The source isn't exactly top rate (a 2013 thread from the archlinuxarm.org forums), but this seems to tally with your linked answer:
For those who don't know the libcofi was a small library that had hand
optimized assembly version of 2 libc functions that could speed
applications using those functions a lot, when set to be preloaded.
Libcofi has now been superseded by arm-mem, it provided more hand
optimized functions and is also faster. This is included in raspbian
by default, so make sure to install this, it would be shame to have
slower performance on Arch compared to Debian...
Looks like some enterprising soul optimised the memcpy and memset functions for Raspberry Pi, which ultimately sped up the programs that used them. Bouncing around Google gave me a few hints towards possible source code:
Source: https://www.raspberrypi.org/forums/viewtopic.php?f=71&t=47832
I believe Raspbian uses the libcofi optimized memcpy and memset via
the ld.so.preload mechanism (/etc/ld.so.preload), which is from
https://github.com/simonjhall/copies-and-fills/
Source: https://github.com/simonjhall/copies-and-fills
Replacement memcpy and memset functionality for the Raspberry Pi with the intention of gaining greater performance.