).
*
* \par
* - The sorting operation is given a pair of key buffers managed by a
* DoubleBuffer structure that indicates which of the two buffers is
* "current" (and thus contains the input data to be sorted).
* - The contents of both buffers may be altered by the sorting operation.
* - Upon completion, the sorting operation will update the "current" indicator
* within the DoubleBuffer wrapper to reference which of the two buffers
* now contains the sorted output sequence (a function of the number of key bits
* specified and the targeted device architecture).
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageP
* - \devicestorage
*
* \par Performance
* The following charts illustrate saturated sorting performance across different
* CUDA architectures for uniform-random \p uint32 and \p uint64 keys, respectively.
*
* \image html lsb_radix_sort_int32_keys.png
* \image html lsb_radix_sort_int64_keys.png
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_key_buf; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_key_alt_buf; // e.g., [ ... ]
* ...
*
* // Create a DoubleBuffer to wrap the pair of device pointers
* cub::DoubleBuffer<int> d_keys(d_key_buf, d_key_alt_buf);
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortKeys(d_temp_storage, temp_storage_bytes, d_keys, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortKeys(d_temp_storage, temp_storage_bytes, d_keys, num_items);
*
* // d_keys.Current() <-- [0, 3, 5, 6, 7, 8, 9]
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
*/
template <typename KeyT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortKeys(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
DoubleBuffer<KeyT> &d_keys, ///< [in,out] Reference to the double-buffer of keys whose "current" device-accessible buffer contains the unsorted input keys and, upon return, is updated to point to the sorted output keys
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
// Null value type
DoubleBuffer<NullType> d_values;
return DispatchRadixSort<false, KeyT, NullType, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
true,
stream,
debug_synchronous);
}
/**
* \brief Sorts keys into descending order. (~<em>2N</em> auxiliary storage required).
*
* \par
* - The contents of the input data are not altered by the sorting operation
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageNP For sorting using only <em>O</em>(<tt>P</tt>) temporary storage, see the sorting interface using DoubleBuffer wrappers below.
* - \devicestorage
*
* \par Performance
* Performance is similar to DeviceRadixSort::SortKeys.
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_keys_in; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_keys_out; // e.g., [ ... ]
* ...
*
* // Create a DoubleBuffer to wrap the pair of device pointers
* cub::DoubleBuffer<int> d_keys(d_key_buf, d_key_alt_buf);
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortKeysDescending(d_temp_storage, temp_storage_bytes, d_keys_in, d_keys_out, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortKeysDescending(d_temp_storage, temp_storage_bytes, d_keys_in, d_keys_out, num_items);
*
* // d_keys_out <-- [9, 8, 7, 6, 5, 3, 0]s
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
*/
template <typename KeyT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortKeysDescending(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
const KeyT *d_keys_in, ///< [in] Pointer to the input data of key data to sort
KeyT *d_keys_out, ///< [out] Pointer to the sorted output sequence of key data
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
DoubleBuffer<KeyT> d_keys(const_cast<KeyT*>(d_keys_in), d_keys_out);
DoubleBuffer<NullType> d_values;
return DispatchRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
false,
stream,
debug_synchronous);
}
/**
* \brief Sorts keys into descending order. (~<em>N </em>auxiliary storage required).
*
* \par
* - The sorting operation is given a pair of key buffers managed by a
* DoubleBuffer structure that indicates which of the two buffers is
* "current" (and thus contains the input data to be sorted).
* - The contents of both buffers may be altered by the sorting operation.
* - Upon completion, the sorting operation will update the "current" indicator
* within the DoubleBuffer wrapper to reference which of the two buffers
* now contains the sorted output sequence (a function of the number of key bits
* specified and the targeted device architecture).
* - An optional bit subrange <tt>[begin_bit, end_bit)</tt> of differentiating key bits can be specified. This can reduce overall sorting overhead and yield a corresponding performance improvement.
* - \devicestorageP
* - \devicestorage
*
* \par Performance
* Performance is similar to DeviceRadixSort::SortKeys.
*
* \par Snippet
* The code snippet below illustrates the sorting of a device vector of \p int keys.
* \par
* \code
* #include <cub/cub.cuh> // or equivalently <cub/device/device_radix_sort.cuh>
*
* // Declare, allocate, and initialize device-accessible pointers for sorting data
* int num_items; // e.g., 7
* int *d_key_buf; // e.g., [8, 6, 7, 5, 3, 0, 9]
* int *d_key_alt_buf; // e.g., [ ... ]
* ...
*
* // Create a DoubleBuffer to wrap the pair of device pointers
* cub::DoubleBuffer<int> d_keys(d_key_buf, d_key_alt_buf);
*
* // Determine temporary device storage requirements
* void *d_temp_storage = NULL;
* size_t temp_storage_bytes = 0;
* cub::DeviceRadixSort::SortKeysDescending(d_temp_storage, temp_storage_bytes, d_keys, num_items);
*
* // Allocate temporary storage
* cudaMalloc(&d_temp_storage, temp_storage_bytes);
*
* // Run sorting operation
* cub::DeviceRadixSort::SortKeysDescending(d_temp_storage, temp_storage_bytes, d_keys, num_items);
*
* // d_keys.Current() <-- [9, 8, 7, 6, 5, 3, 0]
*
* \endcode
*
* \tparam KeyT <b>[inferred]</b> KeyT type
*/
template <typename KeyT>
CUB_RUNTIME_FUNCTION
static cudaError_t SortKeysDescending(
void *d_temp_storage, ///< [in] %Device-accessible allocation of temporary storage. When NULL, the required allocation size is written to \p temp_storage_bytes and no work is done.
size_t &temp_storage_bytes, ///< [in,out] Reference to size in bytes of \p d_temp_storage allocation
DoubleBuffer<KeyT> &d_keys, ///< [in,out] Reference to the double-buffer of keys whose "current" device-accessible buffer contains the unsorted input keys and, upon return, is updated to point to the sorted output keys
int num_items, ///< [in] Number of items to sort
int begin_bit = 0, ///< [in] <b>[optional]</b> The least-significant bit index (inclusive) needed for key comparison
int end_bit = sizeof(KeyT) * 8, ///< [in] <b>[optional]</b> The most-significant bit index (exclusive) needed for key comparison (e.g., sizeof(unsigned int) * 8)
cudaStream_t stream = 0, ///< [in] <b>[optional]</b> CUDA stream to launch kernels within. Default is stream<sub>0</sub>.
bool debug_synchronous = false) ///< [in] <b>[optional]</b> Whether or not to synchronize the stream after every kernel launch to check for errors. Also causes launch configurations to be printed to the console. Default is \p false.
{
// Signed integer type for global offsets
typedef int OffsetT;
// Null value type
DoubleBuffer<NullType> d_values;
return DispatchRadixSort<true, KeyT, NullType, OffsetT>::Dispatch(
d_temp_storage,
temp_storage_bytes,
d_keys,
d_values,
num_items,
begin_bit,
end_bit,
true,
stream,
debug_synchronous);
}
//@} end member group
};
/**
* \example example_device_radix_sort.cu
*/
} // CUB namespace
CUB_NS_POSTFIX // Optional outer namespace(s)
|