Date: Sun, 27 Apr 1997 11:55:34 -0700 (MST) From: Terry Lambert <terry@lambert.org> To: michaelh@cet.co.jp (Michael Hancock) Cc: terry@lambert.org, bde@zeta.org.au, msmith@atrad.adelaide.edu.au, hackers@hub.freebsd.org Subject: Re: namei & hash functions Message-ID: <199704271855.LAA08913@phaeton.artisoft.com> In-Reply-To: <Pine.SV4.3.95.970427102525.25227A-100000@parkplace.cet.co.jp> from "Michael Hancock" at Apr 27, 97 10:42:13 am
next in thread | previous in thread | raw e-mail | index | archive | help
> He's working on another approach that simplifies keeping the name cache > and active vnode in sync by hanging the name cache of the directory vnode. > A lot of us are saying that operations like article list request against > innd would suffer with this approach. Ugh. I do not like this. It breaks locality of reference for cache entry flushing. I don't think it's that big a win for lookup anyway; a hash is a hash, and the fill should be proportionate. There will be a problem in that you can size a "whole FS hash" by knowing how many files are in the FS, but sizing a "per directory hash" is much more problematic. I assume that it's multiply linked? Otherwise you'd have problems on volume flush on unmount, etc.. Regards, Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199704271855.LAA08913>