Дмитрий Изволов
69b54c1a05
[libcxx][algorithm] Optimize std::stable_sort via radix sort algorithm ( #104683 )
...
The radix sort (LSD) algorithm allows to speed up std::stable_sort
dramatically in case we sort integers.
The speed up varies from a relatively small to x10 times, depending on
type of sorted elements and the initial state of the sorted array.
```
Running ./libcxx/test/benchmarks/stable_sort.bench.out
Run on (12 X 2600 MHz CPU s)
CPU Caches:
L1 Data 32 KiB
L1 Instruction 32 KiB
L2 Unified 256 KiB (x6)
L3 Unified 12288 KiB
Load Average: 3.48, 3.38, 3.08
---------------------------------------------------------------------------
Benchmark After Before
---------------------------------------------------------------------------
BM_StableSort_int8_Random_1 3.39 ns 3.58 ns
BM_StableSort_int8_Random_4 21.1 ns 21.9 ns
BM_StableSort_int8_Random_16 142 ns 147 ns
BM_StableSort_int8_Random_64 893 ns 903 ns
BM_StableSort_int8_Random_256 409 ns 5810 ns
BM_StableSort_int8_Random_1024 1235 ns 29973 ns
BM_StableSort_int8_Random_4096 4410 ns 141880 ns
BM_StableSort_int8_Random_16384 18044 ns 620540 ns
BM_StableSort_int8_Random_65536 144030 ns 2592013 ns
BM_StableSort_int8_Random_262144 858350 ns 10935814 ns
BM_StableSort_int8_Random_524288 2929988 ns 27060729 ns
BM_StableSort_int8_Random_1048576 6058292 ns 49622720 ns
BM_StableSort_int8_Ascending_1 3.42 ns 3.92 ns
BM_StableSort_int8_Ascending_4 5.86 ns 8.08 ns
BM_StableSort_int8_Ascending_16 10.6 ns 12.0 ns
BM_StableSort_int8_Ascending_64 28.9 ns 30.6 ns
BM_StableSort_int8_Ascending_256 415 ns 391 ns
BM_StableSort_int8_Ascending_1024 1666 ns 2309 ns
BM_StableSort_int8_Ascending_4096 7748 ns 12269 ns
BM_StableSort_int8_Ascending_16384 40588 ns 60181 ns
BM_StableSort_int8_Ascending_65536 178843 ns 298221 ns
BM_StableSort_int8_Ascending_262144 919959 ns 1402692 ns
BM_StableSort_int8_Ascending_524288 2397397 ns 3036984 ns
BM_StableSort_int8_Ascending_1048576 5080043 ns 7218581 ns
BM_StableSort_int8_Descending_1 3.44 ns 3.53 ns
BM_StableSort_int8_Descending_4 7.94 ns 8.29 ns
BM_StableSort_int8_Descending_16 59.6 ns 57.7 ns
BM_StableSort_int8_Descending_64 1051 ns 1027 ns
BM_StableSort_int8_Descending_256 422 ns 4718 ns
BM_StableSort_int8_Descending_1024 1676 ns 21044 ns
BM_StableSort_int8_Descending_4096 7766 ns 64827 ns
BM_StableSort_int8_Descending_16384 40230 ns 93981 ns
BM_StableSort_int8_Descending_65536 190978 ns 421151 ns
BM_StableSort_int8_Descending_262144 1055141 ns 1918927 ns
BM_StableSort_int8_Descending_524288 2875115 ns 3809153 ns
BM_StableSort_int8_Descending_1048576 5854135 ns 8713690 ns
BM_StableSort_int8_SingleElement_1 3.52 ns 3.46 ns
BM_StableSort_int8_SingleElement_4 6.25 ns 5.79 ns
BM_StableSort_int8_SingleElement_16 10.7 ns 11.4 ns
BM_StableSort_int8_SingleElement_64 29.3 ns 30.3 ns
BM_StableSort_int8_SingleElement_256 858 ns 380 ns
BM_StableSort_int8_SingleElement_1024 3036 ns 2231 ns
BM_StableSort_int8_SingleElement_4096 11580 ns 11866 ns
BM_StableSort_int8_SingleElement_16384 44956 ns 59621 ns
BM_StableSort_int8_SingleElement_65536 182006 ns 297853 ns
BM_StableSort_int8_SingleElement_262144 962181 ns 1432857 ns
BM_StableSort_int8_SingleElement_524288 2256687 ns 2975707 ns
BM_StableSort_int8_SingleElement_1048576 4522556 ns 6949948 ns
BM_StableSort_int8_PipeOrgan_1 3.26 ns 3.64 ns
BM_StableSort_int8_PipeOrgan_4 6.21 ns 6.58 ns
BM_StableSort_int8_PipeOrgan_16 23.7 ns 25.4 ns
BM_StableSort_int8_PipeOrgan_64 250 ns 248 ns
BM_StableSort_int8_PipeOrgan_256 414 ns 2498 ns
BM_StableSort_int8_PipeOrgan_1024 1697 ns 10946 ns
BM_StableSort_int8_PipeOrgan_4096 7840 ns 37238 ns
BM_StableSort_int8_PipeOrgan_16384 41402 ns 74805 ns
BM_StableSort_int8_PipeOrgan_65536 180107 ns 357891 ns
BM_StableSort_int8_PipeOrgan_262144 988273 ns 1647296 ns
BM_StableSort_int8_PipeOrgan_524288 2547374 ns 3245991 ns
BM_StableSort_int8_PipeOrgan_1048576 5128783 ns 7342444 ns
BM_StableSort_int8_QuickSortAdversary_1 3.14 ns 4.01 ns
BM_StableSort_int8_QuickSortAdversary_4 6.05 ns 7.02 ns
BM_StableSort_int8_QuickSortAdversary_16 10.5 ns 11.9 ns
BM_StableSort_int8_QuickSortAdversary_64 520 ns 516 ns
BM_StableSort_int8_QuickSortAdversary_256 920 ns 386 ns
BM_StableSort_int8_QuickSortAdversary_1024 3083 ns 2299 ns
BM_StableSort_int8_QuickSortAdversary_4096 11659 ns 12295 ns
BM_StableSort_int8_QuickSortAdversary_16384 45721 ns 60931 ns
BM_StableSort_int8_QuickSortAdversary_65536 186334 ns 295423 ns
BM_StableSort_int8_QuickSortAdversary_262144 946262 ns 1399973 ns
BM_StableSort_int8_QuickSortAdversary_524288 2282004 ns 2832266 ns
BM_StableSort_int8_QuickSortAdversary_1048576 4691123 ns 6963253 ns
BM_StableSort_uint8_Random_1 3.11 ns 3.44 ns
BM_StableSort_uint8_Random_4 21.9 ns 23.1 ns
BM_StableSort_uint8_Random_16 154 ns 171 ns
BM_StableSort_uint8_Random_64 1000 ns 1051 ns
BM_StableSort_uint8_Random_256 402 ns 6498 ns
BM_StableSort_uint8_Random_1024 1176 ns 35310 ns
BM_StableSort_uint8_Random_4096 4415 ns 164087 ns
BM_StableSort_uint8_Random_16384 17849 ns 686769 ns
BM_StableSort_uint8_Random_65536 146109 ns 2932051 ns
BM_StableSort_uint8_Random_262144 876710 ns 12163988 ns
BM_StableSort_uint8_Random_524288 2858089 ns 26458830 ns
BM_StableSort_uint8_Random_1048576 5766942 ns 54836214 ns
BM_StableSort_uint8_Ascending_1 3.11 ns 3.43 ns
BM_StableSort_uint8_Ascending_4 6.18 ns 7.24 ns
BM_StableSort_uint8_Ascending_16 14.5 ns 17.0 ns
BM_StableSort_uint8_Ascending_64 50.7 ns 59.2 ns
BM_StableSort_uint8_Ascending_256 395 ns 536 ns
BM_StableSort_uint8_Ascending_1024 1752 ns 2956 ns
BM_StableSort_uint8_Ascending_4096 7785 ns 15146 ns
BM_StableSort_uint8_Ascending_16384 41442 ns 74136 ns
BM_StableSort_uint8_Ascending_65536 180879 ns 354261 ns
BM_StableSort_uint8_Ascending_262144 945880 ns 1674256 ns
BM_StableSort_uint8_Ascending_524288 2287832 ns 3138581 ns
BM_StableSort_uint8_Ascending_1048576 4630290 ns 7296278 ns
BM_StableSort_uint8_Descending_1 3.19 ns 3.63 ns
BM_StableSort_uint8_Descending_4 9.60 ns 11.5 ns
BM_StableSort_uint8_Descending_16 78.3 ns 86.0 ns
BM_StableSort_uint8_Descending_64 1265 ns 1308 ns
BM_StableSort_uint8_Descending_256 395 ns 6556 ns
BM_StableSort_uint8_Descending_1024 1712 ns 24669 ns
BM_StableSort_uint8_Descending_4096 7748 ns 83407 ns
BM_StableSort_uint8_Descending_16384 40779 ns 104043 ns
BM_StableSort_uint8_Descending_65536 181560 ns 467680 ns
BM_StableSort_uint8_Descending_262144 1146627 ns 2102769 ns
BM_StableSort_uint8_Descending_524288 2874096 ns 4572229 ns
BM_StableSort_uint8_Descending_1048576 5873195 ns 10170663 ns
BM_StableSort_uint8_SingleElement_1 3.28 ns 3.58 ns
