From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from userp2120.oracle.com ([156.151.31.85]:44204 "EHLO
        userp2120.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1726029AbfIZQIV (ORCPT
        <rfc822;fstests@vger.kernel.org>); Thu, 26 Sep 2019 12:08:21 -0400
Date: Thu, 26 Sep 2019 09:06:13 -0700
From: "Darrick J. Wong" <darrick.wong@oracle.com>
Subject: Re: [PATCH] generic: fallocate two bytes at block boundary
Message-ID: <20190926160613.GF9913@magnolia>
References: <20190926152927.2331-1-mreitz@redhat.com>
 <20190926155513.GE9913@magnolia>
 <c8cf492f-e3e0-e96f-f8f9-49b42b546cf0@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <c8cf492f-e3e0-e96f-f8f9-49b42b546cf0@redhat.com>
Sender: fstests-owner@vger.kernel.org
Content-Transfer-Encoding: quoted-printable
To: Max Reitz <mreitz@redhat.com>
Cc: fstests@vger.kernel.org
List-ID: <fstests@vger.kernel.org>

On Thu, Sep 26, 2019 at 06:02:17PM +0200, Max Reitz wrote:
> On 26.09.19 17:55, Darrick J. Wong wrote:
> > On Thu, Sep 26, 2019 at 05:29:27PM +0200, Max Reitz wrote:
> >> Allocating two bytes at a block boundary with fallocate should alloc=
ate
> >> both blocks involved.  Test this by writing both bytes with dd
> >> afterwards and see whether the on-disk size increases (it should not=
).
> >>
> >> Signed-off-by: Max Reitz <mreitz@redhat.com>
> >> ---
> >>  tests/generic/568     | 63 ++++++++++++++++++++++++++++++++++++++++=
+++
> >>  tests/generic/568.out |  2 ++
> >>  tests/generic/group   |  1 +
> >>  3 files changed, 66 insertions(+)
> >>  create mode 100755 tests/generic/568
> >>  create mode 100644 tests/generic/568.out
> >>
> >> diff --git a/tests/generic/568 b/tests/generic/568
> >> new file mode 100755
> >> index 00000000..8fbdcda0
> >> --- /dev/null
> >> +++ b/tests/generic/568
> >> @@ -0,0 +1,63 @@
> >> +#! /bin/bash
> >> +# SPDX-License-Identifier: GPL-2.0
> >> +# Copyright (c) 2019 Red Hat, Inc.  All Rights Reserved.
> >> +#
> >> +# FS QA Test No. generic/568
> >> +#
> >> +# Test that fallocating an unaligned range allocates all blocks
> >> +# touched by that range
> >> +#
> >> +seq=3D$(basename $0)
> >> +seqres=3D"$RESULT_DIR/$seq"
> >> +echo "QA output created by $seq"
> >> +
> >> +here=3D$PWD
> >> +tmp=3D/tmp/$$
> >> +status=3D1	# failure is the default!
> >> +trap "_cleanup; exit \$status" 0 1 2 3 15
> >> +
> >> +_cleanup()
> >> +{
> >> +	cd /
> >> +	rm -f "$tmp".*
> >> +}
> >> +
> >> +# get standard environment, filters and checks
> >> +. ./common/rc
> >> +. ./common/filter
> >> +
> >> +# real QA test starts here
> >> +_supported_fs generic
> >> +_supported_os Linux
> >> +_require_scratch
> >> +
> >> +testfile=3D"$SCRATCH_MNT/testfile"
> >> +
> >> +_scratch_mkfs > /dev/null 2>&1
> >> +_scratch_mount
> >> +
> >> +# Fallocate 2 bytes across a block boundary
> >> +block_size=3D$(stat -fc '%S' "$SCRATCH_MNT")
> >=20
> > block_size=3D$(_get_file_block_size $SCRATCH_MNT)
>=20
> Ah, nice.
>=20
> >> +fallocate -o $((block_size - 1)) -l 2 "$testfile"
> >=20
> > If you're going to use an external program, you need to gate the test=
 on
> > whether or not the program's installed, by calling _require_command.
>=20
> OK.
>=20
> > Though probably the easier way would be to use xfs_io since fstests
> > requires that xfsprogs be installed:
> >=20
> > 	$XFS_IO_PROG -f -c "falloc $((block_size - 1)) 2" $testfile
> >=20
> > Though you do still have to put at the top of the test:
> >=20
> > 	_require_xfs_io_command "falloc"
> >=20
> > Because not all filesystems support fallocate.
>=20
> So I suppose as long as one doesn=E2=80=99t use special XFS commands, x=
fs_io is
> filesystem-agnostic?

Right.  At this point xfs_io is largely a testing vehicle for fstests
scripts to call various system calls or ioctls, some of which aren't
even supported on XFS itself. :P

--D

> >> +
> >> +# Both the first blocks should be allocated now.  Check that by
> >> +# inquiring whether the file grows when we write to the two bytes w=
e
> >> +# have just fallocated.
> >> +
> >> +allocated_size_before=3D$(($(stat -c '%b * %B' "$testfile")))
> >> +
> >> +dd if=3D/dev/zero of=3D"$testfile" bs=3D1 conv=3Dnotrunc \
> >> +    seek=3D$((block_size - 1)) count=3D2 \
> >> +    2>&1 | _filter_dd
> >=20
> > $XFS_IO_PROG -c "pwrite $((block_size - 1)) 2" $testfile
> >=20
> >> +
> >> +allocated_size_after=3D$(($(stat -c '%b * %B' "$testfile")))
> >> +
> >> +if [ $allocated_size_after -gt $allocated_size_before ]; then
> >> +	echo "ERROR: File grew from ${allocated_size_before} B to" \
> >> +	     "${allocated_size_after} when writing to the fallocated range=
."
> >> +else
> >> +	echo "OK: File did not grow."
> >=20
> > Other than that, the logic makes sense to me.  Thanks for writing thi=
s
> > up!
>=20
> OK, thanks, I=E2=80=99ll prepare a v2.
>=20
> Max
>=20