Date: Thu, 24 Apr 1997 15:42:23 +0200 From: Poul-Henning Kamp <phk@dk.tfs.com> To: Michael Hancock <michaelh@cet.co.jp> Cc: David Greenman <dg@root.com>, Poul-Henning Kamp <phk@dk.tfs.com>, fs@freebsd.org Subject: Re: the namei cache... Message-ID: <357.861889343@critter> In-Reply-To: Your message of "Thu, 24 Apr 1997 21:30:02 %2B0900." <Pine.SV4.3.95.970424212315.13828A-100000@parkplace.cet.co.jp>
next in thread | previous in thread | raw e-mail | index | archive | help
In message <Pine.SV4.3.95.970424212315.13828A-100000@parkplace.cet.co.jp>, Mich ael Hancock writes: > >> It's interesting that the single largest CPU consumer on wcarchive appear >s >> to be the namei cache lookups. I think the hash algorithm needs to be >> re-visited at the very least, perhaps changing the divide by a prime into >> some sort of xor of the filename characters (xored in pairs as int16's). > >I don't think you need to even look at all the characters. You can take >the first and last and couple in the middle. To get the middle ones, >instead of dividing the length by 2 do a bit shift on the length. Then >xor the 2 pairs. Any hash guru's on the list? I'd rather get rid of it entirely... -- Poul-Henning Kamp | phk@FreeBSD.ORG FreeBSD Core-team. http://www.freebsd.org/~phk | phk@login.dknet.dk Private mailbox. whois: [PHK] | phk@tfs.com TRW Financial Systems, Inc. Power and ignorance is a disgusting cocktail.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?357.861889343>