From patchwork Wed Dec 21 17:03:42 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Jeff Layton X-Patchwork-Id: 9483373 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 669F0601B3 for ; Wed, 21 Dec 2016 17:07:32 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 533E82841B for ; Wed, 21 Dec 2016 17:07:32 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 48382281A7; Wed, 21 Dec 2016 17:07:32 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-4.4 required=2.0 tests=BAYES_00, RCVD_IN_DNSWL_HI, SUSPICIOUS_RECIPS autolearn=unavailable version=3.3.1 Received: from vger.kernel.org (vger.kernel.org [209.132.180.67]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id A7DC528424 for ; Wed, 21 Dec 2016 17:07:31 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1757764AbcLURGj (ORCPT ); Wed, 21 Dec 2016 12:06:39 -0500 Received: from mx1.redhat.com ([209.132.183.28]:42168 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S938318AbcLUREM (ORCPT ); Wed, 21 Dec 2016 12:04:12 -0500 Received: from int-mx13.intmail.prod.int.phx2.redhat.com (int-mx13.intmail.prod.int.phx2.redhat.com [10.5.11.26]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id 691CD4ACBF; Wed, 21 Dec 2016 17:04:12 +0000 (UTC) Received: from tleilax.poochiereds.net (ovpn-118-75.rdu2.redhat.com [10.10.118.75]) by int-mx13.intmail.prod.int.phx2.redhat.com (8.14.4/8.14.4) with ESMTP id uBLH3l60022318; Wed, 21 Dec 2016 12:04:11 -0500 From: Jeff Layton To: linux-fsdevel@vger.kernel.org Cc: linux-kernel@vger.kernel.org, linux-nfs@vger.kernel.org, linux-ext4@vger.kernel.org, linux-btrfs@vger.kernel.org, linux-xfs@vger.kernel.org Subject: [RFC PATCH v1 25/30] fs: add a "force" parameter to inode_inc_iversion Date: Wed, 21 Dec 2016 12:03:42 -0500 Message-Id: <1482339827-7882-26-git-send-email-jlayton@redhat.com> In-Reply-To: <1482339827-7882-1-git-send-email-jlayton@redhat.com> References: <1482339827-7882-1-git-send-email-jlayton@redhat.com> X-Scanned-By: MIMEDefang 2.68 on 10.5.11.26 X-Greylist: Sender IP whitelisted, not delayed by milter-greylist-4.5.16 (mx1.redhat.com [10.5.110.38]); Wed, 21 Dec 2016 17:04:12 +0000 (UTC) Sender: linux-btrfs-owner@vger.kernel.org Precedence: bulk List-ID: X-Mailing-List: linux-btrfs@vger.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP In some coming changes, we will want to avoid changing the i_version counter in some cases as that may allow us to avoid having to log a metadata change to the inode. In other cases we will want to forcibly change it even when it's not strictly required (i.e. if we're going to log a time change to disk anyway). Add a parameter to tell the incrementing routine to forcibly bump the counter. For now, everyone sets this to "true" and the value is ignored, but later patches will change this. Signed-off-by: Jeff Layton --- fs/btrfs/file.c | 4 ++-- fs/btrfs/inode.c | 34 +++++++++++++++++----------------- fs/btrfs/ioctl.c | 4 ++-- fs/btrfs/xattr.c | 2 +- fs/ext2/super.c | 2 +- fs/ext4/inode.c | 4 ++-- fs/ext4/namei.c | 2 +- fs/inode.c | 2 +- fs/ocfs2/namei.c | 2 +- include/linux/fs.h | 2 +- 10 files changed, 29 insertions(+), 29 deletions(-) diff --git a/fs/btrfs/file.c b/fs/btrfs/file.c index b5c5da215d05..f301afaf9af5 100644 --- a/fs/btrfs/file.c +++ b/fs/btrfs/file.c @@ -1804,7 +1804,7 @@ static void update_time_for_write(struct inode *inode) inode->i_ctime = now; if (IS_I_VERSION(inode)) - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); } static ssize_t btrfs_file_write_iter(struct kiocb *iocb, @@ -2636,7 +2636,7 @@ static int btrfs_punch_hole(struct inode *inode, loff_t offset, loff_t len) if (!trans) goto out_free; - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); inode->i_mtime = inode->i_ctime = current_time(inode); trans->block_rsv = &fs_info->trans_block_rsv; diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index a56a45a3e992..a03e5a1d5e05 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4086,8 +4086,8 @@ static int __btrfs_unlink_inode(struct btrfs_trans_handle *trans, goto out; btrfs_i_size_write(dir, dir->i_size - name_len * 2); - inode_inc_iversion(inode); - inode_inc_iversion(dir); + inode_inc_iversion(inode, true); + inode_inc_iversion(dir, true); inode->i_ctime = dir->i_mtime = dir->i_ctime = current_time(inode); ret = btrfs_update_inode(trans, root, dir); @@ -4232,7 +4232,7 @@ int btrfs_unlink_subvol(struct btrfs_trans_handle *trans, } btrfs_i_size_write(dir, dir->i_size - name_len * 2); - inode_inc_iversion(dir); + inode_inc_iversion(dir, true); dir->i_mtime = dir->i_ctime = current_time(dir); ret = btrfs_update_inode_fallback(trans, root, dir); if (ret) @@ -5000,7 +5000,7 @@ static int btrfs_setsize(struct inode *inode, struct iattr *attr) * explicitly if it wants a timestamp update. */ if (newsize != oldsize) { - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); if (!(mask & (ATTR_CTIME | ATTR_MTIME))) inode->i_ctime = inode->i_mtime = current_time(inode); @@ -5122,7 +5122,7 @@ static int btrfs_setattr(struct dentry *dentry, struct iattr *attr) if (attr->ia_valid) { setattr_copy(inode, attr); - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); err = btrfs_dirty_inode(inode); if (!err && attr->ia_valid & ATTR_MODE) @@ -6030,7 +6030,7 @@ static int btrfs_update_time(struct inode *inode, struct timespec *now, return -EROFS; if (flags & S_VERSION) - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); if (flags & S_CTIME) inode->i_ctime = *now; if (flags & S_MTIME) @@ -6354,7 +6354,7 @@ int btrfs_add_link(struct btrfs_trans_handle *trans, btrfs_i_size_write(parent_inode, parent_inode->i_size + name_len * 2); - inode_inc_iversion(parent_inode); + inode_inc_iversion(parent_inode, true); parent_inode->i_mtime = parent_inode->i_ctime = current_time(parent_inode); ret = btrfs_update_inode(trans, root, parent_inode); @@ -6576,7 +6576,7 @@ static int btrfs_link(struct dentry *old_dentry, struct inode *dir, /* There are several dir indexes for this inode, clear the cache. */ BTRFS_I(inode)->dir_index = 0ULL; inc_nlink(inode); - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); inode->i_ctime = current_time(inode); ihold(inode); set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags); @@ -9576,10 +9576,10 @@ static int btrfs_rename_exchange(struct inode *old_dir, } /* Update inode version and ctime/mtime. */ - inode_inc_iversion(old_dir); - inode_inc_iversion(new_dir); - inode_inc_iversion(old_inode); - inode_inc_iversion(new_inode); + inode_inc_iversion(old_dir, true); + inode_inc_iversion(new_dir, true); + inode_inc_iversion(old_inode, true); + inode_inc_iversion(new_inode, true); old_dir->i_ctime = old_dir->i_mtime = ctime; new_dir->i_ctime = new_dir->i_mtime = ctime; old_inode->i_ctime = ctime; @@ -9858,9 +9858,9 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, goto out_fail; } - inode_inc_iversion(old_dir); - inode_inc_iversion(new_dir); - inode_inc_iversion(old_inode); + inode_inc_iversion(old_dir, true); + inode_inc_iversion(new_dir, true); + inode_inc_iversion(old_inode, true); old_dir->i_ctime = old_dir->i_mtime = new_dir->i_ctime = new_dir->i_mtime = old_inode->i_ctime = current_time(old_dir); @@ -9887,7 +9887,7 @@ static int btrfs_rename(struct inode *old_dir, struct dentry *old_dentry, } if (new_inode) { - inode_inc_iversion(new_inode); + inode_inc_iversion(new_inode, true); new_inode->i_ctime = current_time(new_inode); if (unlikely(btrfs_ino(new_inode) == BTRFS_EMPTY_SUBVOL_DIR_OBJECTID)) { @@ -10408,7 +10408,7 @@ static int __btrfs_prealloc_file_range(struct inode *inode, int mode, cur_offset += ins.offset; *alloc_hint = ins.objectid + ins.offset; - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); inode->i_ctime = current_time(inode); BTRFS_I(inode)->flags |= BTRFS_INODE_PREALLOC; if (!