From: Eric Sandeen Subject: Re: [PATCH 2/2] ext4: Use slab allocator for sub-page sized allocations Date: Sat, 05 Dec 2009 00:43:13 -0600 Message-ID: <4B1A0101.4090605@redhat.com> References: <1259716451-26901-1-git-send-email-tytso@mit.edu> <1259716451-26901-2-git-send-email-tytso@mit.edu> <4B169B86.4040707@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Cc: Ext4 Developers List To: "Theodore Ts'o" Return-path: Received: from mx1.redhat.com ([209.132.183.28]:26561 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751528AbZLEGnR (ORCPT ); Sat, 5 Dec 2009 01:43:17 -0500 In-Reply-To: <4B169B86.4040707@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: Eric Sandeen wrote: > Theodore Ts'o wrote: >> Now that the SLUB seems to be fixed so that it respects the requested >> alignment, use kmem_cache_alloc() to allocator if the block size of >> the buffer heads to be allocated is less than the page size. >> Previously, we were using 16k page on a Power system for each buffer, >> even when the file system was using 1k or 4k block size. > > So, this undoes commit c089d490dfbf53bc0893dc9ef57cf3ee6448314d > more or less, right: > > JBD: Replace slab allocations with page allocations > > Author: Mingming Cao > > JBD allocate memory for committed_data and frozen_data from slab. However > JBD should not pass slab pages down to the block layer. > Use page allocator pages instead. This will also prepare JBD for the large blocksize patchset. > > Was alignment the only reason that commit went in? Actually, Christoph reminded me that iscsi & co will not like this. See commit 1fa40b01ae4d1b00e366d4949edcc230f5cd6d99 for xfs moving in the opposite direction back in 2007... -Eric