description | none |
repository URL | https://github.com/rofl0r/memcpy-test.git |
owner | retnyg@gmx.net |
last change | Sat, 7 May 2022 21:52:54 +0000 (7 21:52 +0000) |
last refresh | Thu, 21 Nov 2024 07:47:19 +0000 (21 08:47 +0100) |
mirror URL | git://repo.or.cz/rofl0r-memcpy-test.git |
| https://repo.or.cz/rofl0r-memcpy-test.git |
| ssh://git@repo.or.cz/rofl0r-memcpy-test.git |
bundle info | rofl0r-memcpy-test.git downloadable bundles |
content tags
|
|
README
pick one of the memcpy implementations or add a file containing a memcpy function called
`mymemcpy`.
it needs to be called this way in order to keep optimizations/system libc from
interfering.
then run
`CC=gcc OPTS="-O3 -funroll-all-loops" ./compare.py impl1.c [impl2.s ... implX.c]`
(both .s and .c suffixes are supported)
if you use a modern CPU using speedstep, you can get much better results by
putting the CPU into "userspace" governor and then lock the CPU freq to 1 GHz.