BM_StableSort_uint8_SingleElement_4 6.44 ns 7.40 ns
BM_StableSort_uint8_SingleElement_16 14.9 ns 16.4 ns
BM_StableSort_uint8_SingleElement_64 51.2 ns 52.9 ns
BM_StableSort_uint8_SingleElement_256 876 ns 490 ns
BM_StableSort_uint8_SingleElement_1024 3041 ns 2750 ns
BM_StableSort_uint8_SingleElement_4096 11947 ns 14326 ns
BM_StableSort_uint8_SingleElement_16384 46669 ns 69984 ns
BM_StableSort_uint8_SingleElement_65536 197903 ns 328961 ns
BM_StableSort_uint8_SingleElement_262144 1031466 ns 1551436 ns
BM_StableSort_uint8_SingleElement_524288 2447672 ns 3049553 ns
BM_StableSort_uint8_SingleElement_1048576 4793087 ns 7615245 ns
BM_StableSort_uint8_PipeOrgan_1 3.38 ns 3.56 ns
BM_StableSort_uint8_PipeOrgan_4 7.16 ns 8.70 ns
BM_StableSort_uint8_PipeOrgan_16 31.7 ns 35.3 ns
BM_StableSort_uint8_PipeOrgan_64 326 ns 366 ns
BM_StableSort_uint8_PipeOrgan_256 409 ns 2942 ns
BM_StableSort_uint8_PipeOrgan_1024 1994 ns 12571 ns
BM_StableSort_uint8_PipeOrgan_4096 8086 ns 46278 ns
BM_StableSort_uint8_PipeOrgan_16384 41749 ns 79813 ns
BM_StableSort_uint8_PipeOrgan_65536 180697 ns 375120 ns
BM_StableSort_uint8_PipeOrgan_262144 1004899 ns 1676143 ns
BM_StableSort_uint8_PipeOrgan_524288 2456081 ns 3333949 ns
BM_StableSort_uint8_PipeOrgan_1048576 5030857 ns 7591303 ns
BM_StableSort_uint8_QuickSortAdversary_1 3.12 ns 3.46 ns
BM_StableSort_uint8_QuickSortAdversary_4 7.25 ns 6.83 ns
BM_StableSort_uint8_QuickSortAdversary_16 14.6 ns 16.2 ns
BM_StableSort_uint8_QuickSortAdversary_64 650 ns 665 ns
BM_StableSort_uint8_QuickSortAdversary_256 395 ns 2982 ns
BM_StableSort_uint8_QuickSortAdversary_1024 3125 ns 2583 ns
BM_StableSort_uint8_QuickSortAdversary_4096 11797 ns 13929 ns
BM_StableSort_uint8_QuickSortAdversary_16384 45803 ns 66513 ns
BM_StableSort_uint8_QuickSortAdversary_65536 190745 ns 313467 ns
BM_StableSort_uint8_QuickSortAdversary_262144 974646 ns 1469014 ns
BM_StableSort_uint8_QuickSortAdversary_524288 2317553 ns 3022065 ns
BM_StableSort_uint8_QuickSortAdversary_1048576 4898703 ns 6854079 ns
BM_StableSort_int16_Random_1 3.94 ns 3.49 ns
BM_StableSort_int16_Random_4 20.8 ns 23.2 ns
BM_StableSort_int16_Random_16 133 ns 163 ns
BM_StableSort_int16_Random_64 903 ns 953 ns
BM_StableSort_int16_Random_256 5638 ns 6258 ns
BM_StableSort_int16_Random_1024 3056 ns 34587 ns
BM_StableSort_int16_Random_4096 10596 ns 168397 ns
BM_StableSort_int16_Random_16384 49908 ns 753031 ns
BM_StableSort_int16_Random_65536 444605 ns 3838368 ns
BM_StableSort_int16_Random_262144 2419345 ns 15657285 ns
BM_StableSort_int16_Random_524288 7984040 ns 32726933 ns
BM_StableSort_int16_Random_1048576 16092424 ns 67999766 ns
BM_StableSort_int16_Ascending_1 3.40 ns 3.43 ns
BM_StableSort_int16_Ascending_4 5.45 ns 5.79 ns
BM_StableSort_int16_Ascending_16 12.0 ns 15.3 ns
BM_StableSort_int16_Ascending_64 39.6 ns 52.6 ns
BM_StableSort_int16_Ascending_256 470 ns 550 ns
BM_StableSort_int16_Ascending_1024 1686 ns 2707 ns
BM_StableSort_int16_Ascending_4096 5676 ns 14165 ns
BM_StableSort_int16_Ascending_16384 21413 ns 69483 ns
BM_StableSort_int16_Ascending_65536 88010 ns 334466 ns
BM_StableSort_int16_Ascending_262144 567239 ns 1570620 ns
BM_StableSort_int16_Ascending_524288 1553063 ns 3424666 ns
BM_StableSort_int16_Ascending_1048576 3145577 ns 8499649 ns
BM_StableSort_int16_Descending_1 3.22 ns 3.54 ns
BM_StableSort_int16_Descending_4 6.85 ns 10.2 ns
BM_StableSort_int16_Descending_16 62.7 ns 62.2 ns
BM_StableSort_int16_Descending_64 1138 ns 1036 ns
BM_StableSort_int16_Descending_256 5541 ns 4696 ns
BM_StableSort_int16_Descending_1024 3046 ns 19577 ns
BM_StableSort_int16_Descending_4096 10962 ns 79149 ns
BM_StableSort_int16_Descending_16384 58182 ns 327709 ns
BM_StableSort_int16_Descending_65536 447025 ns 1424896 ns
BM_StableSort_int16_Descending_262144 1104973 ns 5921903 ns
BM_StableSort_int16_Descending_524288 2547840 ns 17956789 ns
BM_StableSort_int16_Descending_1048576 5093555 ns 17044318 ns
BM_StableSort_int16_SingleElement_1 3.56 ns 3.96 ns
BM_StableSort_int16_SingleElement_4 5.75 ns 6.72 ns
BM_StableSort_int16_SingleElement_16 12.4 ns 16.1 ns
BM_StableSort_int16_SingleElement_64 36.9 ns 54.4 ns
BM_StableSort_int16_SingleElement_256 473 ns 557 ns
BM_StableSort_int16_SingleElement_1024 1828 ns 2826 ns
BM_StableSort_int16_SingleElement_4096 6239 ns 14252 ns
BM_StableSort_int16_SingleElement_16384 23695 ns 70369 ns
BM_StableSort_int16_SingleElement_65536 93281 ns 361641 ns
BM_StableSort_int16_SingleElement_262144 599078 ns 1640216 ns
BM_StableSort_int16_SingleElement_524288 1659678 ns 3343087 ns
BM_StableSort_int16_SingleElement_1048576 3184033 ns 7770271 ns
BM_StableSort_int16_PipeOrgan_1 3.75 ns 3.76 ns
BM_StableSort_int16_PipeOrgan_4 5.94 ns 7.74 ns
BM_StableSort_int16_PipeOrgan_16 26.7 ns 25.9 ns
BM_StableSort_int16_PipeOrgan_64 300 ns 263 ns
BM_StableSort_int16_PipeOrgan_256 2769 ns 2760 ns
BM_StableSort_int16_PipeOrgan_1024 2996 ns 10544 ns
BM_StableSort_int16_PipeOrgan_4096 11641 ns 44750 ns
BM_StableSort_int16_PipeOrgan_16384 57224 ns 200464 ns
BM_StableSort_int16_PipeOrgan_65536 416873 ns 887631 ns
BM_StableSort_int16_PipeOrgan_262144 843264 ns 3588669 ns
BM_StableSort_int16_PipeOrgan_524288 2027741 ns 11056924 ns
BM_StableSort_int16_PipeOrgan_1048576 4223773 ns 13261276 ns
BM_StableSort_int16_QuickSortAdversary_1 3.83 ns 3.68 ns
BM_StableSort_int16_QuickSortAdversary_4 5.55 ns 6.93 ns
BM_StableSort_int16_QuickSortAdversary_16 12.3 ns 15.2 ns
BM_StableSort_int16_QuickSortAdversary_64 646 ns 632 ns
BM_StableSort_int16_QuickSortAdversary_256 2751 ns 2542 ns
BM_StableSort_int16_QuickSortAdversary_1024 3028 ns 16901 ns
BM_StableSort_int16_QuickSortAdversary_4096 10862 ns 80222 ns
BM_StableSort_int16_QuickSortAdversary_16384 57753 ns 317281 ns
BM_StableSort_int16_QuickSortAdversary_65536 94064 ns 328502 ns
BM_StableSort_int16_QuickSortAdversary_262144 557796 ns 1613208 ns
BM_StableSort_int16_QuickSortAdversary_524288 1518451 ns 3479740 ns
BM_StableSort_int16_QuickSortAdversary_1048576 3165129 ns 7655880 ns
BM_StableSort_uint16_Random_1 3.26 ns 3.44 ns
BM_StableSort_uint16_Random_4 21.1 ns 22.2 ns
BM_StableSort_uint16_Random_16 157 ns 156 ns
BM_StableSort_uint16_Random_64 955 ns 947 ns
BM_StableSort_uint16_Random_256 5886 ns 6097 ns
BM_StableSort_uint16_Random_1024 2787 ns 30776 ns
BM_StableSort_uint16_Random_4096 9973 ns 155652 ns
BM_StableSort_uint16_Random_16384 48628 ns 741072 ns
BM_StableSort_uint16_Random_65536 439609 ns 3478966 ns
BM_StableSort_uint16_Random_262144 2336983 ns 15197642 ns
BM_StableSort_uint16_Random_524288 7888701 ns 34234254 ns
