[PATCH v3 2/2] powerpc: Add optimized stpncpy for POWER9
Raphael M Zinsly
rzinsly@linux.ibm.com
Tue Sep 29 15:23:21 GMT 2020
generic_stpncpy __stpncpy_power9
__stpncpy_power8 __stpncpy_power7 __stpncpy_ppc
Length 16, n 16, alignment 1/ 1: 7.04141 2.66905 2.71071 5.33257
9.45193
Length 16, n 16, alignment 1/ 1: 7.01728 2.54349 2.70763 5.35555
9.40601
Length 16, n 16, alignment 1/ 2: 6.76331 2.56894 2.70649 5.28715
9.19534
Length 16, n 16, alignment 2/ 1: 6.41285 2.52953 2.86392 5.25868
9.24343
Length 2, n 4, alignment 7/ 2: 7.76627 2.36037 4.34749 4.05757
8.45648
Length 4, n 2, alignment 2/ 7: 6.15257 1.734 2.66932 2.81884 6.61486
Length 2, n 4, alignment 7/ 2: 7.69004 2.34779 3.90224 4.08693
8.51617
Length 4, n 2, alignment 2/ 7: 6.14888 1.73738 2.66929 2.81777
6.39066
Length 16, n 16, alignment 2/ 2: 7.25765 2.5434 2.8759 4.7084 9.43171
Length 16, n 16, alignment 2/ 2: 6.41274 2.52681 2.87939 5.2894 9.2505
Length 16, n 16, alignment 2/ 4: 6.74797 2.6683 2.82869 5.27608 9.43391
Length 16, n 16, alignment 4/ 2: 7.6281 2.54368 3.52982 5.26862 8.7369
Length 4, n 8, alignment 6/ 4: 7.79233 2.33099 5.64785 4.21131 9.03
Length 8, n 4, alignment 4/ 6: 6.01824 1.73782 2.81779 2.81777
7.90004
Length 4, n 8, alignment 6/ 4: 7.94851 2.33098 4.90456 3.75698
8.89379
Length 8, n 4, alignment 4/ 6: 6.0183 1.73715 2.81777 2.41521 7.83867
Length 16, n 16, alignment 3/ 3: 6.93178 2.66854 3.22004 5.31673
9.09542
Length 16, n 16, alignment 3/ 3: 6.99998 2.67084 3.22862 5.48294 9.2366
Length 16, n 16, alignment 3/ 6: 7.14689 2.6615 3.21888 5.25964 9.1277
Length 16, n 16, alignment 6/ 3: 6.46654 2.65885 4.57873 5.25391
7.75507
Length 8, n 16, alignment 5/ 6: 7.37286 2.33316 3.92971 4.50331
10.1496
Length 16, n 8, alignment 6/ 5: 5.73663 1.87991 2.633 4.09291 5.91732
Length 8, n 16, alignment 5/ 6: 7.77512 2.33361 3.67636 4.50091 10.147
Length 16, n 8, alignment 6/ 5: 5.73662 1.88001 2.57119 4.10496
6.15016
Length 16, n 16, alignment 4/ 4: 7.55115 2.65827 3.5838 5.25628 8.81586
Length 16, n 16, alignment 4/ 4: 7.61232 2.66851 3.62508 5.32044
8.73914
Length 16, n 16, alignment 4/ 0: 7.54588 2.54345 3.48987 5.27812
8.77989
Length 16, n 16, alignment 0/ 4: 6.82387 1.88425 2.41569 5.27746
7.19847
Length 16, n 32, alignment 4/ 0: 10.1135 3.10868 6.01894 6.66693
11.7681
Length 32, n 16, alignment 0/ 4: 6.93527 1.8793 2.4162 5.29155 6.50752
Length 16, n 32, alignment 4/ 0: 10.1565 3.16134 5.78062 6.81425
11.2226
Length 32, n 16, alignment 0/ 4: 6.76758 1.87928 2.41649 5.30161
7.22291
Length 16, n 16, alignment 5/ 5: 7.22753 2.56593 4.22659 5.30415
9.86703
Length 16, n 16, alignment 5/ 5: 6.76256 2.54348 4.23108 5.43866
9.53557
Length 16, n 16, alignment 5/ 2: 7.23702 2.52833 4.23011 5.26711
9.52126
Length 16, n 16, alignment 2/ 5: 6.68084 2.66311 2.84314 5.2709 9.24495
Length 32, n 64, alignment 3/ 2: 12.4989 3.84198 6.40671 10.4545 14.317
Length 64, n 32, alignment 2/ 3: 10.1464 2.78457 3.17933 7.67569
12.4356
Length 32, n 64, alignment 3/ 2: 12.4991 3.83968 7.08471 10.451 15.8984
Length 64, n 32, alignment 2/ 3: 9.61285 2.78401 3.18834 7.66606
13.9602
Length 16, n 16, alignment 6/ 6: 7.24557 2.66839 4.55951 5.25563 7.7369
Length 16, n 16, alignment 6/ 6: 6.76327 2.65836 4.5127 5.63264 7.80333
