"kernel: Improve __copy_to_user and __copy_from_user performance
Provide a __copy_from_user that uses memcpy. On BCM2708, use
optimised memcpy/memmove/memcmp/memset implementations."
This update gives a significant performance boost to the memory system. On a Raspberry Pi running at 1 GHz with turbo overclocking I can get for the very first time cached reads above 300 MB/s from hdparm. Previously I have seen a result slightly above 230 MB/s at best. As this has been optimized already earlier I did not expect so large gains still to be achieved! Thank you very much!
Code: Select all
hdparm -Tt /dev/mmcblk0p2
/dev/mmcblk0p2:
Timing cached reads: 674 MB in 2.00 seconds = 336.59 MB/sec
Timing buffered disk reads: 54 MB in 3.09 seconds = 17.47 MB/sec