From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mx3-rdu2.redhat.com ([66.187.233.73]:48040 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S968944AbeE3Js6 (ORCPT ); Wed, 30 May 2018 05:48:58 -0400 From: Andreas Gruenbacher To: cluster-devel@redhat.com, Christoph Hellwig Cc: linux-fsdevel@vger.kernel.org, Andreas Gruenbacher , Dave Chinner Subject: [PATCH v5 10/14] iomap: Complete partial direct I/O writes synchronously Date: Wed, 30 May 2018 11:48:38 +0200 Message-Id: <20180530094842.13559-11-agruenba@redhat.com> In-Reply-To: <20180530094842.13559-1-agruenba@redhat.com> References: <20180530094842.13559-1-agruenba@redhat.com> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: According to xfstest generic/240, applications seem to expect direct I/O writes to either complete as a whole or to fail; short direct I/O writes are apparently not appreciated. This means that when only part of an asynchronous direct I/O write succeeds, we can either fail the entire write, or we can wait for the partial write to complete and retry the remaining write using buffered I/O. The old __blockdev_direct_IO helper has code for waiting for partial writes to complete; the new iomap_dio_rw iomap helper does not. The above mentioned fallback mode is used by gfs2, which doesn't allow block allocations under direct I/O to avoid taking cluster-wide exclusive locks. As a consequence, an asynchronous direct I/O write to a file range that ends in a hole will result in a short write. When that happens, we want to complete the remaining write as buffered I/O. To allow that, change iomap_dio_rw to wait for short direct I/O writes like __blockdev_direct_IO does instead of returning -EIOCBQUEUED. This makes xfstest generic/240 work on gfs2. Signed-off-by: Andreas Gruenbacher Cc: Dave Chinner --- fs/iomap.c | 22 ++++++++++++---------- 1 file changed, 12 insertions(+), 10 deletions(-) diff --git a/fs/iomap.c b/fs/iomap.c index 27d97a290623..fcd3e26aadb7 100644 --- a/fs/iomap.c +++ b/fs/iomap.c @@ -720,6 +720,7 @@ struct iomap_dio { atomic_t ref; unsigned flags; int error; + bool wait_for_completion; union { /* used during submission and for synchronous completion: */ @@ -821,9 +822,8 @@ static void iomap_dio_bio_end_io(struct bio *bio) iomap_dio_set_error(dio, blk_status_to_errno(bio->bi_status)); if (atomic_dec_and_test(&dio->ref)) { - if (is_sync_kiocb(dio->iocb)) { + if (dio->wait_for_completion) { struct task_struct *waiter = dio->submit.waiter; - WRITE_ONCE(dio->submit.waiter, NULL); wake_up_process(waiter); } else if (dio->flags & IOMAP_DIO_WRITE) { @@ -1014,13 +1014,12 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, dio->end_io = end_io; dio->error = 0; dio->flags = 0; + dio->wait_for_completion = is_sync_kiocb(iocb); dio->submit.iter = iter; - if (is_sync_kiocb(iocb)) { - dio->submit.waiter = current; - dio->submit.cookie = BLK_QC_T_NONE; - dio->submit.last_queue = NULL; - } + dio->submit.waiter = current; + dio->submit.cookie = BLK_QC_T_NONE; + dio->submit.last_queue = NULL; if (iov_iter_rw(iter) == READ) { if (pos >= dio->i_size) @@ -1057,7 +1056,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, dio_warn_stale_pagecache(iocb->ki_filp); ret = 0; - if (iov_iter_rw(iter) == WRITE && !is_sync_kiocb(iocb) && + if (iov_iter_rw(iter) == WRITE && !dio->wait_for_completion && !inode->i_sb->s_dio_done_wq) { ret = sb_init_dio_done_wq(inode->i_sb); if (ret < 0) @@ -1072,8 +1071,10 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, iomap_dio_actor); if (ret <= 0) { /* magic error code to fall back to buffered I/O */ - if (ret == -ENOTBLK) + if (ret == -ENOTBLK) { + dio->wait_for_completion = true; ret = 0; + } break; } pos += ret; @@ -1086,8 +1087,9 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *iter, if (ret < 0) iomap_dio_set_error(dio, ret); + smp_mb__before_atomic(); if (!atomic_dec_and_test(&dio->ref)) { - if (!is_sync_kiocb(iocb)) + if (!dio->wait_for_completion) return -EIOCBQUEUED; for (;;) { -- 2.17.0