BM_StableSort_uint16_Random_1048576 14865180 ns 68516386 ns
BM_StableSort_uint16_Ascending_1 3.33 ns 4.00 ns
BM_StableSort_uint16_Ascending_4 5.79 ns 6.64 ns
BM_StableSort_uint16_Ascending_16 14.9 ns 15.5 ns
BM_StableSort_uint16_Ascending_64 50.2 ns 52.5 ns
BM_StableSort_uint16_Ascending_256 538 ns 546 ns
BM_StableSort_uint16_Ascending_1024 1645 ns 2652 ns
BM_StableSort_uint16_Ascending_4096 5559 ns 14517 ns
BM_StableSort_uint16_Ascending_16384 22803 ns 70275 ns
BM_StableSort_uint16_Ascending_65536 83109 ns 333446 ns
BM_StableSort_uint16_Ascending_262144 562667 ns 1568670 ns
BM_StableSort_uint16_Ascending_524288 1564646 ns 3059839 ns
BM_StableSort_uint16_Ascending_1048576 3178826 ns 7048327 ns
BM_StableSort_uint16_Descending_1 3.34 ns 3.93 ns
BM_StableSort_uint16_Descending_4 8.75 ns 9.73 ns
BM_StableSort_uint16_Descending_16 55.9 ns 55.5 ns
BM_StableSort_uint16_Descending_64 1021 ns 1035 ns
BM_StableSort_uint16_Descending_256 4752 ns 4931 ns
BM_StableSort_uint16_Descending_1024 2982 ns 19727 ns
BM_StableSort_uint16_Descending_4096 10432 ns 83165 ns
BM_StableSort_uint16_Descending_16384 56593 ns 326131 ns
BM_StableSort_uint16_Descending_65536 439134 ns 1371346 ns
BM_StableSort_uint16_Descending_262144 1220925 ns 5735665 ns
BM_StableSort_uint16_Descending_524288 2767234 ns 16758330 ns
BM_StableSort_uint16_Descending_1048576 5673769 ns 17541715 ns
BM_StableSort_uint16_SingleElement_1 3.53 ns 3.73 ns
BM_StableSort_uint16_SingleElement_4 6.27 ns 5.81 ns
BM_StableSort_uint16_SingleElement_16 14.8 ns 15.1 ns
BM_StableSort_uint16_SingleElement_64 51.5 ns 50.9 ns
BM_StableSort_uint16_SingleElement_256 536 ns 540 ns
BM_StableSort_uint16_SingleElement_1024 1669 ns 2690 ns
BM_StableSort_uint16_SingleElement_4096 5840 ns 14230 ns
BM_StableSort_uint16_SingleElement_16384 22468 ns 68524 ns
BM_StableSort_uint16_SingleElement_65536 89845 ns 332187 ns
BM_StableSort_uint16_SingleElement_262144 590736 ns 1550868 ns
BM_StableSort_uint16_SingleElement_524288 1573677 ns 3095703 ns
BM_StableSort_uint16_SingleElement_1048576 3183421 ns 8251180 ns
BM_StableSort_uint16_PipeOrgan_1 3.70 ns 3.64 ns
BM_StableSort_uint16_PipeOrgan_4 7.01 ns 6.81 ns
BM_StableSort_uint16_PipeOrgan_16 25.7 ns 26.4 ns
BM_StableSort_uint16_PipeOrgan_64 283 ns 277 ns
BM_StableSort_uint16_PipeOrgan_256 2562 ns 2852 ns
BM_StableSort_uint16_PipeOrgan_1024 2863 ns 10892 ns
BM_StableSort_uint16_PipeOrgan_4096 10585 ns 45668 ns
BM_StableSort_uint16_PipeOrgan_16384 59151 ns 194358 ns
BM_StableSort_uint16_PipeOrgan_65536 508579 ns 854692 ns
BM_StableSort_uint16_PipeOrgan_262144 901294 ns 3606346 ns
BM_StableSort_uint16_PipeOrgan_524288 2192498 ns 10449279 ns
BM_StableSort_uint16_PipeOrgan_1048576 4204368 ns 11956606 ns
BM_StableSort_uint16_QuickSortAdversary_1 3.20 ns 3.63 ns
BM_StableSort_uint16_QuickSortAdversary_4 5.30 ns 6.38 ns
BM_StableSort_uint16_QuickSortAdversary_16 14.5 ns 15.3 ns
BM_StableSort_uint16_QuickSortAdversary_64 575 ns 611 ns
BM_StableSort_uint16_QuickSortAdversary_256 2423 ns 2577 ns
BM_StableSort_uint16_QuickSortAdversary_1024 2794 ns 16854 ns
BM_StableSort_uint16_QuickSortAdversary_4096 10511 ns 75952 ns
BM_StableSort_uint16_QuickSortAdversary_16384 56214 ns 333824 ns
BM_StableSort_uint16_QuickSortAdversary_65536 422512 ns 1354867 ns
BM_StableSort_uint16_QuickSortAdversary_262144 583301 ns 1564443 ns
BM_StableSort_uint16_QuickSortAdversary_524288 1584319 ns 3265575 ns
BM_StableSort_uint16_QuickSortAdversary_1048576 3197732 ns 7945245 ns
BM_StableSort_int32_Random_1 3.81 ns 3.70 ns
BM_StableSort_int32_Random_4 20.8 ns 23.4 ns
BM_StableSort_int32_Random_16 134 ns 161 ns
BM_StableSort_int32_Random_64 895 ns 984 ns
BM_StableSort_int32_Random_256 5640 ns 5897 ns
BM_StableSort_int32_Random_1024 6994 ns 32118 ns
BM_StableSort_int32_Random_4096 27367 ns 168960 ns
BM_StableSort_int32_Random_16384 183261 ns 843240 ns
BM_StableSort_int32_Random_65536 950914 ns 3953588 ns
BM_StableSort_int32_Random_262144 3673311 ns 16790171 ns
BM_StableSort_int32_Random_524288 11515700 ns 36023098 ns
BM_StableSort_int32_Random_1048576 24492515 ns 78116028 ns
BM_StableSort_int32_Ascending_1 3.31 ns 4.48 ns
BM_StableSort_int32_Ascending_4 5.96 ns 6.99 ns
BM_StableSort_int32_Ascending_16 13.0 ns 16.0 ns
BM_StableSort_int32_Ascending_64 36.7 ns 53.0 ns
BM_StableSort_int32_Ascending_256 391 ns 471 ns
BM_StableSort_int32_Ascending_1024 2705 ns 2682 ns
BM_StableSort_int32_Ascending_4096 8773 ns 14231 ns
BM_StableSort_int32_Ascending_16384 34709 ns 70625 ns
BM_StableSort_int32_Ascending_65536 142907 ns 344482 ns
BM_StableSort_int32_Ascending_262144 745483 ns 1591418 ns
BM_StableSort_int32_Ascending_524288 1873701 ns 3190305 ns
BM_StableSort_int32_Ascending_1048576 3851590 ns 7570095 ns
BM_StableSort_int32_Descending_1 3.22 ns 4.23 ns
BM_StableSort_int32_Descending_4 7.58 ns 11.2 ns
BM_StableSort_int32_Descending_16 63.9 ns 58.6 ns
BM_StableSort_int32_Descending_64 1133 ns 1017 ns
BM_StableSort_int32_Descending_256 4850 ns 4464 ns
BM_StableSort_int32_Descending_1024 7023 ns 18954 ns
BM_StableSort_int32_Descending_4096 28550 ns 75163 ns
BM_StableSort_int32_Descending_16384 200880 ns 341104 ns
BM_StableSort_int32_Descending_65536 1095910 ns 1398021 ns
BM_StableSort_int32_Descending_262144 3818864 ns 5695486 ns
BM_StableSort_int32_Descending_524288 5606779 ns 17593982 ns
BM_StableSort_int32_Descending_1048576 16416366 ns 26649503 ns
BM_StableSort_int32_SingleElement_1 3.81 ns 3.71 ns
BM_StableSort_int32_SingleElement_4 6.57 ns 6.61 ns
BM_StableSort_int32_SingleElement_16 14.0 ns 15.8 ns
BM_StableSort_int32_SingleElement_64 38.7 ns 53.5 ns
BM_StableSort_int32_SingleElement_256 386 ns 554 ns
BM_StableSort_int32_SingleElement_1024 2761 ns 3046 ns
BM_StableSort_int32_SingleElement_4096 9179 ns 15188 ns
BM_StableSort_int32_SingleElement_16384 34794 ns 70119 ns
BM_StableSort_int32_SingleElement_65536 135190 ns 354755 ns
BM_StableSort_int32_SingleElement_262144 760995 ns 1644072 ns
BM_StableSort_int32_SingleElement_524288 1969575 ns 3343419 ns
BM_StableSort_int32_SingleElement_1048576 4423816 ns 8346971 ns
BM_StableSort_int32_PipeOrgan_1 3.79 ns 3.63 ns
BM_StableSort_int32_PipeOrgan_4 6.21 ns 6.73 ns
BM_StableSort_int32_PipeOrgan_16 27.5 ns 26.0 ns
BM_StableSort_int32_PipeOrgan_64 291 ns 265 ns
BM_StableSort_int32_PipeOrgan_256 2557 ns 2518 ns
BM_StableSort_int32_PipeOrgan_1024 6765 ns 10976 ns
BM_StableSort_int32_PipeOrgan_4096 26373 ns 44537 ns
BM_StableSort_int32_PipeOrgan_16384 201466 ns 188582 ns
BM_StableSort_int32_PipeOrgan_65536 1148533 ns 802368 ns
BM_StableSort_int32_PipeOrgan_262144 2255177 ns 3477829 ns