(mode & FALLOC_FL_KEEP_SIZE) && diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index 33f967d30b2a..c2d845b73e58 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -348,7 +348,7 @@ static int btrfs_ioctl_setflags(struct file *file, void __user *arg) } btrfs_update_iflags(inode); - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); inode->i_ctime = current_time(inode); ret = btrfs_update_inode(trans, root, inode); @@ -3294,7 +3294,7 @@ static int clone_finish_inode_update(struct btrfs_trans_handle *trans, struct btrfs_root *root = BTRFS_I(inode)->root; int ret; - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); if (!no_time_update) inode->i_mtime = inode->i_ctime = current_time(inode); /* diff --git a/fs/btrfs/xattr.c b/fs/btrfs/xattr.c index 9621c7f2503e..b9a5475754b8 100644 --- a/fs/btrfs/xattr.c +++ b/fs/btrfs/xattr.c @@ -253,7 +253,7 @@ int __btrfs_setxattr(struct btrfs_trans_handle *trans, if (ret) goto out; - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); inode->i_ctime = current_time(inode); set_bit(BTRFS_INODE_COPY_EVERYTHING, &BTRFS_I(inode)->runtime_flags); ret = btrfs_update_inode(trans, root, inode); diff --git a/fs/ext2/super.c b/fs/ext2/super.c index 5c9ba3b3f77b..db8251e9fbd9 100644 --- a/fs/ext2/super.c +++ b/fs/ext2/super.c @@ -1542,7 +1542,7 @@ static ssize_t ext2_quota_write(struct super_block *sb, int type, return err; if (inode->i_size < off+len-towrite) i_size_write(inode, off+len-towrite); - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); inode->i_mtime = inode->i_ctime = current_time(inode); mark_inode_dirty(inode); return len - towrite; diff --git a/fs/ext4/inode.c b/fs/ext4/inode.c index 5603e1782c65..cd1c255944b3 100644 --- a/fs/ext4/inode.c +++ b/fs/ext4/inode.c @@ -5253,7 +5253,7 @@ int ext4_setattr(struct dentry *dentry, struct iattr *attr) return -EINVAL; if (IS_I_VERSION(inode) && attr->ia_size != inode->i_size) - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); if (ext4_should_order_data(inode) && (attr->ia_size < inode->i_size)) { @@ -5488,7 +5488,7 @@ int ext4_mark_iloc_dirty(handle_t *handle, int err = 0; if (IS_I_VERSION(inode)) - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); /* the do_update_inode consumes one bh->b_count */ get_bh(iloc->bh); diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c index a8f812162dd6..3c5110817a55 100644 --- a/fs/ext4/namei.c +++ b/fs/ext4/namei.c @@ -3379,7 +3379,7 @@ static int ext4_setent(handle_t *handle, struct ext4_renament *ent, ent->de->inode = cpu_to_le32(ino); if (ext4_has_feature_filetype(ent->dir->i_sb)) ent->de->file_type = file_type; - inode_inc_iversion(ent->dir); + inode_inc_iversion(ent->dir, true); ent->dir->i_ctime = ent->dir->i_mtime = current_time(ent->dir); ext4_mark_inode_dirty(handle, ent->dir); diff --git a/fs/inode.c b/fs/inode.c index 88110fd0b282..2484d91ae8ef 100644 --- a/fs/inode.c +++ b/fs/inode.c @@ -1625,7 +1625,7 @@ int generic_update_time(struct inode *inode, struct timespec *time, int flags) if (flags & S_ATIME) inode->i_atime = *time; if (flags & S_VERSION) - inode_inc_iversion(inode); + inode_inc_iversion(inode, true); if (flags & S_CTIME) inode->i_ctime = *time; if (flags & S_MTIME) diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c index c045826b716a..387cfa36c7b7 100644 --- a/fs/ocfs2/namei.c +++ b/fs/ocfs2/namei.c @@ -1520,7 +1520,7 @@ static int ocfs2_rename(struct inode *old_dir, mlog_errno(status); goto bail; } - inode_inc_iversion(new_dir); + inode_inc_iversion(new_dir, true); if (S_ISDIR(new_inode->i_mode)) ocfs2_set_links_count(newfe, 0); diff --git a/include/linux/fs.h b/include/linux/fs.h index 075c915fe2b1..75323e7b6954 100644 --- a/include/linux/fs.h +++ b/include/linux/fs.h @@ -2013,7 +2013,7 @@ inode_set_iversion_read(struct inode *inode, const u64 new) * The filesystem has to be mounted with MS_I_VERSION flag. */ static inline bool -inode_inc_iversion(struct inode *inode) +inode_inc_iversion(struct inode *inode, bool force) { spin_lock(&inode->i_lock); inode_inc_iversion_locked(inode);