On Mon, 6 Apr 2015, Adrian Chadd wrote:
Hi Bruce,
There's been a bunch of research into that field lately.
Eg: http://daim.idi.ntnu.no/masteroppgaver/009/9231/masteroppgave.pdf
The best part (imho)- the length of your USER environment string
influencing code runtime costs. :)
I fix the latter in an FP benchmark by aligning the stack to a page
boundary, and further adjusting it by a magic amount of a few hundred.
This only works for microbenchmarks, since in general the stack pointer
is too variable for a single adjustment to keep it from hitting things.
Bruce
_______________________________________________
svn-src-head@freebsd.org mailing list
http://lists.freebsd.org/mailman/listinfo/svn-src-head
To unsubscribe, send any mail to "svn-src-head-unsubscr...@freebsd.org"
