From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sunil Mushran Date: Wed, 12 May 2010 14:09:28 -0700 Subject: [Ocfs2-devel] [PATCH 2/2] ocfs2: Block signals for mkdir/link/symlink/O_CREAT. In-Reply-To: <20100510190031.GE25670@mail.oracle.com> References: <20100510185919.GC25670@mail.oracle.com> <20100510190031.GE25670@mail.oracle.com> Message-ID: <4BEB1908.1030806@oracle.com> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: ocfs2-devel@oss.oracle.com Signed-off-by: Sunil Mushran On 05/10/2010 12:00 PM, Joel Becker wrote: > Once file or link creation gets going, it can't be interrupted by a > signal. They're not idempotent. > > This blocks signals in ocfs2_mknod(), ocfs2_link(), and ocfs2_symlink() > once we start actually changing things. ocfs2_mknod() covers mknod(), > creat(), mkdir(), and open(O_CREAT). > > Signed-off-by: Joel Becker > --- > fs/ocfs2/namei.c | 21 +++++++++++++++++++++ > 1 files changed, 21 insertions(+), 0 deletions(-) > > diff --git a/fs/ocfs2/namei.c b/fs/ocfs2/namei.c > index 21d4a33..607084b 100644 > --- a/fs/ocfs2/namei.c > +++ b/fs/ocfs2/namei.c > @@ -239,6 +239,8 @@ static int ocfs2_mknod(struct inode *dir, > }; > int did_quota_inode = 0; > struct ocfs2_dir_lookup_result lookup = { NULL, }; > + sigset_t oldset; > + int did_block_signals = 0; > > mlog_entry("(0x%p, 0x%p, %d, %lu, '%.*s')\n", dir, dentry, mode, > (unsigned long)dev, dentry->d_name.len, > @@ -350,6 +352,10 @@ static int ocfs2_mknod(struct inode *dir, > goto leave; > } > > + /* Starting to change things, restart is no longer possible. */ > + ocfs2_block_signals(&oldset); > + did_block_signals = 1; > + > status = dquot_alloc_inode(inode); > if (status) > goto leave; > @@ -430,6 +436,8 @@ leave: > ocfs2_commit_trans(osb, handle); > > ocfs2_inode_unlock(dir, 1); > + if (did_block_signals) > + ocfs2_unblock_signals(&oldset); > > if (status == -ENOSPC) > mlog(0, "Disk is full\n"); > @@ -618,6 +626,7 @@ static int ocfs2_link(struct dentry *old_dentry, > struct ocfs2_dinode *fe = NULL; > struct ocfs2_super *osb = OCFS2_SB(dir->i_sb); > struct ocfs2_dir_lookup_result lookup = { NULL, }; > + sigset_t oldset; > > mlog_entry("(inode=%lu, old='%.*s' new='%.*s')\n", inode->i_ino, > old_dentry->d_name.len, old_dentry->d_name.name, > @@ -674,6 +683,9 @@ static int ocfs2_link(struct dentry *old_dentry, > goto out_unlock_inode; > } > > + /* Starting to change things, restart is no longer possible. */ > + ocfs2_block_signals(&oldset); > + > err = ocfs2_journal_access_di(handle, INODE_CACHE(inode), fe_bh, > OCFS2_JOURNAL_ACCESS_WRITE); > if (err< 0) { > @@ -710,6 +722,7 @@ static int ocfs2_link(struct dentry *old_dentry, > > out_commit: > ocfs2_commit_trans(osb, handle); > + ocfs2_unblock_signals(&oldset); > out_unlock_inode: > ocfs2_inode_unlock(inode, 1); > > @@ -1568,6 +1581,8 @@ static int ocfs2_symlink(struct inode *dir, > }; > int did_quota = 0, did_quota_inode = 0; > struct ocfs2_dir_lookup_result lookup = { NULL, }; > + sigset_t oldset; > + int did_block_signals = 0; > > mlog_entry("(0x%p, 0x%p, symname='%s' actual='%.*s')\n", dir, > dentry, symname, dentry->d_name.len, dentry->d_name.name); > @@ -1663,6 +1678,10 @@ static int ocfs2_symlink(struct inode *dir, > goto bail; > } > > + /* Starting to change things, restart is no longer possible. */ > + ocfs2_block_signals(&oldset); > + did_block_signals = 1; > + > status = dquot_alloc_inode(inode); > if (status) > goto bail; > @@ -1766,6 +1785,8 @@ bail: > ocfs2_commit_trans(osb, handle); > > ocfs2_inode_unlock(dir, 1); > + if (did_block_signals) > + ocfs2_unblock_signals(&oldset); > > brelse(new_fe_bh); > brelse(parent_fe_bh); >