From: Josef Bacik <jbacik@fb.com>
To: Koen De Wit <koen.de.wit@oracle.com>, <xfs@oss.sgi.com>
Cc: <linux-btrfs@vger.kernel.org>
Subject: Re: [PATCH v2] xfstests: Btrfs: add test for large metadata blocks
Date: Mon, 10 Feb 2014 14:54:00 -0500 [thread overview]
Message-ID: <52F92E58.3060908@fb.com> (raw)
In-Reply-To: <1391848230-26761-1-git-send-email-koen.de.wit@oracle.com>
On 02/08/2014 03:30 AM, Koen De Wit wrote:
> Tests Btrfs filesystems with all possible metadata block sizes, by
> setting large extended attributes on files.
>
> Signed-off-by: Koen De Wit <koen.de.wit@oracle.com>
> ---
>
> v1->v2:
> - Fix indentation: 8 spaces instead of 4
> - Move _scratch_unmount to end of loop, add _check_scratch_fs
> - Sending failure messages of mkfs.btrfs to output instead of
> $seqres.full
>
> diff --git a/tests/btrfs/036 b/tests/btrfs/036
> new file mode 100644
> index 0000000..b14697d
> --- /dev/null
> +++ b/tests/btrfs/036
> @@ -0,0 +1,137 @@
> +#! /bin/bash
> +# FS QA Test No. 036
> +#
> +# Tests large metadata blocks in btrfs, which allows large extended
> +# attributes.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2014, Oracle and/or its affiliates. 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`
> +status=1 # failure is the default!
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +
> +_supported_fs btrfs
> +_supported_os Linux
> +_require_scratch
> +_need_to_be_root
> +
> +rm -f $seqres.full
> +
> +pagesize=`$here/src/feature -s`
> +pagesize_kb=`expr $pagesize / 1024`
> +
> +# Test all valid leafsizes
> +for leafsize in `seq $pagesize_kb $pagesize_kb 64`; do
> + _scratch_mkfs -l ${leafsize}K >/dev/null
> + _scratch_mount
> + # Calculate the size of the extended attribute value, leaving
> + # 512 bytes for other metadata.
> + xattr_size=`expr $leafsize \* 1024 - 512`
> +
> + touch $SCRATCH_MNT/emptyfile
> + # smallfile will be inlined, bigfile not.
> + $XFS_IO_PROG -f -c "pwrite 0 100" $SCRATCH_MNT/smallfile \
> + >/dev/null
> + $XFS_IO_PROG -f -c "pwrite 0 9000" $SCRATCH_MNT/bigfile \
> + >/dev/null
> + ln -s $SCRATCH_MNT/bigfile $SCRATCH_MNT/bigfile_softlink
> +
> + files=(emptyfile smallfile bigfile bigfile_softlink)
> + chars=(a b c d)
> + for i in `seq 0 1 3`; do
> + char=${chars[$i]}
> + file=$SCRATCH_MNT/${files[$i]}
> + lnkfile=${file}_hardlink
> + ln $file $lnkfile
> + xattr_value=`head -c $xattr_size < /dev/zero \
> + | tr '\0' $char`
> +
> + set_md5=`echo -n "$xattr_value" | md5sum`
> + ${ATTR_PROG} -Lq -s attr_$char -V $xattr_value $file
> + get_md5=`${ATTR_PROG} -Lq -g attr_$char $file | md5sum`
> + get_ln_md5=`${ATTR_PROG} -Lq -g attr_$char $lnkfile \
> + | md5sum`
> +
> + # Using md5sums for comparison instead of the values
> + # themselves because bash command lines cannot be larger
> + # than 64K chars.
> + if [ "$set_md5" != "$get_md5" ]; then
> + echo -n "Got unexpected xattr value for "
> + echo -n "attr_$char from file ${file}. "
> + echo "(leafsize is ${leafsize}K)"
> + fi
> + if [ "$set_md5" != "$get_ln_md5" ]; then
> + echo -n "Value for attr_$char differs for "
> + echo -n "$file and ${lnkfile}. "
> + echo "(leafsize is ${leafsize}K)"
> + fi
> + done
> +
> + # Test attributes with a size larger than the leafsize.
> + # Should result in an error.
> + if [ "$leafsize" -lt "64" ]; then
> + # Bash command lines cannot be larger than 64K
> + # characters, so we do not test attribute values
> + # with a size >64KB.
> + xattr_size=`expr $leafsize \* 1024 + 512`
> + xattr_value=`head -c $xattr_size < /dev/zero | tr '\0' x`
> + ${ATTR_PROG} -q -s attr_toobig -V $xattr_value \
> + $SCRATCH_MNT/emptyfile >> $seqres.full 2>&1
> + if [ "$?" -eq "0" ]; then
> + echo -n "Expected error, xattr_size is bigger "
> + echo "than ${leafsize}K"
> + fi
> + fi
> +
> + _scratch_unmount >/dev/null 2>&1
> + _check_scratch_fs
> +done
> +
> +_scratch_mount
> +
> +# Illegal attribute name (more than 256 characters)
> +attr_name=`head -c 260 < /dev/zero | tr '\0' n`
> +${ATTR_PROG} -s $attr_name -V attribute_name_too_big \
> + $SCRATCH_MNT/emptyfile 2>&1 | head -n 1
> +
> +_scratch_unmount
> +
> +_filter_size_numbers() {
> + sed "s/$pagesize/PAGESIZE/g" | sed "s/$1/LEAFSIZE/g"
> +}
> +
> +# Some illegal leafsizes
> +_scratch_mkfs -l 0 2>&1 | _filter_size_numbers 0
> +_scratch_mkfs -l 5678 2>&1 | _filter_size_numbers 5678
> +lsize=`expr $pagesize / 2 + $pagesize`
> +_scratch_mkfs -l $lsize 2>&1 | _filter_size_numbers $lsize
> +_scratch_mkfs -l 128K 2>&1 | _filter_size_numbers 131072
> +_scratch_mkfs -l K
This fails if I have MKFS_OPTIONS="-O skinny-metadata" because the
output from turning on skinny metadata is showing up. Thanks,
Josef
prev parent reply other threads:[~2014-02-10 19:54 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-02-08 8:30 [PATCH v2] xfstests: Btrfs: add test for large metadata blocks Koen De Wit
2014-02-10 19:54 ` Josef Bacik [this message]
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=52F92E58.3060908@fb.com \
--to=jbacik@fb.com \
--cc=koen.de.wit@oracle.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=xfs@oss.sgi.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).