Martin Storsjö 4664f5aa66 aarch64: Improve scheduling in sad_x3/sad_x4
Cortex A53    A72    A73
8 bpc:
Before:
sad_x3_4x4_neon:      580    303    204
sad_x3_4x8_neon:     1065    516    323
sad_x3_8x4_neon:      668    262    282
sad_x3_8x8_neon:     1238    454    471
sad_x3_8x16_neon:    2378    842    847
sad_x3_16x8_neon:    2136    738    776
sad_x3_16x16_neon:   4162   1378   1463
After:
sad_x3_4x4_neon:      477    298    206
sad_x3_4x8_neon:      842    515    327
sad_x3_8x4_neon:      603    260    279
sad_x3_8x8_neon:     1110    451    464
sad_x3_8x16_neon:    2125    841    843
sad_x3_16x8_neon:    2124    730    766
sad_x3_16x16_neon:   4145   1370   1434

10 bpc:
Before:
sad_x3_4x4_neon:      632    247    254
sad_x3_4x8_neon:     1162    419    443
sad_x3_8x4_neon:      890    358    416
sad_x3_8x8_neon:     1670    632    759
sad_x3_8x16_neon:    3230   1179   1458
sad_x3_16x8_neon:    3070   1209   1403
sad_x3_16x16_neon:   6030   2333   2699

After:
sad_x3_4x4_neon:      522    253    255
sad_x3_4x8_neon:      932    443    431
sad_x3_8x4_neon:      880    354    406
sad_x3_8x8_neon:     1660    626    736
sad_x3_8x16_neon:    3220   1170   1397
sad_x3_16x8_neon:    3060   1184   1362
sad_x3_16x16_neon:   6020   2272   2579

Thus, this is around a 20-25% speedup on Cortex A53 for the small
sizes (much smaller difference for bigger sizes though), while it
doesn't make much of a difference at all (mostly within measurement
noise) for the out-of-order cores (A72 and A73).
2023-11-02 13:27:08 +02:00
2021-09-29 21:11:16 +00:00
2023-10-24 22:07:14 +03:00
2023-01-28 16:37:02 +03:00
2023-01-28 16:37:02 +03:00
2023-01-28 16:37:02 +03:00
2023-01-28 16:37:02 +03:00
2018-01-17 18:31:04 +01:00
2021-09-29 21:11:16 +00:00
2023-01-28 16:37:02 +03:00
2023-10-10 09:00:09 +08:00
2023-10-10 09:00:09 +08:00
2004-06-03 19:29:33 +00:00
2023-01-28 16:37:02 +03:00
2015-08-19 01:00:22 +03:00
2023-04-02 15:51:50 +03:00
2023-01-28 16:37:02 +03:00
2023-01-28 16:37:02 +03:00
2023-01-28 16:37:02 +03:00
S
Description
No description provided
8.1 MiB
Languages
C 64.5%
Assembly 33.6%
Perl 0.8%
Python 0.8%
Makefile 0.2%