From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S264352AbTKZWRX (ORCPT ); Wed, 26 Nov 2003 17:17:23 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S264351AbTKZWRX (ORCPT ); Wed, 26 Nov 2003 17:17:23 -0500 Received: from dp.samba.org ([66.70.73.150]:30130 "EHLO lists.samba.org") by vger.kernel.org with ESMTP id S264352AbTKZWRW (ORCPT ); Wed, 26 Nov 2003 17:17:22 -0500 Date: Wed, 26 Nov 2003 18:25:53 +1100 From: Anton Blanchard To: Jack Steiner Cc: Jes Sorensen , Alexander Viro , Andrew Morton , "William Lee Irwin, III" , linux-kernel@vger.kernel.org, jbarnes@sgi.com Subject: Re: hash table sizes Message-ID: <20031126072553.GF26811@krispykreme> References: <16323.23221.835676.999857@gargle.gargle.HOWL> <20031125204814.GA19397@sgi.com> <20031125211611.GE26811@krispykreme> <20031125231108.GA5675@sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20031125231108.GA5675@sgi.com> User-Agent: Mutt/1.5.4i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org > That was a concern to me too. However, on IA64, all page structs are > in the vmalloc region (it isnt allocated by vmalloc but is in the same > region as vmalloc'ed pages. They are mapped with 16k pages instead of > the 64MB pages used for memory allocated by kmalloc). > > Before switching to 16K pages for the page structs, we made numerous > performance measurements. As far as we could tell, there was no > performce degradation caused by the smaller pages. It seems to me that > if page structs are ok being mapped by 16k pages, the hash tables > would be ok too. OK, on ppc64 with a 4kB base pagesize Id be more worried about using the vmalloc region. Anton