From: Eryu Guan <guaneryu@gmail.com>
To: Zorro Lang <zlang@redhat.com>
Cc: fstests@vger.kernel.org
Subject: Re: [PATCH] generic: verify FIBMAP address overlap
Date: Sun, 28 Oct 2018 22:39:14 +0800 [thread overview]
Message-ID: <20181028143914.GG3876@desktop> (raw)
In-Reply-To: <20181024092356.25406-1-zlang@redhat.com>
On Wed, Oct 24, 2018 at 05:23:56PM +0800, Zorro Lang wrote:
> xfstests doesn't cover FIBMAP test, it cause we brought in a
> regression bug fixed by "79b3dbe4adb3 fs: fix iomap_bmap position
> calculation".
>
> Although FIBMAP is old, there're still some programs use it, likes
> LILO. This case tests if there's physical address overlap returned
> by FIBMAP.
>
> Signed-off-by: Zorro Lang <zlang@redhat.com>
> ---
> tests/generic/999 | 117 ++++++++++++++++++++++++++++++++++++++++++++++++++
> tests/generic/999.out | 3 ++
> tests/generic/group | 1 +
> 3 files changed, 121 insertions(+)
> create mode 100755 tests/generic/999
> create mode 100644 tests/generic/999.out
>
> diff --git a/tests/generic/999 b/tests/generic/999
> new file mode 100755
> index 00000000..3f70452d
> --- /dev/null
> +++ b/tests/generic/999
> @@ -0,0 +1,117 @@
> +#! /bin/bash
> +# SPDX-License-Identifier: GPL-2.0
> +# Copyright (c) 2018 Red Hat Inc. All Rights Reserved.
> +#
> +# FS QA Test No. 999
> +#
> +# Verify if there's physical address overlap returned by FIBMAP, cover:
> +# 79b3dbe4adb3 fs: fix iomap_bmap position calculation
> +#
> +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 15
> +
> +_cleanup()
> +{
> + cd /
> + rm -f $tmp.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# remove previous $seqres.full before test
> +rm -f $seqres.full
> +
> +# real QA test starts here
> +_supported_fs generic
> +_supported_os Linux
> +_require_scratch
> +_require_command "$FILEFRAG_PROG" filefrag
> +
> +TESTFILE="$SCRATCH_MNT/$seq-testfile"
Use lower case name for local variables.
> +
> +# Use filefrag -B option to force the use of the older FIBMAP ioctl instead of
> +# the FIEMAP ioctl. Then verify if there's map overlap.
> +verify_filefrag()
> +{
> + local n=1
> +
> + # record details in .full file
> + ${FILEFRAG_PROG} -Bes -v $TESTFILE >> $seqres.full
> +
> + # Due to physical things can't be golden image, so only output logical
> + # information at here
> + ${FILEFRAG_PROG} -Be $TESTFILE | _filter_filefrag | \
> + cut -d# -f1-2 > $tmp.filefrag
> +
> + # Verify there's not physical address overlay
> + for i in `cat $tmp.filefrag`; do
> + # Get the start(v1) and end(v2) addresses will be verified
> + v1=`sed -n "${n}p" $tmp.filefrag | cut -d# -f1`
> + v2=`sed -n "${n}p" $tmp.filefrag | cut -d# -f2`
> + # The 2nd value is length, so the real end addr is:
> + v2=$((v1 + v2))
> + # The end address maybe same with the next start address, so
> + # reduce 1 at here
> + ((v2--))
I don't think this is needed, we just have the extent range as [v1, v2),
which v1 is inclusive and v2 is exclusive.
> +
> + # Remove the line need to be verified ($i), compare with other
> + # lines one by one
> + sed -e "${n}d" $tmp.filefrag > $tmp.filefrag.tmp
> + for j in `cat $tmp.filefrag.tmp`; do
> + # Get 'next' line start(e1) and end(e2) addresses
> + e1=`echo $j | cut -d# -f1`
> + e2=`echo $j | cut -d# -f2`
> + # The 2nd value is length, so the real end addr is:
> + e2=$((e1 + e2))
> + # Verify there's not:
> + # [ e1 ... e2 ]
> + # [ v1 ... v2 ]
> + if [ ${v1} -ge ${e1} -a ${v1} -le ${e2} ]; then
> + echo "find physical addr overlap [$i] vs [$j]"
> + fi
> + # Verify there's not:
> + # [ e1 ... e2 ]
> + # [ v1 ... v2 ]
> + if [ ${v2} -ge ${e1} -a ${v2} -le ${e2} ]; then
> + echo "find physical addr overlap [$i] vs [$j]"
> + fi
So we only allow v1 >= e2 or e1 >= v2, otherwise we have a addr overlap.
> + done
> + ((n++))
> + done
> +}
> +
> +_scratch_mkfs > $seqres.full 2>&1
> +_scratch_mount
> +
> +# Check if FIBMAP is supported
> +echo "XX" > $TESTFILE
> +${FILEFRAG_PROG} -B $TESTFILE 2>&1 | \
> + grep -q "FIBMAP[[:space:]]*unsupported"
> +if [ $? -eq 0 ]; then
> + _notrun "FIBMAP isn't supported on current fs"
> +fi
> +rm -f $TESTFILE
Make it a new _require_fibmap() rule?
Thanks,
Eryu
> +
> +# Test
> +echo "== FIBMAP on empty file =="
> +$XFS_IO_PROG -f -c "truncate 0" $TESTFILE > /dev/null
> +verify_filefrag
> +
> +echo "== FIBMAP on sparse file =="
> +$XFS_IO_PROG -f -t -c "pwrite -S 0xaa 0 1m" \
> + -c "pwrite -S 0xaa 2m 1m" \
> + -c "pwrite -S 0xaa 4m 1m" \
> + $TESTFILE > /dev/null
> +verify_filefrag
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/999.out b/tests/generic/999.out
> new file mode 100644
> index 00000000..d3edbccf
> --- /dev/null
> +++ b/tests/generic/999.out
> @@ -0,0 +1,3 @@
> +QA output created by 999
> +== FIBMAP on empty file ==
> +== FIBMAP on sparse file ==
> diff --git a/tests/generic/group b/tests/generic/group
> index 47de9782..45f1d534 100644
> --- a/tests/generic/group
> +++ b/tests/generic/group
> @@ -519,3 +519,4 @@
> 514 auto quick clone
> 515 auto quick clone
> 516 auto quick dedupe clone
> +999 auto quick
> --
> 2.14.5
>
next prev parent reply other threads:[~2018-10-28 23:24 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-24 9:23 [PATCH] generic: verify FIBMAP address overlap Zorro Lang
2018-10-28 14:39 ` Eryu Guan [this message]
2018-11-01 7:48 ` Zorro Lang
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=20181028143914.GG3876@desktop \
--to=guaneryu@gmail.com \
--cc=fstests@vger.kernel.org \
--cc=zlang@redhat.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