BM_StableSort_int32_PipeOrgan_524288 3947015 ns 10356637 ns
BM_StableSort_int32_PipeOrgan_1048576 10274312 ns 16405366 ns
BM_StableSort_int32_QuickSortAdversary_1 3.32 ns 4.36 ns
BM_StableSort_int32_QuickSortAdversary_4 5.98 ns 7.44 ns
BM_StableSort_int32_QuickSortAdversary_16 13.0 ns 16.3 ns
BM_StableSort_int32_QuickSortAdversary_64 657 ns 616 ns
BM_StableSort_int32_QuickSortAdversary_256 2569 ns 2483 ns
BM_StableSort_int32_QuickSortAdversary_1024 6898 ns 19635 ns
BM_StableSort_int32_QuickSortAdversary_4096 27092 ns 75108 ns
BM_StableSort_int32_QuickSortAdversary_16384 190379 ns 316463 ns
BM_StableSort_int32_QuickSortAdversary_65536 1109040 ns 1319018 ns
BM_StableSort_int32_QuickSortAdversary_262144 4361925 ns 5472779 ns
BM_StableSort_int32_QuickSortAdversary_524288 6528215 ns 17538983 ns
BM_StableSort_int32_QuickSortAdversary_1048576 18345325 ns 27223926 ns
BM_StableSort_uint32_Random_1 3.67 ns 3.82 ns
BM_StableSort_uint32_Random_4 22.3 ns 21.8 ns
BM_StableSort_uint32_Random_16 155 ns 153 ns
BM_StableSort_uint32_Random_64 946 ns 976 ns
BM_StableSort_uint32_Random_256 5824 ns 6019 ns
BM_StableSort_uint32_Random_1024 4525 ns 32764 ns
BM_StableSort_uint32_Random_4096 17223 ns 158608 ns
BM_StableSort_uint32_Random_16384 134821 ns 748525 ns
BM_StableSort_uint32_Random_65536 716644 ns 3453325 ns
BM_StableSort_uint32_Random_262144 3628062 ns 16065414 ns
BM_StableSort_uint32_Random_524288 10971334 ns 36567712 ns
BM_StableSort_uint32_Random_1048576 22688377 ns 77533497 ns
BM_StableSort_uint32_Ascending_1 3.57 ns 3.44 ns
BM_StableSort_uint32_Ascending_4 5.73 ns 5.33 ns
BM_StableSort_uint32_Ascending_16 14.5 ns 14.0 ns
BM_StableSort_uint32_Ascending_64 50.3 ns 51.3 ns
BM_StableSort_uint32_Ascending_256 465 ns 467 ns
BM_StableSort_uint32_Ascending_1024 3042 ns 2530 ns
BM_StableSort_uint32_Ascending_4096 9842 ns 12207 ns
BM_StableSort_uint32_Ascending_16384 37994 ns 61726 ns
BM_StableSort_uint32_Ascending_65536 148890 ns 294385 ns
BM_StableSort_uint32_Ascending_262144 855080 ns 1422167 ns
BM_StableSort_uint32_Ascending_524288 2154903 ns 3203018 ns
BM_StableSort_uint32_Ascending_1048576 5002518 ns 7563817 ns
BM_StableSort_uint32_Descending_1 3.51 ns 3.40 ns
BM_StableSort_uint32_Descending_4 9.09 ns 7.95 ns
BM_StableSort_uint32_Descending_16 54.8 ns 74.4 ns
BM_StableSort_uint32_Descending_64 1003 ns 1305 ns
BM_StableSort_uint32_Descending_256 4545 ns 5300 ns
BM_StableSort_uint32_Descending_1024 4361 ns 21884 ns
BM_StableSort_uint32_Descending_4096 16018 ns 90534 ns
BM_StableSort_uint32_Descending_16384 146274 ns 381943 ns
BM_StableSort_uint32_Descending_65536 938248 ns 1536806 ns
BM_StableSort_uint32_Descending_262144 3899300 ns 6387843 ns
BM_StableSort_uint32_Descending_524288 5808157 ns 21959858 ns
BM_StableSort_uint32_Descending_1048576 17520047 ns 26351912 ns
BM_StableSort_uint32_SingleElement_1 4.03 ns 3.97 ns
BM_StableSort_uint32_SingleElement_4 6.55 ns 6.41 ns
BM_StableSort_uint32_SingleElement_16 15.6 ns 15.8 ns
BM_StableSort_uint32_SingleElement_64 52.3 ns 58.7 ns
BM_StableSort_uint32_SingleElement_256 473 ns 485 ns
BM_StableSort_uint32_SingleElement_1024 3020 ns 2407 ns
BM_StableSort_uint32_SingleElement_4096 9998 ns 12527 ns
BM_StableSort_uint32_SingleElement_16384 38072 ns 62228 ns
BM_StableSort_uint32_SingleElement_65536 153706 ns 295662 ns
BM_StableSort_uint32_SingleElement_262144 836532 ns 1477099 ns
BM_StableSort_uint32_SingleElement_524288 2144900 ns 3157204 ns
BM_StableSort_uint32_SingleElement_1048576 4995525 ns 7617233 ns
BM_StableSort_uint32_PipeOrgan_1 4.02 ns 3.99 ns
BM_StableSort_uint32_PipeOrgan_4 6.97 ns 6.84 ns
BM_StableSort_uint32_PipeOrgan_16 26.1 ns 29.7 ns
BM_StableSort_uint32_PipeOrgan_64 266 ns 333 ns
BM_StableSort_uint32_PipeOrgan_256 2462 ns 2892 ns
BM_StableSort_uint32_PipeOrgan_1024 4291 ns 12431 ns
BM_StableSort_uint32_PipeOrgan_4096 15638 ns 51449 ns
BM_StableSort_uint32_PipeOrgan_16384 154563 ns 217460 ns
BM_StableSort_uint32_PipeOrgan_65536 907724 ns 925873 ns
BM_StableSort_uint32_PipeOrgan_262144 2394580 ns 4103575 ns
BM_StableSort_uint32_PipeOrgan_524288 4177145 ns 13947158 ns
BM_StableSort_uint32_PipeOrgan_1048576 11848224 ns 18807297 ns
BM_StableSort_uint32_QuickSortAdversary_1 3.50 ns 3.43 ns
BM_StableSort_uint32_QuickSortAdversary_4 5.88 ns 4.96 ns
BM_StableSort_uint32_QuickSortAdversary_16 14.6 ns 14.0 ns
BM_StableSort_uint32_QuickSortAdversary_64 576 ns 715 ns
BM_StableSort_uint32_QuickSortAdversary_256 2353 ns 2797 ns
BM_StableSort_uint32_QuickSortAdversary_1024 4176 ns 21775 ns
BM_StableSort_uint32_QuickSortAdversary_4096 15565 ns 96188 ns
BM_StableSort_uint32_QuickSortAdversary_16384 149092 ns 398332 ns
BM_StableSort_uint32_QuickSortAdversary_65536 902488 ns 1552393 ns
BM_StableSort_uint32_QuickSortAdversary_262144 3946517 ns 6560414 ns
BM_StableSort_uint32_QuickSortAdversary_524288 6247114 ns 22420977 ns
BM_StableSort_uint32_QuickSortAdversary_1048576 19892446 ns 26529576 ns
BM_StableSort_int64_Random_1 3.83 ns 3.98 ns
BM_StableSort_int64_Random_4 21.1 ns 24.0 ns
BM_StableSort_int64_Random_16 129 ns 136 ns
BM_StableSort_int64_Random_64 890 ns 906 ns
BM_StableSort_int64_Random_256 5542 ns 5901 ns
BM_StableSort_int64_Random_1024 16085 ns 33112 ns
BM_StableSort_int64_Random_4096 63895 ns 162181 ns
BM_StableSort_int64_Random_16384 348827 ns 790045 ns
BM_StableSort_int64_Random_65536 1488237 ns 3557506 ns
BM_StableSort_int64_Random_262144 8195713 ns 16315808 ns
BM_StableSort_int64_Random_524288 16586833 ns 38274075 ns
BM_StableSort_int64_Random_1048576 40346644 ns 79182089 ns
BM_StableSort_int64_Ascending_1 3.76 ns 3.55 ns
BM_StableSort_int64_Ascending_4 5.82 ns 6.19 ns
BM_StableSort_int64_Ascending_16 11.7 ns 11.8 ns
BM_StableSort_int64_Ascending_64 32.9 ns 36.8 ns
BM_StableSort_int64_Ascending_256 415 ns 550 ns
BM_StableSort_int64_Ascending_1024 5352 ns 3347 ns
BM_StableSort_int64_Ascending_4096 17516 ns 19134 ns
BM_StableSort_int64_Ascending_16384 64147 ns 91099 ns
BM_StableSort_int64_Ascending_65536 322126 ns 434009 ns
BM_StableSort_int64_Ascending_262144 1554669 ns 2057056 ns
BM_StableSort_int64_Ascending_524288 3656527 ns 5016650 ns
BM_StableSort_int64_Ascending_1048576 10469979 ns 12908613 ns
BM_StableSort_int64_Descending_1 4.09 ns 3.35 ns
BM_StableSort_int64_Descending_4 9.13 ns 8.01 ns
BM_StableSort_int64_Descending_16 76.8 ns 92.9 ns
BM_StableSort_int64_Descending_64 1336 ns 1417 ns
BM_StableSort_int64_Descending_256 5525 ns 5674 ns
BM_StableSort_int64_Descending_1024 17461 ns 22558 ns
BM_StableSort_int64_Descending_4096 64285 ns 102360 ns
BM_StableSort_int64_Descending_16384 336946 ns 388940 ns
