From: "Darrick J. Wong" <darrick.wong@oracle.com>
To: Qu Wenruo <quwenruo@cn.fujitsu.com>
Cc: eguan@redhat.com, linux-xfs@vger.kernel.org,
"Darrick J. Wong" <djwong@djwong.org>,
fstests@vger.kernel.org
Subject: Re: [PATCH 3/4] reflink: test adjacency of reflinked blocks
Date: Tue, 21 Mar 2017 10:23:52 -0400 [thread overview]
Message-ID: <20170321142352.GA5722@birch.djwong.org> (raw)
In-Reply-To: <6511f176-ee8e-09d8-d955-8aed36abaa29@cn.fujitsu.com>
On Tue, Mar 21, 2017 at 03:14:36PM +0800, Qu Wenruo wrote:
> Hi Darrick
>
> At 02/25/2017 09:12 AM, Darrick J. Wong wrote:
> >From: Darrick J. Wong <darrick.wong@oracle.com>
> >
> >If we reflink a file with N blocks to another file one block at a time,
> >does the destination file end up with the same number of extents as the
> >source file? In other words, does the filesystem succeed at combining
> >adjacent mappings into a maximal extents?
>
> Btrfs can't pass the test since it doesn't merge such mapping, not only the
> fiemap result, but also on-disk extents.
Weirdly, I observed that if you unmount and remount the btrfs it
actually /will/ merge the extents.
> IIRC ocfs also merges it, so I wonder if this is a documented behavior to
> merge any extent maps that are adjacent.
Generally, yes you'd expect that two logically and physically contiguous
extents to be merged together. That might change a bit for compressed
or encrypted extents.
> Further more, if a fs has a limit on extent size on disk, e.g 128M.
> But 2 file extents are also adjacent on disk, then should fiemap return a
> large 256M map or just 2 adjacent 128M maps?
It's up to the fs, really... XFS will merge the extent records (up to
maximum extent size) and then unmerge the shared/not-shared FIEMAP
records for accurate shared extent reporting.
--D
>
> Thanks,
> Qu
> >
> >Signed-off-by: Darrick J. Wong <djwong@djwong.org>
> >---
> > tests/generic/930 | 106 +++++++++++++++++++++++++++++++++++++++++++++++++
> > tests/generic/930.out | 11 +++++
> > tests/generic/group | 1
> > 3 files changed, 118 insertions(+)
> > create mode 100755 tests/generic/930
> > create mode 100644 tests/generic/930.out
> >
> >
> >diff --git a/tests/generic/930 b/tests/generic/930
> >new file mode 100755
> >index 0000000..15d8cbf
> >--- /dev/null
> >+++ b/tests/generic/930
> >@@ -0,0 +1,106 @@
> >+#! /bin/bash
> >+# FS QA Test No. 930
> >+#
> >+# Check that reflinking adjacent blocks in a file produces a single
> >+# block mapping extent.
> >+#
> >+#-----------------------------------------------------------------------
> >+# Copyright (c) 2017 Oracle, Inc. All Rights Reserved.
> >+#
> >+# This program is free software; you can redistribute it and/or
> >+# modify it under the terms of the GNU General Public License as
> >+# published by the Free Software Foundation.
> >+#
> >+# This program is distributed in the hope that it would be useful,
> >+# but WITHOUT ANY WARRANTY; without even the implied warranty of
> >+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> >+# GNU General Public License for more details.
> >+#
> >+# You should have received a copy of the GNU General Public License
> >+# along with this program; if not, write the Free Software Foundation,
> >+# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
> >+#-----------------------------------------------------------------------
> >+#
> >+
> >+seq=`basename $0`
> >+seqres=$RESULT_DIR/$seq
> >+echo "QA output created by $seq"
> >+
> >+here=`pwd`
> >+tmp=/tmp/$$
> >+status=1 # failure is the default!
> >+trap "_cleanup; exit \$status" 0 1 2 3 7 15
> >+
> >+_cleanup()
> >+{
> >+ cd /
> >+ rm -rf $tmp.*
> >+ wait
> >+}
> >+
> >+# get standard environment, filters and checks
> >+. ./common/rc
> >+. ./common/filter
> >+. ./common/reflink
> >+
> >+# real QA test starts here
> >+_supported_os Linux
> >+_supported_fs generic
> >+_require_scratch_reflink
> >+_require_fiemap
> >+
> >+echo "Format and mount"
> >+_scratch_mkfs > $seqres.full 2>&1
> >+_scratch_mount >> $seqres.full 2>&1
> >+
> >+testdir=$SCRATCH_MNT/test-$seq
> >+mkdir $testdir
> >+
> >+blocks=32
> >+blksz=65536
> >+sz=$((blocks * blksz))
> >+
> >+echo "Create the original files"
> >+$XFS_IO_PROG -f -c "falloc 0 $sz" $testdir/file1 >> $seqres.full
> >+_pwrite_byte 0x61 0 $sz $testdir/file1 >> $seqres.full
> >+seq 0 $blksz $((sz - blksz)) | while read offset; do
> >+ _reflink_range $testdir/file1 $offset $testdir/file2 $offset $blksz >> $seqres.full
> >+done
> >+
> >+echo "Compare files"
> >+md5sum $testdir/file1 | _filter_scratch
> >+md5sum $testdir/file2 | _filter_scratch
> >+
> >+echo "Check extent counts"
> >+f1=$(_count_extents $testdir/file1)
> >+f2=$(_count_extents $testdir/file2)
> >+s1=$($XFS_IO_PROG -c 'fiemap -v' $testdir/file1 | awk '{print $5}' | grep -c '0x.*[2367aAbBfF]...$')
> >+s2=$($XFS_IO_PROG -c 'fiemap -v' $testdir/file2 | awk '{print $5}' | grep -c '0x.*[2367aAbBfF]...$')
> >+
> >+# Did the fs combine the extent mappings when we made f2?
