From mboxrd@z Thu Jan 1 00:00:00 1970 From: Sasha Levin Subject: [PATCH AUTOSEL 5.5 20/58] ceph: do not execute direct write in parallel if O_APPEND is specified Date: Sat, 22 Feb 2020 21:20:41 -0500 Message-ID: <20200223022119.707-20-sashal@kernel.org> References: <20200223022119.707-1-sashal@kernel.org> Mime-Version: 1.0 Content-Transfer-Encoding: 8bit Return-path: Received: from mail.kernel.org ([198.145.29.99]:50252 "EHLO mail.kernel.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1727689AbgBWCVp (ORCPT ); Sat, 22 Feb 2020 21:21:45 -0500 In-Reply-To: <20200223022119.707-1-sashal@kernel.org> Sender: ceph-devel-owner@vger.kernel.org List-ID: To: linux-kernel@vger.kernel.org, stable@vger.kernel.org Cc: Xiubo Li , Jeff Layton , Ilya Dryomov , Sasha Levin , ceph-devel@vger.kernel.org From: Xiubo Li [ Upstream commit 8e4473bb50a1796c9c32b244e5dbc5ee24ead937 ] In O_APPEND & O_DIRECT mode, the data from different writers will be possibly overlapping each other since they take the shared lock. For example, both Writer1 and Writer2 are in O_APPEND and O_DIRECT mode: Writer1 Writer2 shared_lock() shared_lock() getattr(CAP_SIZE) getattr(CAP_SIZE) iocb->ki_pos = EOF iocb->ki_pos = EOF write(data1) write(data2) shared_unlock() shared_unlock() The data2 will overlap the data1 from the same file offset, the old EOF. Switch to exclusive lock instead when O_APPEND is specified. Signed-off-by: Xiubo Li Reviewed-by: Jeff Layton Signed-off-by: Ilya Dryomov Signed-off-by: Sasha Levin --- fs/ceph/file.c | 17 +++++++++++------ 1 file changed, 11 insertions(+), 6 deletions(-) diff --git a/fs/ceph/file.c b/fs/ceph/file.c index 11929d2bb594c..cd09e63d682b7 100644 --- a/fs/ceph/file.c +++ b/fs/ceph/file.c @@ -1418,6 +1418,7 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) struct ceph_cap_flush *prealloc_cf; ssize_t count, written = 0; int err, want, got; + bool direct_lock = false; loff_t pos; loff_t limit = max(i_size_read(inode), fsc->max_file_size); @@ -1428,8 +1429,11 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) if (!prealloc_cf) return -ENOMEM; + if ((iocb->ki_flags & (IOCB_DIRECT | IOCB_APPEND)) == IOCB_DIRECT) + direct_lock = true; + retry_snap: - if (iocb->ki_flags & IOCB_DIRECT) + if (direct_lock) ceph_start_io_direct(inode); else ceph_start_io_write(inode); @@ -1519,14 +1523,15 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) /* we might need to revert back to that point */ data = *from; - if (iocb->ki_flags & IOCB_DIRECT) { + if (iocb->ki_flags & IOCB_DIRECT) written = ceph_direct_read_write(iocb, &data, snapc, &prealloc_cf); - ceph_end_io_direct(inode); - } else { + else written = ceph_sync_write(iocb, &data, pos, snapc); + if (direct_lock) + ceph_end_io_direct(inode); + else ceph_end_io_write(inode); - } if (written > 0) iov_iter_advance(from, written); ceph_put_snap_context(snapc); @@ -1577,7 +1582,7 @@ static ssize_t ceph_write_iter(struct kiocb *iocb, struct iov_iter *from) goto out_unlocked; out: - if (iocb->ki_flags & IOCB_DIRECT) + if (direct_lock) ceph_end_io_direct(inode); else ceph_end_io_write(inode); -- 2.20.1