From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Thu, 19 Jul 2007 21:12:15 -0700 (PDT) Received: from sandeen.net (sandeen.net [209.173.210.139]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id l6K4CAbm027829 for ; Thu, 19 Jul 2007 21:12:11 -0700 Received: from liberator.sandeen.net (liberator.sandeen.net [10.0.0.4]) (using TLSv1 with cipher DHE-RSA-AES256-SHA (256/256 bits)) (No client certificate requested) by sandeen.net (Postfix) with ESMTP id B56871801056F for ; Thu, 19 Jul 2007 23:12:13 -0500 (CDT) Message-ID: <46A0361D.3090103@sandeen.net> Date: Thu, 19 Jul 2007 23:12:13 -0500 From: Eric Sandeen MIME-Version: 1.0 Subject: [PATCH 3/7] - pick single default inode cluster Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: xfs-oss Remove scaling of inode "clusters" based on machine memory; small cluster cut-point was an unrealistic 32MB and was probably never tested. Removes another user of xfs_physmem. Signed-off-by: Eric Sandeen Index: linux/fs/xfs/xfs_ialloc.h =================================================================== --- linux.orig/fs/xfs/xfs_ialloc.h +++ linux/fs/xfs/xfs_ialloc.h @@ -30,14 +30,9 @@ struct xfs_trans; #define XFS_IALLOC_BLOCKS(mp) (mp)->m_ialloc_blks /* - * For small block file systems, move inodes in clusters of this size. - * When we don't have a lot of memory, however, we go a bit smaller - * to reduce the number of AGI and ialloc btree blocks we need to keep - * around for xfs_dilocate(). We choose which one to use in - * xfs_mount_int(). + * Move inodes in clusters of this size. */ #define XFS_INODE_BIG_CLUSTER_SIZE 8192 -#define XFS_INODE_SMALL_CLUSTER_SIZE 4096 #define XFS_INODE_CLUSTER_SIZE(mp) (mp)->m_inode_cluster_size /* Index: linux/fs/xfs/xfs_mount.c =================================================================== --- linux.orig/fs/xfs/xfs_mount.c +++ linux/fs/xfs/xfs_mount.c @@ -885,15 +885,12 @@ xfs_mountfs( mp->m_writeio_blocks = 1 << (mp->m_writeio_log - sbp->sb_blocklog); /* - * Set the inode cluster size based on the physical memory - * size. This may still be overridden by the file system + * Set the inode cluster size. + * This may still be overridden by the file system * block size if it is larger than the chosen cluster size. */ - if (xfs_physmem <= btoc(32 * 1024 * 1024)) { /* <= 32 MB */ - mp->m_inode_cluster_size = XFS_INODE_SMALL_CLUSTER_SIZE; - } else { - mp->m_inode_cluster_size = XFS_INODE_BIG_CLUSTER_SIZE; - } + mp->m_inode_cluster_size = XFS_INODE_BIG_CLUSTER_SIZE; + /* * Set whether we're using inode alignment. */