djbsort-speed(1)

πŸ“¦
this version Β· 2 versions
, in epel 10, epel 9, fedora 43, fedora 44

πŸ’– djbsort-speed β€” the speed test page, bestie edition πŸ’–

okay so real quick before anything: this is the manual page for djbsort-speed, from section 1 of the manual (the "commands you just type" section, yk). now let's get into it 🫢


✨ NAME ✨

djbsort-speed β€” runs some speed tests on the djbsort library, no cap πŸ’…


πŸŽ€ SYNOPSIS πŸŽ€

djbsort-speed

that's it. that's the whole vibe. no flags, no args, just send it ✨


🌸 DESCRIPTION 🌸

djbsort-speed runs a bunch of tests measuring how fast the djbsort library actually goes. it's giving benchmark energy, it's giving performance check, it's giving "let's see what this CPU can do" πŸ”₯

heads up bestie: the output format of djbsort-speed is subject to change. so don't go parsing it in your lil scripts like it's a stable API, because it is Not That πŸ™ˆ

here are some example output lines, in the current format, from a djbsort-speed run on an AMD Zen 3 CPU core:

sort_int32 selected implementation avx2 compiler gcc -Wall -fPIC -fwrapv -O2 -mmmx -msse -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -mavx -mbmi -mbmi2 -mpopcnt -mavx2 -mtune=haswell; gcc (Debian 12.2.0-14+deb12u1) 12.2.0; Copyright (C) 2022 Free Software Foundation, Inc.; This is free software; see the source for copying conditions.  There is NO; warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
sort_int32 selected 1597 8769 +1335+200-12+36-20-70+20-61-25+57-55+26+4-22+10+0+26-1-70-41+53-59-25+57-55+26+4-22+10+0+26-1

so what are these lines even saying?? let me translate 🧚

  • they're telling you that the implementation automatically selected for djbsort_int32 was the avx2 implementation, and it got compiled with the clang options shown above πŸ‘‘
  • and that it took an interquartile mean (IQM) of 8769 CPU cycles to sort an int32[1597] array ⭐
  • and then those numbers β€” 8769 +1335+200+... β€” are the timings for a series of consecutive operations: the IQM was 8769 cycles, the first run took 1335 cycles more than the IQM, the second run took 200 cycles more than the IQM, etc. so like, a lil deviation report card for each run πŸ’€

πŸ¦‹ SEE ALSO πŸ¦‹


and that's the whole page, bestie!! this was djbsort-speed(1) β€” go forth and benchmark, mwah πŸ’–βœ¨πŸŽ€πŸ¦‹πŸŒΈπŸ’…πŸ«Ά

πŸ“¦ where this page comes from