BM_StableSort_int64_Descending_65536 837912 ns 1662169 ns
BM_StableSort_int64_Descending_262144 3680806 ns 7494323 ns
BM_StableSort_int64_Descending_524288 11023784 ns 24935033 ns
BM_StableSort_int64_Descending_1048576 20023568 ns 33220712 ns
BM_StableSort_int64_SingleElement_1 3.37 ns 3.98 ns
BM_StableSort_int64_SingleElement_4 5.32 ns 6.92 ns
BM_StableSort_int64_SingleElement_16 10.9 ns 13.3 ns
BM_StableSort_int64_SingleElement_64 32.1 ns 43.8 ns
BM_StableSort_int64_SingleElement_256 420 ns 541 ns
BM_StableSort_int64_SingleElement_1024 5689 ns 3381 ns
BM_StableSort_int64_SingleElement_4096 19199 ns 17989 ns
BM_StableSort_int64_SingleElement_16384 75754 ns 91963 ns
BM_StableSort_int64_SingleElement_65536 357106 ns 500326 ns
BM_StableSort_int64_SingleElement_262144 1672975 ns 2417734 ns
BM_StableSort_int64_SingleElement_524288 3642891 ns 5200878 ns
BM_StableSort_int64_SingleElement_1048576 11172007 ns 13729511 ns
BM_StableSort_int64_PipeOrgan_1 3.38 ns 3.94 ns
BM_StableSort_int64_PipeOrgan_4 5.73 ns 6.44 ns
BM_StableSort_int64_PipeOrgan_16 27.5 ns 29.0 ns
BM_StableSort_int64_PipeOrgan_64 310 ns 321 ns
BM_StableSort_int64_PipeOrgan_256 2761 ns 2918 ns
BM_StableSort_int64_PipeOrgan_1024 16105 ns 12525 ns
BM_StableSort_int64_PipeOrgan_4096 65289 ns 59990 ns
BM_StableSort_int64_PipeOrgan_16384 341757 ns 270636 ns
BM_StableSort_int64_PipeOrgan_65536 587452 ns 1126132 ns
BM_StableSort_int64_PipeOrgan_262144 2837955 ns 5034180 ns
BM_StableSort_int64_PipeOrgan_524288 6617313 ns 15267354 ns
BM_StableSort_int64_PipeOrgan_1048576 15208796 ns 23162989 ns
BM_StableSort_int64_QuickSortAdversary_1 3.77 ns 3.45 ns
BM_StableSort_int64_QuickSortAdversary_4 5.55 ns 5.20 ns
BM_StableSort_int64_QuickSortAdversary_16 12.5 ns 11.5 ns
BM_StableSort_int64_QuickSortAdversary_64 646 ns 750 ns
BM_StableSort_int64_QuickSortAdversary_256 2655 ns 3539 ns
BM_StableSort_int64_QuickSortAdversary_1024 16373 ns 22349 ns
BM_StableSort_int64_QuickSortAdversary_4096 62306 ns 97248 ns
BM_StableSort_int64_QuickSortAdversary_16384 321755 ns 388084 ns
BM_StableSort_int64_QuickSortAdversary_65536 1374694 ns 1596091 ns
BM_StableSort_int64_QuickSortAdversary_262144 4374661 ns 6894139 ns
BM_StableSort_int64_QuickSortAdversary_524288 12736074 ns 23932229 ns
BM_StableSort_int64_QuickSortAdversary_1048576 22615219 ns 33355629 ns
BM_StableSort_uint64_Random_1 3.82 ns 3.49 ns
BM_StableSort_uint64_Random_4 22.4 ns 23.4 ns
BM_StableSort_uint64_Random_16 154 ns 146 ns
BM_StableSort_uint64_Random_64 924 ns 926 ns
BM_StableSort_uint64_Random_256 5864 ns 5913 ns
BM_StableSort_uint64_Random_1024 7168 ns 31746 ns
BM_StableSort_uint64_Random_4096 27668 ns 154224 ns
BM_StableSort_uint64_Random_16384 219526 ns 755205 ns
BM_StableSort_uint64_Random_65536 965251 ns 3490165 ns
BM_StableSort_uint64_Random_262144 6262162 ns 15889589 ns
BM_StableSort_uint64_Random_524288 12530078 ns 36458581 ns
BM_StableSort_uint64_Random_1048576 38462191 ns 75168445 ns
BM_StableSort_uint64_Ascending_1 3.30 ns 3.35 ns
BM_StableSort_uint64_Ascending_4 5.65 ns 5.84 ns
BM_StableSort_uint64_Ascending_16 14.7 ns 12.6 ns
BM_StableSort_uint64_Ascending_64 55.3 ns 34.6 ns
BM_StableSort_uint64_Ascending_256 513 ns 533 ns
BM_StableSort_uint64_Ascending_1024 5541 ns 3189 ns
BM_StableSort_uint64_Ascending_4096 17706 ns 20326 ns
BM_StableSort_uint64_Ascending_16384 66420 ns 93757 ns
BM_StableSort_uint64_Ascending_65536 341425 ns 435016 ns
BM_StableSort_uint64_Ascending_262144 1595691 ns 2088317 ns
BM_StableSort_uint64_Ascending_524288 3808703 ns 5092832 ns
BM_StableSort_uint64_Ascending_1048576 11060417 ns 13023250 ns
BM_StableSort_uint64_Descending_1 3.29 ns 3.35 ns
BM_StableSort_uint64_Descending_4 8.65 ns 7.92 ns
BM_StableSort_uint64_Descending_16 54.7 ns 80.2 ns
BM_StableSort_uint64_Descending_64 1028 ns 1307 ns
BM_StableSort_uint64_Descending_256 4521 ns 5635 ns
BM_StableSort_uint64_Descending_1024 7122 ns 23323 ns
BM_StableSort_uint64_Descending_4096 30538 ns 95892 ns
BM_StableSort_uint64_Descending_16384 195565 ns 392721 ns
BM_StableSort_uint64_Descending_65536 852002 ns 1720358 ns
BM_StableSort_uint64_Descending_262144 3737884 ns 7484130 ns
BM_StableSort_uint64_Descending_524288 11159345 ns 25690770 ns
BM_StableSort_uint64_Descending_1048576 20648864 ns 33057383 ns
BM_StableSort_uint64_SingleElement_1 3.62 ns 4.10 ns
BM_StableSort_uint64_SingleElement_4 6.73 ns 6.64 ns
BM_StableSort_uint64_SingleElement_16 14.9 ns 11.3 ns
BM_StableSort_uint64_SingleElement_64 52.0 ns 33.0 ns
BM_StableSort_uint64_SingleElement_256 511 ns 582 ns
BM_StableSort_uint64_SingleElement_1024 6499 ns 3287 ns
BM_StableSort_uint64_SingleElement_4096 22190 ns 17616 ns
BM_StableSort_uint64_SingleElement_16384 84378 ns 86885 ns
BM_StableSort_uint64_SingleElement_65536 466257 ns 457144 ns
BM_StableSort_uint64_SingleElement_262144 1993687 ns 2361999 ns
BM_StableSort_uint64_SingleElement_524288 4759565 ns 5096771 ns
BM_StableSort_uint64_SingleElement_1048576 12426111 ns 13468453 ns
BM_StableSort_uint64_PipeOrgan_1 3.73 ns 3.94 ns
BM_StableSort_uint64_PipeOrgan_4 7.18 ns 7.54 ns
BM_StableSort_uint64_PipeOrgan_16 25.2 ns 29.1 ns
BM_StableSort_uint64_PipeOrgan_64 260 ns 321 ns
BM_StableSort_uint64_PipeOrgan_256 2468 ns 2970 ns
BM_StableSort_uint64_PipeOrgan_1024 7025 ns 12912 ns
BM_StableSort_uint64_PipeOrgan_4096 28968 ns 53379 ns
BM_StableSort_uint64_PipeOrgan_16384 194156 ns 239790 ns
BM_StableSort_uint64_PipeOrgan_65536 599491 ns 993800 ns
BM_StableSort_uint64_PipeOrgan_262144 2648585 ns 4689680 ns
BM_StableSort_uint64_PipeOrgan_524288 7621109 ns 15401808 ns
BM_StableSort_uint64_PipeOrgan_1048576 15608814 ns 23484821 ns
BM_StableSort_uint64_QuickSortAdversary_1 3.38 ns 3.54 ns
BM_StableSort_uint64_QuickSortAdversary_4 5.50 ns 6.03 ns
BM_StableSort_uint64_QuickSortAdversary_16 14.2 ns 11.0 ns
BM_StableSort_uint64_QuickSortAdversary_64 597 ns 688 ns
BM_StableSort_uint64_QuickSortAdversary_256 2446 ns 2818 ns
BM_StableSort_uint64_QuickSortAdversary_1024 7266 ns 20319 ns
BM_StableSort_uint64_QuickSortAdversary_4096 31155 ns 89112 ns
BM_StableSort_uint64_QuickSortAdversary_16384 201033 ns 390574 ns
BM_StableSort_uint64_QuickSortAdversary_65536 871014 ns 1685639 ns
BM_StableSort_uint64_QuickSortAdversary_262144 3978535 ns 7265830 ns
BM_StableSort_uint64_QuickSortAdversary_524288 10279721 ns 25350004 ns
BM_StableSort_uint64_QuickSortAdversary_1048576 20256585 ns 33054393 ns
```
2025-01-09 19:02:35 +01:00
Vitaly Buka
570f03096a
Revert "Reapply "[libc++] Explicitly convert to masks in SIMD code ( #107983 )"" ( #122022 )
...