Length 16, n 16, alignment 6/ 4: 7.15127 2.54397 4.57355 5.32957
7.51005
Length 16, n 16, alignment 4/ 6: 7.51733 2.5615 3.67299 5.31244 8.73893
Length 64, n 128, alignment 2/ 4: 14.0745 4.98021 7.33878 11.4384
17.1572
Length 128, n 64, alignment 4/ 2: 11.7179 3.59088 4.89414 10.2021
11.6637
Length 64, n 128, alignment 2/ 4: 14.0841 4.99105 7.28507 11.4365
21.7537
Length 128, n 64, alignment 4/ 2: 11.7142 3.59211 4.83864 9.87632
19.4664
Length 16, n 16, alignment 7/ 7: 7.12738 2.53533 5.62213 5.30017
7.90888
Length 16, n 16, alignment 7/ 7: 6.82635 2.53529 5.60694 5.27111
8.88482
Length 16, n 16, alignment 7/ 6: 6.9193 2.54376 5.48117 5.24785 8.04263
Length 16, n 16, alignment 6/ 7: 6.89261 2.55078 4.51003 5.32471
7.81768
Length 128, n 256, alignment 1/ 6: 16.2686 7.68983 9.35727 16.2843
19.8458
Length 256, n 128, alignment 6/ 1: 13.4356 4.94899 7.94404 15.0122
15.0231
Length 128, n 256, alignment 1/ 6: 16.2511 7.69025 9.35528 16.2859
37.8453
Length 256, n 128, alignment 6/ 1: 13.4332 4.94446 8.02757 12.2879
34.1949
Length 8, n 16, alignment 0/ 0: 7.26102 2.33285 3.75702 3.85762
7.72869
Length 32, n 16, alignment 0/ 0: 7.049 1.88689 2.42187 2.41537 6.58192
Length 8, n 16, alignment 7/ 2: 8.09344 2.31269 3.67403 4.31612
8.21018
Length 32, n 16, alignment 7/ 2: 6.822 2.45733 5.59593 5.33252 6.53496
Length 16, n 32, alignment 0/ 0: 9.99648 3.36432 4.70547 4.55746 10.148
Length 64, n 32, alignment 0/ 0: 7.89408 2.4309 2.58854 2.70519 8.89171
Length 16, n 32, alignment 6/ 4: 9.31969 3.15547 7.24937 9.47362
10.0091
Length 64, n 32, alignment 6/ 4: 9.91687 2.78234 4.64259 7.00062
10.5972
Length 32, n 64, alignment 0/ 0: 11.0651 3.81484 4.4379 4.91663 11.8363
Length 128, n 64, alignment 0/ 0: 9.25821 3.20129 3.55296 4.22664
9.63556
Length 32, n 64, alignment 5/ 6: 12.5097 3.83422 7.29892 9.09849
13.2517
Length 128, n 64, alignment 5/ 6: 11.6165 3.60246 5.35542 8.90704
13.3207
Length 64, n 128, alignment 0/ 0: 12.372 4.91681 5.41951 6.91629 15.0813
Length 256, n 128, alignment 0/ 0: 7.93075 4.5247 6.29502 5.58357 12.5963
Length 64, n 128, alignment 4/ 0: 12.569 5.00092 7.25225 10.4764 15.9366
Length 256, n 128, alignment 4/ 0: 12.2963 4.90654 7.57109 12.0953
16.7672
Length 128, n 256, alignment 0/ 0: 13.9015 7.34814 7.88738 9.15353
19.4141
Length 512, n 256, alignment 0/ 0: 10.6865 6.52749 9.15011 9.71701
20.9021
Length 128, n 256, alignment 3/ 2: 16.3681 7.53318 9.89911 18.5309
20.8335
Length 512, n 256, alignment 3/ 2: 17.0249 7.10063 10.1568 22.6063
25.1262
Length 256, n 512, alignment 0/ 0: 16.5169 12.3406 13.6056 14.5875
29.2826
Length 1024, n 512, alignment 0/ 0: 16.3619 10.8422 16.7061 17.1025
37.7908
Length 256, n 512, alignment 2/ 4: 21.162 12.9621 14.3306 26.0856 30.0397
Length 1024, n 512, alignment 2/ 4: 25.5543 11.9978 17.7424 42.4293
47.7581
Length 512, n 1024, alignment 0/ 0: 20.5504 17.3132 19.5751 21.3633
42.7215
Length 2048, n 1024, alignment 0/ 0: 28.5197 19.3708 37.1801 35.3122
67.9792
Length 512, n 1024, alignment 1/ 6: 29.9875 17.7823 22.3228 47.3516
51.3697
Length 2048, n 1024, alignment 1/ 6: 42.9443 21.6004 38.7767 78.1732
83.9784
More information about the Libc-alpha
mailing list