> >+test $f1 -eq $f2 || echo "f1 ($f1) != f2 ($f2)"
> >+test $s1 -eq $s2 || echo "s1 ($s1) != s2 ($s2)"
> >+test $f1 -eq $s1 || echo "f1 ($f1) != s1 ($f1)"
> >+test $f2 -eq $s2 || echo "f2 ($f2) != s2 ($f2)"
> >+
> >+_scratch_cycle_mount
> >+
> >+echo "Compare files after remounting"
> >+md5sum $testdir/file1 | _filter_scratch
> >+md5sum $testdir/file2 | _filter_scratch
> >+
> >+echo "Check extent counts"
> >+f1=$(_count_extents $testdir/file1)
> >+f2=$(_count_extents $testdir/file2)
> >+s1=$($XFS_IO_PROG -c 'fiemap -v' $testdir/file1 | awk '{print $5}' | grep -c '0x.*[2367aAbBfF]...$')
> >+s2=$($XFS_IO_PROG -c 'fiemap -v' $testdir/file2 | awk '{print $5}' | grep -c '0x.*[2367aAbBfF]...$')
> >+
> >+# Are the mappings still combined?
> >+test $f1 -eq $f2 || echo "f1 ($f1) != f2 ($f2)"
> >+test $s1 -eq $s2 || echo "s1 ($s1) != s2 ($s2)"
> >+test $f1 -eq $s1 || echo "f1 ($f1) != s1 ($f1)"
> >+test $f2 -eq $s2 || echo "f2 ($f2) != s2 ($f2)"
> >+
> >+# success, all done
> >+status=0
> >+exit
> >diff --git a/tests/generic/930.out b/tests/generic/930.out
> >new file mode 100644
> >index 0000000..556108a
> >--- /dev/null
> >+++ b/tests/generic/930.out
> >@@ -0,0 +1,11 @@
> >+QA output created by 930
> >+Format and mount
> >+Create the original files
> >+Compare files
> >+de89461b64701958984c95d1bfb0065a SCRATCH_MNT/test-930/file1
> >+de89461b64701958984c95d1bfb0065a SCRATCH_MNT/test-930/file2
> >+Check extent counts
> >+Compare files after remounting
> >+de89461b64701958984c95d1bfb0065a SCRATCH_MNT/test-930/file1
> >+de89461b64701958984c95d1bfb0065a SCRATCH_MNT/test-930/file2
> >+Check extent counts
> >diff --git a/tests/generic/group b/tests/generic/group
> >index d0bc47d..4b4df66 100644
> >--- a/tests/generic/group
> >+++ b/tests/generic/group
> >@@ -411,3 +411,4 @@
> > 406 auto quick dangerous
> > 407 auto quick clone metadata
> > 408 auto quick clone dedupe metadata
> >+930 auto quick clone
> >
> >--
> >To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> >the body of a message to majordomo@vger.kernel.org
> >More majordomo info at http://vger.kernel.org/majordomo-info.html
> >
> >
>
>
> --
> To unsubscribe from this list: send the line "unsubscribe linux-xfs" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
next prev parent reply other threads:[~2017-03-21 14:25 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-02-25 1:12 [PATCH 0/4] miscellaneous tests Darrick J. Wong
2017-02-25 1:12 ` [PATCH 1/4] xfs/348: classify this as a fuzzer test Darrick J. Wong
2017-02-28 14:05 ` Christoph Hellwig
2017-02-25 1:12 ` [PATCH 2/4] xfs: minor fixes to the fsmap tests Darrick J. Wong
2017-02-25 1:12 ` [PATCH 3/4] reflink: test adjacency of reflinked blocks Darrick J. Wong
2017-02-28 8:15 ` Eryu Guan
2017-02-28 16:31 ` Darrick J. Wong
2017-03-01 3:20 ` Eryu Guan
2017-03-21 7:14 ` Qu Wenruo
2017-03-21 14:23 ` Darrick J. Wong [this message]
2017-03-22 8:13 ` Qu Wenruo
2017-03-22 13:06 ` Darrick J. Wong
2017-02-25 1:13 ` [PATCH 4/4] xfs: stress test xfs_scrub Darrick J. Wong
2017-02-28 8:46 ` Eryu Guan
2017-02-28 17:42 ` Darrick J. Wong
2017-02-28 20:57 ` [PATCH v2 " Darrick J. Wong
2017-03-01 3:33 ` Eryu Guan
2017-03-01 6:08 ` Darrick J. Wong
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20170321142352.GA5722@birch.djwong.org \
--to=darrick.wong@oracle.com \
--cc=djwong@djwong.org \
--cc=eguan@redhat.com \
--cc=fstests@vger.kernel.org \
--cc=linux-xfs@vger.kernel.org \
--cc=quwenruo@cn.fujitsu.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).