Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Sasha Levin <Alexander.Levin@microsoft.com>
To:     "stable@vger.kernel.org" <stable@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
CC:     Andreas Gruenbacher <agruenba@redhat.com>,
        Christoph Hellwig <hch@lst.de>,
        "Darrick J . Wong" <darrick.wong@oracle.com>,
        Sasha Levin <Alexander.Levin@microsoft.com>
Subject: [PATCH AUTOSEL 4.14 74/87] iomap: complete partial direct I/O writes
 synchronously
Thread-Topic: [PATCH AUTOSEL 4.14 74/87] iomap: complete partial direct I/O
 writes synchronously
Thread-Index: AQHUTjL41lQb7aNja0O9+230DB7mBA==
Date:   Mon, 17 Sep 2018 03:03:12 +0000
Message-ID: <20180917030220.245686-74-alexander.levin@microsoft.com>
References: <20180917030220.245686-1-alexander.levin@microsoft.com>
In-Reply-To: <20180917030220.245686-1-alexander.levin@microsoft.com>
Accept-Language: en-US
Content-Language: en-US
received-spf: None (protection.outlook.com: microsoft.com does not designate
 permitted sender hosts)
spamdiagnosticoutput: 1:99
spamdiagnosticmetadata: NSPM
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: b23e3a1c-6c4c-42ff-04f0-08d61c4a5b01
X-MS-Exchange-CrossTenant-originalarrivaltime: 17 Sep 2018 03:03:12.1184
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 72f988bf-86f1-41af-91ab-2d7cd011db47
X-MS-Exchange-Transport-CrossTenantHeadersStamped: CY4PR21MB0133
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

From: Andreas Gruenbacher <agruenba@redhat.com>

[ Upstream commit ebf00be37de35788cad72f4f20b4a39e30c0be4a ]

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 as 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 needed for 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 contains a hole will result in a short write.  In that
case, wait for the short write to complete to allow gfs2 to recover.

Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
 fs/iomap.c | 21 +++++++++++----------
 1 file changed, 11 insertions(+), 10 deletions(-)

diff --git a/fs/iomap.c b/fs/iomap.c
index d4801f8dd4fd..8f7673a69273 100644
--- a/fs/iomap.c
+++ b/fs/iomap.c
@@ -693,6 +693,7 @@ struct iomap_dio {
 	atomic_t		ref;
 	unsigned		flags;
 	int			error;
+	bool			wait_for_completion;
=20
 	union {
 		/* used during submission and for synchronous completion: */
@@ -793,9 +794,8 @@ static void iomap_dio_bio_end_io(struct bio *bio)
 		iomap_dio_set_error(dio, blk_status_to_errno(bio->bi_status));
=20
 	if (atomic_dec_and_test(&dio->ref)) {
-		if (is_sync_kiocb(dio->iocb)) {
+		if (dio->wait_for_completion) {
 			struct task_struct *waiter =3D dio->submit.waiter;
-
 			WRITE_ONCE(dio->submit.waiter, NULL);
 			wake_up_process(waiter);
 		} else if (dio->flags & IOMAP_DIO_WRITE) {
@@ -980,13 +980,12 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *ite=
r,
 	dio->end_io =3D end_io;
 	dio->error =3D 0;
 	dio->flags =3D 0;
+	dio->wait_for_completion =3D is_sync_kiocb(iocb);
=20
 	dio->submit.iter =3D iter;
-	if (is_sync_kiocb(iocb)) {
-		dio->submit.waiter =3D current;
-		dio->submit.cookie =3D BLK_QC_T_NONE;
-		dio->submit.last_queue =3D NULL;
-	}
+	dio->submit.waiter =3D current;
+	dio->submit.cookie =3D BLK_QC_T_NONE;
+	dio->submit.last_queue =3D NULL;
=20
 	if (iov_iter_rw(iter) =3D=3D READ) {
 		if (pos >=3D dio->i_size)
@@ -1016,7 +1015,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *ite=
r,
 	WARN_ON_ONCE(ret);
 	ret =3D 0;
=20
-	if (iov_iter_rw(iter) =3D=3D WRITE && !is_sync_kiocb(iocb) &&
+	if (iov_iter_rw(iter) =3D=3D WRITE && !dio->wait_for_completion &&
 	    !inode->i_sb->s_dio_done_wq) {
 		ret =3D sb_init_dio_done_wq(inode->i_sb);
 		if (ret < 0)
@@ -1031,8 +1030,10 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *it=
er,
 				iomap_dio_actor);
 		if (ret <=3D 0) {
 			/* magic error code to fall back to buffered I/O */
-			if (ret =3D=3D -ENOTBLK)
+			if (ret =3D=3D -ENOTBLK) {
+				dio->wait_for_completion =3D true;
 				ret =3D 0;
+			}
 			break;
 		}
 		pos +=3D ret;
@@ -1046,7 +1047,7 @@ iomap_dio_rw(struct kiocb *iocb, struct iov_iter *ite=
r,
 		iomap_dio_set_error(dio, ret);
=20
 	if (!atomic_dec_and_test(&dio->ref)) {
-		if (!is_sync_kiocb(iocb))
+		if (!dio->wait_for_completion)
 			return -EIOCBQUEUED;
=20
 		for (;;) {
--=20
2.17.1