Reverts llvm/llvm-project#121352
Triggers "vector type should not be a bool!" on:
```
bool a[100];
bool b[100];
auto t = std::mismatch(std::begin(a), std::end(a), std::begin(b), std::end(b));
```
https://godbolt.org/z/Y73s3sdef
2025-01-08 18:14:39 +01:00
Nikolas Klauser
f69585235e
[libc++] Put _LIBCPP_NODEBUG on all internal aliases ( #118710 )
...
This significantly reduces the amount of debug information generated
for codebases using libc++, without hurting the debugging experience.
2025-01-08 11:12:59 -05:00
Vitaly Buka
ed572f2003
Reapply "[libc++] Explicitly convert to masks in SIMD code ( #107983 )" ( #121352 )
...
This reverts commit 0ea40bf02138c02e7680ce6fa8169502f2a8bd42.
Passes with https://github.com/llvm/llvm-project/issues/121365 fix:
https://lab.llvm.org/buildbot/#/builders/55/builds/4930
2025-01-01 11:30:35 +01:00
Nikolas Klauser
b905bcc509
[libc++] Remove some unused includes ( #120219 )
2024-12-18 21:10:27 +01:00
Nikolas Klauser
59890c1334
[libc++] Granularize <new> includes ( #119964 )
2024-12-17 11:29:16 +01:00
Nikolas Klauser
a2042521a0
[libc++] Remove _AlgPolicy from std::copy and algorithms using std::copy ( #115887 )
...
`std::copy` doesn't use the `_AlgPolicy` for anything other than calling
itself with it, so we can just remove the argument. This also removes
the need in a few other algorithms which had an `_AlgPolicy` argument
only to call `copy`.
2024-11-12 23:03:52 +01:00
Nikolas Klauser
5b67372aec
[libc++] Remove a few unused includes from <__algorithm/find_end.h>
2024-11-12 22:11:15 +01:00
Nikolas Klauser
eab7be5d42
[libc++] Forward more algorithms to the classic algorithms ( #114674 )
...
This partially addresses #105687 .
2024-11-06 12:10:06 +01:00
Nikolas Klauser
c6f3b7bcd0
[libc++] Refactor the configuration macros to being always defined ( #112094 )
...
This is a follow-up to #89178 . This updates the `<__config_site>`
macros.
2024-11-06 10:39:19 +01:00
Nikolas Klauser
e99c4906e4
[libc++] Granularize <cstddef> includes ( #108696 )
2024-10-31 02:20:10 +01:00
Nikolas Klauser
0fb76bae6b
Reapply "[libc++] Simplify the implementation of std::sort a bit ( #104902 )" ( #114023 )
...
This reverts commit ef44e4659878f2. The patch was originally reverted
because it was
deemed to introduce a performance regression for small inputs, however
it also fixed
a previous performance regression for larger inputs. So overall, this
patch is desirable.
2024-10-30 11:51:55 +01:00
Louis Dionne
8e6bba230e
[libc++][NFC] Rename fold.h to ranges_fold.h ( #109696 )
...
This follows the pattern we use consistently for ranges algorithms.
This is a re-application of 24bc3244d4e which had been reverted in
f11abac65 due to unrelated failures.
2024-09-30 08:30:16 -04:00
Chris B
f11abac652
Revert "[libc++][modules] Rewrite the modulemap to have fewer top-level modules ( #107638 )" ( #110384 )
...
This reverts 3 commits:
45a09d1811d5d6597385ef02ecf2d4b7320c37c5
24bc3244d4e221f4e6740f45e2bf15a1441a3076
bc6bd3bc1e99c7ec9e22dff23b4f4373fa02cae3
The GitHub pre-merge CI has been broken since this PR went in. This
change reverts it to see if I can get the pre-merge CI working again.
2024-09-28 21:47:09 -05:00
Louis Dionne
24bc3244d4
[libc++][NFC] Rename fold.h to ranges_fold.h ( #109696 )
...
This follows the pattern we use consistently for ranges algorithms.
2024-09-27 01:02:21 -04:00
Louis Dionne
ef44e46598
Revert "[libc++] Simplify the implementation of std::sort a bit ( #104902 )"
...
This reverts commit d4ffccfce103b01401b8a9222e373f2d404f8439, which
caused a performance regression that needs to be investigated further.
2024-09-20 09:48:58 -04:00
Thurston Dang
0ea40bf021
Revert "[libc++] Explicitly convert to masks in SIMD code ( #107983 )"
...
This reverts commit 1603f99a37c5b179a21dbb8000c39a471a950927.
Reason: buildbot breakage e.g., https://lab.llvm.org/buildbot/#/builders/55/builds/2061
llvm-libc++-shared.cfg.in :: std/algorithms/alg.nonmodifying/alg.starts_with/ranges.starts_with.pass.cpp
llvm-libc++-shared.cfg.in :: std/algorithms/alg.nonmodifying/mismatch/mismatch.pass.cpp
llvm-libc++-shared.cfg.in :: std/algorithms/alg.nonmodifying/mismatch/ranges_mismatch.pass.cpp
...
(Buildbot re-run passed with the previous revision, 1fc288bf481726393c73133eef9aa73c0f78312e)
2024-09-17 21:52:33 +00:00
Nikolas Klauser
1603f99a37
[libc++] Explicitly convert to masks in SIMD code ( #107983 )
...
This makes it clearer when we use masks and avoids MSan complaining.
2024-09-17 12:04:54 +02:00
Louis Dionne
09e3a36058
[libc++][modules] Fix missing and incorrect includes ( #108850 )
...
This patch adds a large number of missing includes in the libc++ headers
and the test suite. Those were found as part of the effort to move
towards a mostly monolithic top-level std module.
2024-09-16 15:06:20 -04:00
A. Jiang
94e7c0b051
[libc++] Remove get_temporary_buffer and return_temporary_buffer ( #100914 )
...
Works towards P0619R4 / #99985 .
The use of `std::get_temporary_buffer` and `std::return_temporary_buffer`
are replaced with `unique_ptr`-based RAII buffer holder.
Escape hatches:
- `_LIBCPP_ENABLE_CXX20_REMOVED_TEMPORARY_BUFFER` restores
`std::get_temporary_buffer` and `std::return_temporary_buffer`.
Drive-by changes:
- In `<syncstream>`, states that `get_temporary_buffer` is now removed,
because `<syncstream>` is added in C++20.
2024-09-16 11:53:05 -04:00
Nikolas Klauser
17e0686ab1
[libc++][NFC] Use [[__nodiscard__]] unconditionally ( #80454 )
...
`__has_cpp_attribute(__nodiscard__)` is always true now, so we might as
well replace `_LIBCPP_NODISCARD`. It's one less macro that can result in
bad diagnostics.
2024-09-12 21:18:43 +02:00
Louis Dionne
d6832a611a
[libc++][modules] Modularize <cstddef> ( #107254 )
...
Many headers include `<cstddef>` just for size_t, and pulling in
additional content (e.g. the traits used for std::byte) is unnecessary.
To solve this problem, this patch splits up `<cstddef>` into
subcomponents so that headers can include only the parts that they
actually require.
This has the added benefit of making the modules build a lot stricter
with respect to IWYU, and also providing a canonical location where we
define `std::size_t` and friends (which were previously defined in
multiple headers like `<cstddef>` and `<ctime>`).
After this patch, there's still many places in the codebase where we
include `<cstddef>` when `<__cstddef/size_t.h>` would be sufficient.
This patch focuses on removing `<cstddef>` includes from __type_traits
to make these headers non-circular with `<cstddef>`. Additional
refactorings can be tackled separately.
2024-09-05 08:28:33 -04:00
Louis Dionne
0df78123fd
[libc++] Add missing include to three_way_comp_ref_type.h
...
We were using a `_LIBCPP_ASSERT_FOO` macro without including `<__assert>`.
rdar://134425695
2024-08-27 14:22:49 -04:00
Nikolas Klauser
d4ffccfce1
[libc++] Simplify the implementation of std::sort a bit ( #104902 )
...
This does a few things to canonicalize the library a bit. Specifically
- use `__desugars_to_v` instead of the custom `__is_simple_comparator`
- make `__use_branchless_sort` an inline variable
- remove the `_maybe_branchless` versions of the `__sortN` functions and
overload based on whether we can do branchless sorting instead.
2024-08-27 16:54:05 +02:00
Louis Dionne
f73050e722
[libc++] Fix several double-moves in the code base ( #104616 )
...
This patch hardens the "test iterators" we use to test algorithms by
ensuring that they don't get double-moved. As a result of this
hardening, the tests started reporting multiple failures where we would
double-move iterators, which are being fixed in this patch.
In particular:
- Fixed a double-move in pstl.partition
- Add coverage for begin()/end() in subrange tests
- Fix tests for ranges::ends_with and ranges::contains, which were
incorrectly calling begin() twice on the same subrange containing
non-copyable input iterators.
Fixes #100709
2024-08-20 14:36:11 -04:00
Louis Dionne
257831582c
[libc++] Check correctly ref-qualified __is_callable in algorithms ( #101553 )
...
We were only checking that the comparator was rvalue callable,
when in reality the algorithms always call comparators as lvalues.
This patch also refactors the tests for callable requirements and
expands it to a few missing algorithms.
This is take 2 of #73451 , which was reverted because it broke some
CI bots. The issue was that we checked __is_callable with arguments
in the wrong order inside std::upper_bound. This has now been fixed
and a test was added.
Fixes #69554
2024-08-05 11:23:06 -04:00
Nikolas Klauser
d07fdf9779
[libc++] Optimize lexicographical_compare ( #65279 )
...
