From patchwork Mon May 14 18:47:19 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Josef Bacik X-Patchwork-Id: 159116 Return-Path: X-Original-To: patchwork-incoming@ozlabs.org Delivered-To: patchwork-incoming@ozlabs.org Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by ozlabs.org (Postfix) with ESMTP id A26E5B702A for ; Tue, 15 May 2012 04:47:30 +1000 (EST) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757584Ab2ENSr1 (ORCPT ); Mon, 14 May 2012 14:47:27 -0400 Received: from mx1.redhat.com ([209.132.183.28]:27170 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757548Ab2ENSr0 (ORCPT ); Mon, 14 May 2012 14:47:26 -0400 Received: from int-mx01.intmail.prod.int.phx2.redhat.com (int-mx01.intmail.prod.int.phx2.redhat.com [10.5.11.11]) by mx1.redhat.com (8.14.4/8.14.4) with ESMTP id q4EIlLat001101 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=OK); Mon, 14 May 2012 14:47:21 -0400 Received: from localhost.localdomain.com (vpn-11-79.rdu.redhat.com [10.11.11.79]) by int-mx01.intmail.prod.int.phx2.redhat.com (8.13.8/8.13.8) with ESMTP id q4EIlJfn028711; Mon, 14 May 2012 14:47:20 -0400 From: Josef Bacik To: linux-ext4@vger.kernel.org, linux-nfs@vger.kernel.org, bfields@fieldses.org, adilger@dilger.ca, tytso@mit.edu Subject: [PATCH] ext4: fix how i_version is modified and turn it on by default Date: Mon, 14 May 2012 14:47:19 -0400 Message-Id: <1337021239-2790-1-git-send-email-josef@redhat.com> X-Scanned-By: MIMEDefang 2.67 on 10.5.11.11 Sender: linux-ext4-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-ext4@vger.kernel.org This makes MS_I_VERSION be turned on by default. Ext4 had been unconditionally doing i_version++ in a few cases anway so the mount option was kind of silly. This patch also removes the update in mark_inode_dirty and makes all of the cases where we update ctime also do inode_inc_iversion. file_update_time takes care of the write case and all the places where we update iversion are protected by the i_mutex so there should be no extra i_lock overhead in the normal non-exported fs case. Thanks, Signed-off-by: Josef Bacik --- fs/ext4/acl.c | 1 + fs/ext4/extents.c | 3 +++ fs/ext4/indirect.c | 1 + fs/ext4/inode.c | 3 --- fs/ext4/ioctl.c | 2 ++ fs/ext4/namei.c | 14 ++++++++++---- fs/ext4/super.c | 5 +++-- fs/ext4/xattr.c | 1 + 8 files changed, 21 insertions(+), 9 deletions(-) diff --git a/fs/ext4/acl.c b/fs/ext4/acl.c index a5c29bb..fb3ff80 100644 --- a/fs/ext4/acl.c +++ b/fs/ext4/acl.c @@ -202,6 +202,7 @@ ext4_set_acl(handle_t *handle, struct inode *inode, int type, if (error < 0) return error; else { + inode_inc_iversion(inode); inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); if (error == 0) diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c index 74f23c2..a9ac029 100644 --- a/fs/ext4/extents.c +++ b/fs/ext4/extents.c @@ -4249,6 +4249,7 @@ out_stop: if (inode->i_nlink) ext4_orphan_del(handle, inode); + inode_inc_iversion(inode); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); @@ -4261,6 +4262,7 @@ static void ext4_falloc_update_inode(struct inode *inode, if (update_ctime) { now = current_fs_time(inode->i_sb); + inode_inc_iversion(inode); if (!timespec_equal(&inode->i_ctime, &now)) inode->i_ctime = now; } @@ -4905,6 +4907,7 @@ int ext4_ext_punch_hole(struct file *file, loff_t offset, loff_t length) out: ext4_orphan_del(handle, inode); + inode_inc_iversion(inode); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); diff --git a/fs/ext4/indirect.c b/fs/ext4/indirect.c index 830e1b2..45ed6d4 100644 --- a/fs/ext4/indirect.c +++ b/fs/ext4/indirect.c @@ -1476,6 +1476,7 @@ do_indirects: out_unlock: up_write(&ei->i_data_sem); + inode_inc_iversion(inode); inode->i_mtime = inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index feaa82f..d11b4e5 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -4314,9 +4314,6 @@ int ext4_mark_iloc_dirty(handle_t *handle, { int err = 0; - if (test_opt(inode->i_sb, I_VERSION)) - inode_inc_iversion(inode); - /* the do_update_inode consumes one bh->b_count */ get_bh(iloc->bh); diff --git a/fs/ext4/ioctl.c b/fs/ext4/ioctl.c index 6eee255..c79b874 100644 --- a/fs/ext4/ioctl.c +++ b/fs/ext4/ioctl.c @@ -120,6 +120,7 @@ long ext4_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) ei->i_flags = flags; ext4_set_inode_flags(inode); + inode_inc_iversion(inode); inode->i_ctime = ext4_current_time(inode); err = ext4_mark_iloc_dirty(handle, inode, &iloc); @@ -168,6 +169,7 @@ flags_out: } err = ext4_reserve_inode_write(handle, inode, &iloc); if (err == 0) { + inode_inc_iversion(inode); inode->i_ctime = ext4_current_time(inode); inode->i_generation = generation; err = ext4_mark_iloc_dirty(handle, inode, &iloc); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index 2043f48..9bb533a 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -1316,9 +1316,9 @@ static int add_dirent_to_buf(handle_t *handle, struct dentry *dentry, * happen is that the times are slightly out of date * and/or different from the directory change time. */ + inode_inc_iversion(dir); dir->i_mtime = dir->i_ctime = ext4_current_time(dir); ext4_update_dx_flag(dir); - dir->i_version++; ext4_mark_inode_dirty(handle, dir); BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); err = ext4_handle_dirty_metadata(handle, dir, bh); @@ -1668,7 +1668,6 @@ static int ext4_delete_entry(handle_t *handle, blocksize); else de->inode = 0; - dir->i_version++; BUFFER_TRACE(bh, "call ext4_handle_dirty_metadata"); err = ext4_handle_dirty_metadata(handle, dir, bh); if (unlikely(err)) { @@ -2158,13 +2157,14 @@ static int ext4_rmdir(struct inode *dir, struct dentry *dentry) ext4_warning(inode->i_sb, "empty directory has too many links (%d)", inode->i_nlink); - inode->i_version++; clear_nlink(inode); /* There's no need to set i_disksize: the fact that i_nlink is * zero will ensure that the right thing happens during any * recovery. */ inode->i_size = 0; ext4_orphan_add(handle, inode); + inode_inc_iversion(inode); + inode_inc_iversion(dir); inode->i_ctime = dir->i_ctime = dir->i_mtime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); ext4_dec_count(handle, dir); @@ -2218,12 +2218,14 @@ static int ext4_unlink(struct inode *dir, struct dentry *dentry) retval = ext4_delete_entry(handle, dir, de, bh); if (retval) goto end_unlink; + inode_inc_iversion(dir); dir->i_ctime = dir->i_mtime = ext4_current_time(dir); ext4_update_dx_flag(dir); ext4_mark_inode_dirty(handle, dir); drop_nlink(inode); if (!inode->i_nlink) ext4_orphan_add(handle, inode); + inode_inc_iversion(inode); inode->i_ctime = ext4_current_time(inode); ext4_mark_inode_dirty(handle, inode); retval = 0; @@ -2363,6 +2365,7 @@ retry: if (IS_DIRSYNC(dir)) ext4_handle_sync(handle); + inode_inc_iversion(inode); inode->i_ctime = ext4_current_time(inode); ext4_inc_count(handle, inode); ihold(inode); @@ -2470,7 +2473,7 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, if (EXT4_HAS_INCOMPAT_FEATURE(new_dir->i_sb, EXT4_FEATURE_INCOMPAT_FILETYPE)) new_de->file_type = old_de->file_type; - new_dir->i_version++; + inode_inc_iversion(new_dir); new_dir->i_ctime = new_dir->i_mtime = ext4_current_time(new_dir); ext4_mark_inode_dirty(handle, new_dir); @@ -2488,6 +2491,7 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, * Like most other Unix systems, set the ctime for inodes on a * rename. */ + inode_inc_iversion(old_inode); old_inode->i_ctime = ext4_current_time(old_inode); ext4_mark_inode_dirty(handle, old_inode); @@ -2521,8 +2525,10 @@ static int ext4_rename(struct inode *old_dir, struct dentry *old_dentry, if (new_inode) { ext4_dec_count(handle, new_inode); + inode_inc_iversion(new_inode); new_inode->i_ctime = ext4_current_time(new_inode); } + inode_inc_iversion(old_dir); old_dir->i_ctime = old_dir->i_mtime = ext4_current_time(old_dir); ext4_update_dx_flag(old_dir); if (dir_bh) { diff --git a/fs/ext4/super.c b/fs/ext4/super.c index 502c61f..5866d38 100644 --- a/fs/ext4/super.c +++ b/fs/ext4/super.c @@ -1813,8 +1813,7 @@ set_qf_format: "Ignoring deprecated bh option"); break; case Opt_i_version: - set_opt(sb, I_VERSION); - sb->s_flags |= MS_I_VERSION; + /* On by default now */ break; case Opt_nodelalloc: clear_opt(sb, DELALLOC); @@ -3132,6 +3131,7 @@ static int ext4_fill_super(struct super_block *sb, void *data, int silent) goto out_free_orig; } sb->s_fs_info = sbi; + sb->s_flags |= MS_I_VERSION; sbi->s_mount_opt = 0; sbi->s_resuid = EXT4_DEF_RESUID; sbi->s_resgid = EXT4_DEF_RESGID; @@ -4831,6 +4831,7 @@ static int ext4_quota_off(struct super_block *sb, int type) handle = ext4_journal_start(inode, 1); if (IS_ERR(handle)) goto out; + inode_inc_iversion(inode); inode->i_mtime = inode->i_ctime = CURRENT_TIME; ext4_mark_inode_dirty(handle, inode); ext4_journal_stop(handle); diff --git a/fs/ext4/xattr.c b/fs/ext4/xattr.c index 93a00d8..48a3412 100644 --- a/fs/ext4/xattr.c +++ b/fs/ext4/xattr.c @@ -1048,6 +1048,7 @@ ext4_xattr_set_handle(handle_t *handle, struct inode *inode, int name_index, } if (!error) { ext4_xattr_update_super_block(handle, inode->i_sb); + inode_inc_iversion(inode); inode->i_ctime = ext4_current_time(inode); if (!value) ext4_clear_inode_state(inode, EXT4_STATE_NO_EXPAND);