From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Thu, 8 Feb 2001 03:22:43 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Thu, 8 Feb 2001 03:22:33 -0500 Received: from adsl-63-205-85-133.dsl.snfc21.pacbell.net ([63.205.85.133]:26128 "EHLO schmee.sfgoth.com") by vger.kernel.org with ESMTP id ; Thu, 8 Feb 2001 03:22:18 -0500 Date: Thu, 8 Feb 2001 00:22:12 -0800 From: Mitchell Blank Jr To: "David S. Miller" Cc: dean gaudet , linux-kernel@vger.kernel.org Subject: Re: dentry cache order 7 is broken Message-ID: <20010208002212.Q12227@sfgoth.com> In-Reply-To: <14978.21605.98365.252519@pizda.ninka.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-Mailer: Mutt 1.0i In-Reply-To: <14978.21605.98365.252519@pizda.ninka.net>; from davem@redhat.com on Thu, Feb 08, 2001 at 12:10:13AM -0800 Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org David S. Miller wrote: > - hash = hash ^ (hash >> D_HASHBITS) ^ (hash >> D_HASHBITS*2); > + hash = hash ^ (hash >> D_HASHBITS) ^ > + (hash >> (D_HASHBITS+(D_HASHBITS/2))); Two comments: 1. The inode-cache has the exact same problem, but it'll require a lot of RAM to run into it. The buffer and page caches don't have the same problem. 2. Given that D_HASHBITS is not a constant I wonder if there isn't a more efficient hash to be found. But I guess I'll leave that to the hashing experts. -Mitch - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org Please read the FAQ at http://www.tux.org/lkml/