If the comparison operation is equivalent to < and that is a total
order, we know that we can use equality comparison on that type instead
to extract some information. Furthermore, if equality comparison on that
type is trivial, the user can't observe that we're calling it. So
instead of using the user-provided total order, we use std::mismatch,
which uses equality comparison (and is vertorized). Additionally, if the
type is trivially lexicographically comparable, we can go one step
further and use std::memcmp directly instead of calling std::mismatch.
Benchmarks:
```
-------------------------------------------------------------------------------------
Benchmark old new
-------------------------------------------------------------------------------------
bm_lexicographical_compare<unsigned char>/1 1.17 ns 2.34 ns
bm_lexicographical_compare<unsigned char>/2 1.64 ns 2.57 ns
bm_lexicographical_compare<unsigned char>/3 2.23 ns 2.58 ns
bm_lexicographical_compare<unsigned char>/4 2.82 ns 2.57 ns
bm_lexicographical_compare<unsigned char>/5 3.34 ns 2.11 ns
bm_lexicographical_compare<unsigned char>/6 3.94 ns 2.21 ns
bm_lexicographical_compare<unsigned char>/7 4.56 ns 2.11 ns
bm_lexicographical_compare<unsigned char>/8 5.25 ns 2.11 ns
bm_lexicographical_compare<unsigned char>/16 9.88 ns 2.11 ns
bm_lexicographical_compare<unsigned char>/64 38.9 ns 2.36 ns
bm_lexicographical_compare<unsigned char>/512 317 ns 6.54 ns
bm_lexicographical_compare<unsigned char>/4096 2517 ns 41.4 ns
bm_lexicographical_compare<unsigned char>/32768 20052 ns 488 ns
bm_lexicographical_compare<unsigned char>/262144 159579 ns 4409 ns
bm_lexicographical_compare<unsigned char>/1048576 640456 ns 20342 ns
bm_lexicographical_compare<signed char>/1 1.18 ns 2.37 ns
bm_lexicographical_compare<signed char>/2 1.65 ns 2.60 ns
bm_lexicographical_compare<signed char>/3 2.23 ns 2.83 ns
bm_lexicographical_compare<signed char>/4 2.81 ns 3.06 ns
bm_lexicographical_compare<signed char>/5 3.35 ns 3.30 ns
bm_lexicographical_compare<signed char>/6 3.90 ns 3.99 ns
bm_lexicographical_compare<signed char>/7 4.56 ns 3.78 ns
bm_lexicographical_compare<signed char>/8 5.20 ns 4.02 ns
bm_lexicographical_compare<signed char>/16 9.80 ns 6.21 ns
bm_lexicographical_compare<signed char>/64 39.0 ns 3.16 ns
bm_lexicographical_compare<signed char>/512 318 ns 7.58 ns
bm_lexicographical_compare<signed char>/4096 2514 ns 47.4 ns
bm_lexicographical_compare<signed char>/32768 20096 ns 504 ns
bm_lexicographical_compare<signed char>/262144 156617 ns 4146 ns
bm_lexicographical_compare<signed char>/1048576 624265 ns 19810 ns
bm_lexicographical_compare<int>/1 1.15 ns 2.12 ns
bm_lexicographical_compare<int>/2 1.60 ns 2.36 ns
bm_lexicographical_compare<int>/3 2.21 ns 2.59 ns
bm_lexicographical_compare<int>/4 2.74 ns 2.83 ns
bm_lexicographical_compare<int>/5 3.26 ns 3.06 ns
bm_lexicographical_compare<int>/6 3.81 ns 4.53 ns
bm_lexicographical_compare<int>/7 4.41 ns 4.72 ns
bm_lexicographical_compare<int>/8 5.08 ns 2.36 ns
bm_lexicographical_compare<int>/16 9.54 ns 3.08 ns
bm_lexicographical_compare<int>/64 37.8 ns 4.71 ns
bm_lexicographical_compare<int>/512 309 ns 24.6 ns
bm_lexicographical_compare<int>/4096 2422 ns 204 ns
bm_lexicographical_compare<int>/32768 19362 ns 1947 ns
bm_lexicographical_compare<int>/262144 155727 ns 19793 ns
bm_lexicographical_compare<int>/1048576 623614 ns 80180 ns
bm_ranges_lexicographical_compare<unsigned char>/1 1.07 ns 2.35 ns
bm_ranges_lexicographical_compare<unsigned char>/2 1.72 ns 2.13 ns
bm_ranges_lexicographical_compare<unsigned char>/3 2.46 ns 2.12 ns
bm_ranges_lexicographical_compare<unsigned char>/4 3.17 ns 2.12 ns
bm_ranges_lexicographical_compare<unsigned char>/5 3.86 ns 2.12 ns
bm_ranges_lexicographical_compare<unsigned char>/6 4.55 ns 2.12 ns
bm_ranges_lexicographical_compare<unsigned char>/7 5.25 ns 2.12 ns
bm_ranges_lexicographical_compare<unsigned char>/8 5.95 ns 2.13 ns
bm_ranges_lexicographical_compare<unsigned char>/16 11.7 ns 2.13 ns
bm_ranges_lexicographical_compare<unsigned char>/64 45.5 ns 2.36 ns
bm_ranges_lexicographical_compare<unsigned char>/512 366 ns 6.35 ns
bm_ranges_lexicographical_compare<unsigned char>/4096 2886 ns 40.9 ns
bm_ranges_lexicographical_compare<unsigned char>/32768 23054 ns 489 ns
bm_ranges_lexicographical_compare<unsigned char>/262144 185302 ns 4339 ns
bm_ranges_lexicographical_compare<unsigned char>/1048576 741576 ns 19430 ns
bm_ranges_lexicographical_compare<signed char>/1 1.10 ns 2.12 ns
bm_ranges_lexicographical_compare<signed char>/2 1.66 ns 2.35 ns
bm_ranges_lexicographical_compare<signed char>/3 2.23 ns 2.58 ns
bm_ranges_lexicographical_compare<signed char>/4 2.82 ns 2.82 ns
bm_ranges_lexicographical_compare<signed char>/5 3.34 ns 3.06 ns
bm_ranges_lexicographical_compare<signed char>/6 3.92 ns 3.99 ns
bm_ranges_lexicographical_compare<signed char>/7 4.64 ns 4.10 ns
bm_ranges_lexicographical_compare<signed char>/8 5.21 ns 4.61 ns
bm_ranges_lexicographical_compare<signed char>/16 9.79 ns 7.42 ns
bm_ranges_lexicographical_compare<signed char>/64 38.9 ns 2.93 ns
bm_ranges_lexicographical_compare<signed char>/512 317 ns 7.31 ns
bm_ranges_lexicographical_compare<signed char>/4096 2500 ns 47.5 ns
bm_ranges_lexicographical_compare<signed char>/32768 19940 ns 496 ns
bm_ranges_lexicographical_compare<signed char>/262144 159166 ns 4393 ns
bm_ranges_lexicographical_compare<signed char>/1048576 638206 ns 19786 ns
bm_ranges_lexicographical_compare<int>/1 1.10 ns 2.12 ns
bm_ranges_lexicographical_compare<int>/2 1.64 ns 3.04 ns
bm_ranges_lexicographical_compare<int>/3 2.23 ns 2.58 ns
bm_ranges_lexicographical_compare<int>/4 2.81 ns 2.81 ns
bm_ranges_lexicographical_compare<int>/5 3.35 ns 3.05 ns
bm_ranges_lexicographical_compare<int>/6 3.94 ns 4.60 ns
bm_ranges_lexicographical_compare<int>/7 4.60 ns 4.81 ns
bm_ranges_lexicographical_compare<int>/8 5.19 ns 2.35 ns
bm_ranges_lexicographical_compare<int>/16 9.85 ns 2.87 ns
bm_ranges_lexicographical_compare<int>/64 38.9 ns 4.70 ns
bm_ranges_lexicographical_compare<int>/512 318 ns 24.5 ns
bm_ranges_lexicographical_compare<int>/4096 2494 ns 202 ns
bm_ranges_lexicographical_compare<int>/32768 20000 ns 1939 ns
bm_ranges_lexicographical_compare<int>/262144 160433 ns 19730 ns
bm_ranges_lexicographical_compare<int>/1048576 642636 ns 80760 ns
```
2024-08-04 10:02:43 +02:00
Louis Dionne
451bba6fbf
[libc++] Revert "Check correctly ref-qualified __is_callable in algorithms ( #73451 )"
...
This reverts commit 8d151f804ff43aaed1edf810bb2a07607b8bba14, which
broke some build bots. I think that is caused by an invalid argument
order when checking __is_comparable in upper_bound.
2024-08-01 15:56:06 -04:00
Nhat Nguyen
8d151f804f
[libc++] Check correctly ref-qualified __is_callable in algorithms ( #73451 )
...
We were only checking that the comparator was rvalue callable,
when in reality the algorithms always call comparators as lvalues.
This patch also refactors the tests for callable requirements and
expands it to a few missing algorithms.
Fixes #69554
2024-08-01 14:08:21 -04:00
Christopher Di Bella
d10dc5a06f
[libc++] Remove dedicated namespaces for ranges functions ( #76543 )
...
We originally put implementation-detail function objects into individual
namespaces for `std::ranges` without a good reason for doing so. This
practice was continued, presumably because there was prior art. Since
there's no reason to keep these namespaces, this commit removes them,
which will slightly impact binary size.
This commit does not apply to CPOs, some of which need additional work.
2024-08-01 08:54:06 -04:00
Hewill Kang
5b6b48800e
[libc++][NFC] Remove two unused implementation details __find_end ( #100685 )
...
Those two `__find_end` functions are no longer used after 101d1e9b3c86.
After that commit, `std::find_end` started dispatching to `__find_end_classic`,
and `ranges::find_end` to `__find_end_impl`, which means that the two `__find_end`
functions were no longer necessary.
Fixes #100569
2024-07-31 10:34:19 -04:00
nicole mazzuca
04760bfadb
[libc++][ranges] P1223R5: find_last ( #99312 )
...
Implements [P1223R5][] completely.
Includes an implementation of `find_last`, `find_last_if`, and
`find_last_if_not`.
[P1223R5]: https://wg21.link/p1223r5
2024-07-19 09:42:16 -07:00
Iuri Chaer
a0662176a9
[libc++] Speed up set_intersection() by fast-forwarding over ranges of non-matching elements with one-sided binary search. ( #75230 )
...
One-sided binary search, aka meta binary search, has been in the public
domain for decades, and has the general advantage of being constant time
in the best case, with the downside of executing at most 2*log(N)
comparisons vs classic binary search's exact log(N). There are two
scenarios in which it really shines: the first one is when operating
over non-random-access iterators, because the classic algorithm requires
knowing the container's size upfront, which adds N iterator increments
to the complexity. The second one is when traversing the container in
order, trying to fast-forward to the next value: in that case the
classic algorithm requires at least O(N*log(N)) comparisons and, for
non-random-access iterators, O(N^2) iterator increments, whereas the
one-sided version will yield O(N) operations on both counts, with a
best-case of O(log(N)) comparisons which is very common in practice.
2024-07-18 16:11:24 -04:00
Louis Dionne
e2c2ffbe7a
[libc++][NFC] Run clang-format on libcxx/include again ( #95874 )
...
As time went by, a few files have become mis-formatted w.r.t.
clang-format. This was made worse by the fact that formatting was not
being enforced in extensionless headers. This commit simply brings all
of libcxx/include in-line with clang-format again.
We might have to do this from time to time as we update our clang-format
version, but frankly this is really low effort now that we've formatted
everything once.
2024-06-18 09:13:45 -04:00
Louis Dionne
acb896a344
[libc++] Remove unnecessary #ifdef guards around PSTL implementation details ( #95268 )
...
We want the PSTL implementation details to be available regardless of
the Standard mode or whether the experimental PSTL is enabled. This
patch guards the inclusion of the PSTL to the top-level headers that
define the public API in `__numeric` and `__algorithm`.
2024-06-12 17:25:43 -04:00
Louis Dionne
fe4cd104a8
[libc++][NFC] Fix typo in concept PSTL concept check
2024-06-12 12:31:05 -04:00
Louis Dionne
9540950a45
[libc++] Overhaul the PSTL dispatching mechanism ( #88131 )
...
The experimental PSTL's current dispatching mechanism was designed with
flexibility in mind. However, while reviewing the in-progress OpenMP
backend, I realized that the dispatching mechanism based on ADL and
default definitions in the frontend had several downsides. To name a
few:
1. The dispatching of an algorithm to the back-end and its default
implementation is bundled together via `_LIBCPP_PSTL_CUSTOMIZATION_POINT`.
This makes the dispatching really confusing and leads to annoyances
such as variable shadowing and weird lambda captures in the front-end.
2. The distinction between back-end functions and front-end algorithms
is not as clear as it could be, which led us to call one where we meant
the other in a few cases. This is bad due to the exception requirements
of the PSTL: calling a front-end algorithm inside the implementation of
a back-end is incorrect for exception-safety.
3. There are two levels of back-end dispatching in the PSTL, which treat
CPU backends as a special case. This was confusing and not as flexible
as we'd like. For example, there was no straightforward way to dispatch
all uses of `unseq` to a specific back-end from the OpenMP backend,
or for CPU backends to fall back on each other.
This patch rewrites the backend dispatching mechanism to solve these
problems, but doesn't touch any of the actual implementation of
algorithms. Specifically, this rewrite has the following
characteristics:
- There is a single level of backend dispatching, however partial backends can
be stacked to provide a full implementation of the PSTL. The two-level dispatching
that was used for CPU-based backends is handled by providing CPU-based basis
operations as simple helpers that can easily be reused when defining any PSTL
backend.
- The default definitions for algorithms are separated from their dispatching logic.
- The front-end is thus simplified a whole lot and made very consistent
for all algorithms, which makes it easier to audit the front-end for
things like exception-correctness, appropriate forwarding, etc.
Fixes #70718
2024-06-12 12:24:34 -04:00
Zibi Sarbinowski
ffc3a6b286
[libc++] Fix endianness for algorithm mismatch ( #93082 )
...
This PR is required to fix
`std/algorithms/alg.nonmodifying/mismatch/mismatch.pass.cpp` test for
big endian platrofrms such as z/OS.
2024-06-11 08:29:12 -04:00
Louis Dionne
e406d5ed9c
[libc++][pstl] Merge all frontend functions for the PSTL ( #89219 )
...
This is an intermediate step towards the PSTL dispatching mechanism
rework. It will make it a lot easier to track the upcoming front-end
changes. After the rework, there are basically no implementation details
in the front-end, so the definition of each algorithm will become much
simpler. Otherwise, it wouldn't make sense to define all the algorithms
in the same header.
2024-05-27 17:51:12 -04:00
Louis Dionne
72417920d3
[libc++] Remove a few unused includes of trivially_copyable.h ( #93200 )
2024-05-23 15:58:51 -04:00
Louis Dionne
bd3f5a4bd3
[libc++][pstl] Improve exception handling ( #88998 )
...
There were various places where we incorrectly handled exceptions in the
PSTL. Typical issues were missing `noexcept` and taking iterators by
value instead of by reference.
This patch fixes those inconsistent and incorrect instances, and adds
proper tests for all of those. Note that the previous tests were often
incorrectly turned into no-ops by the compiler due to copy ellision,
which doesn't happen with these new tests.
2024-05-22 12:39:21 -07:00
zibi2
af57ad6536
[libc++][z/OS] Correct a definition of __native_vector_size ( #91995 )
...
Fix `std/ranges/range.adaptors/range.lazy.split/general.pass.cpp` which
started failing on z/OS after this
[commit](https://github.com/llvm/llvm-project/commit/985c1a44f8d49e0af ).
This test case is passing on other platforms such as AIX. This is
because the `__ALTIVEC__` macro is defined and `__mismatch` under
`_LIBCPP_VECTORIZE_ALGORITHMS` guard is compiled out. However, on z/OS
`_LIBCPP_VECTORIZE_ALGORITHMS` is defined. Analyzing the algorithm of
`__mismatch` shows that the culprit is the definition of
`__native_vector_size` which was defined wrongly as 1. This PR corrects
the definition of `__native_vector_size` and fixes the affected test.
2024-05-16 08:58:44 -04:00
Nikolas Klauser
05cc2d5fe1
[libc++] Vectorize std::mismatch with trivially equality comparable types ( #87716 )
2024-05-11 23:32:48 +02:00
Nikolas Klauser
840032419d
[libc++][NFC] Rename __find_impl to __find ( #90163 )
...
For most algorithms we've just added underscores to the detail function.
This changes `std::find` to match that pattern.
2024-04-27 09:51:59 +02:00
Nikolas Klauser
83bc7b5771
[libc++] Remove _LIBCPP_DISABLE_NODISCARD_EXTENSIONS and refactor the tests ( #87094 )
...
This also adds a few tests that were missing.
2024-04-22 22:13:58 +02:00
Louis Dionne
0e08bce142
[libc++][pstl] Move the CPU algorithm implementations to __pstl ( #89109 )
...
This colocates the CPU algorithms closer to the rest of the PSTL
implementation details.
2024-04-18 07:50:57 -04:00
Louis Dionne
d423d80e56
[libc++][pstl] Promote CPU backends to top-level backends ( #88968 )
...
This patch removes the two-level backend dispatching mechanism we had in
the PSTL. Instead of selecting both a PSTL backend and a PSTL CPU
backend, we now only select a top-level PSTL backend. This greatly
simplifies the PSTL configuration layer.
While this patch technically removes some flexibility from the PSTL
configuration mechanism because CPU backends are not considered
separately, it opens the door to a much more powerful configuration
mechanism based on chained backends in a follow-up patch.
This is a step towards overhauling the PSTL dispatching mechanism.
2024-04-17 13:36:53 -04:00
Louis Dionne
d57907d0b4
[libc++] Add missing iterator requirement checks in the PSTL ( #88127 )
...
Also add tests for those, and add a few missing requirements to testing
iterators in the test suite.
2024-04-17 08:21:48 -04:00
Louis Dionne
5b811562a5
[libc++] Rename __cpu_traits functions ( #88741 )
...
Functions inside __cpu_traits were needlessly prefixed with __parallel,
which doesn't serve a real purpose anymore now that they are inside a
traits class.
2024-04-16 10:33:39 +02:00
Louis Dionne
a3ce29f7bb
[libc++][PSTL] Introduce cpu traits ( #88134 )
...
Currently, CPU backends in the PSTL are created by defining functions
in the __par_backend namespace. Then, the PSTL includes the CPU backend
that gets configured via CMake and gets those definitions.
This prevents CPU backends from easily co-existing and is a bit
confusing.
To solve this problem, this patch introduces the notion of __cpu_traits,
which is a cheap encapsulation of the basis operations required to
implement a CPU-based PSTL. Different backends can now define their own
tag and coexist, and the CPU-based PSTL will simply use __cpu_traits to
dispatch to the right implementation of e.g. __for_each.
Note that this patch doesn't change the actual implementation of the
backends in any way, it only modifies how that implementation is
accessed
to implement PSTL algorithms.
This patch is a step towards #88131 .
2024-04-15 10:30:00 -04:00
