Re: [PATCH 05/11] xfs: Check for extent overflow when adding/removing xattrs

[Chandan Babu R <chandanrlinux@gmail.com>]

On Saturday 14 November 2020 6:04:40 AM IST Darrick J. Wong wrote:
> On Fri, Nov 13, 2020 at 04:56:57PM +0530, Chandan Babu R wrote:
> > This test verifies that XFS does not cause inode fork's extent count to
> > overflow when adding/removing xattrs.
> > 
> > Signed-off-by: Chandan Babu R <chandanrlinux@gmail.com>
> > ---
> >  tests/xfs/525     | 154 ++++++++++++++++++++++++++++++++++++++++++++++
> >  tests/xfs/525.out |  16 +++++
> >  tests/xfs/group   |   1 +
> >  3 files changed, 171 insertions(+)
> >  create mode 100755 tests/xfs/525
> >  create mode 100644 tests/xfs/525.out
> > 
> > diff --git a/tests/xfs/525 b/tests/xfs/525
> > new file mode 100755
> > index 00000000..1d5d6e7c
> > --- /dev/null
> > +++ b/tests/xfs/525
> > @@ -0,0 +1,154 @@
> > +#! /bin/bash
> > +# SPDX-License-Identifier: GPL-2.0
> > +# Copyright (c) 2020 Chandan Babu R.  All Rights Reserved.
> > +#
> > +# FS QA Test 525
> > +#
> > +# Verify that XFS does not cause inode fork's extent count to overflow when
> > +# Adding/removing xattrs.
> > +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
> > +. ./common/attr
> > +. ./common/inject
> > +
> > +# remove previous $seqres.full before test
> > +rm -f $seqres.full
> > +
> > +# real QA test starts here
> > +
> > +_supported_fs xfs
> > +_require_scratch
> > +_require_attrs
> > +_require_xfs_debug
> > +_require_test_program "punch-alternating"
> > +_require_xfs_io_error_injection "reduce_max_iextents"
> > +_require_xfs_io_error_injection "bmap_alloc_minlen_extent"
> > +
> > +attr_set()
> > +{
> > +	echo "* Set xattrs"
> > +
> > +	echo "Format and mount fs"
> > +	_scratch_mkfs_sized $((1024 * 1024 * 1024)) >> $seqres.full
> > +	_scratch_mount >> $seqres.full
> > +
> > +	bsize=$(_get_block_size $SCRATCH_MNT)
> > +
> > +	testfile=$SCRATCH_MNT/testfile
> > +
> > +	echo "Consume free space"
> > +	dd if=/dev/zero of=${testfile} bs=${bsize} >> $seqres.full 2>&1
> > +	sync
> > +
> > +	echo "Create fragmented filesystem"
> > +	$here/src/punch-alternating $testfile >> $seqres.full
> > +	sync
> > +
> > +	echo "Inject reduce_max_iextents error tag"
> > +	xfs_io -x -c 'inject reduce_max_iextents' $SCRATCH_MNT
> > +
> > +	echo "Inject bmap_alloc_minlen_extent error tag"
> > +	xfs_io -x -c 'inject bmap_alloc_minlen_extent' $SCRATCH_MNT
> > +
> > +	echo "Create xattrs"
> > +
> > +	attr_len=$(uuidgen | wc -c)
> > +	nr_attrs=$((bsize * 20 / attr_len))
> > +	for i in $(seq 1 $nr_attrs); do
> > +		$SETFATTR_PROG -n "trusted.""$(uuidgen)" $testfile \
> 
> Does this test require UUIDs in the attr names?  Why wouldn't
> $(printf "%037d" $i) suffice for this purpose?

You are right. I can replace executing uuidgen with calls to printf.

> 
> Though if you insist upon using uuids, please call $UUIDGEN_PROG per
> fstest custom.
> 
> > +			 >> $seqres.full 2>&1
> > +		[[ $? != 0 ]] && break
> > +	done
> > +
> > +	testino=$(stat -c "%i" $testfile)
> > +
> > +	_scratch_unmount >> $seqres.full
> > +
> > +	echo "Verify uquota inode's extent count"
> 
> Huh?  I thought we were testing file attrs?

Sorry, I will fix that.

> 
> > +	nextents=$(_scratch_get_iext_count $testino attr || \
> > +			_fail "Unable to obtain inode fork's extent count")
> > +	if (( $nextents > 10 )); then
> > +		echo "Extent count overflow check failed: nextents = $nextents"
> > +		exit 1
> > +	fi
> > +}
> > +
> > +attr_remove()
> > +{
> > +	echo "* Remove xattrs"
> > +
> > +	echo "Format and mount fs"
> > +	_scratch_mkfs_sized $((1024 * 1024 * 1024)) >> $seqres.full
> > +	_scratch_mount >> $seqres.full
> > +
> > +	bsize=$(_get_block_size $SCRATCH_MNT)
> > +
> > +	testfile=$SCRATCH_MNT/testfile
> > +
> > +	echo "Consume free space"
> > +	dd if=/dev/zero of=${testfile} bs=${bsize} >> $seqres.full 2>&1
> > +	sync
> > +
> > +	echo "Create fragmented filesystem"
> > +	$here/src/punch-alternating $testfile >> $seqres.full
> > +	sync
> > +
> > +	testino=$(stat -c "%i" $testfile)
> > +
> > +	naextents=0
> > +	last=""
> > +
> > +	attr_len=$(uuidgen | wc -c)
> > +	nr_attrs=$((bsize / attr_len))
> > +
> > +	echo "Create initial xattr extents"
> > +	while (( $naextents < 4 )); do
> > +		xfs_io -x -c 'inject bmap_alloc_minlen_extent' $SCRATCH_MNT
> > +
> > +		for i in $(seq 1 $nr_attrs); do
> > +			last="trusted.""$(uuidgen)"
> > +			$SETFATTR_PROG -n $last $testfile
> > +		done
> > +
> > +		_scratch_unmount >> $seqres.full
> > +
> > +		naextents=$(_scratch_get_iext_count $testino attr || \
> > +				_fail "Unable to obtain inode fork's extent count")
> > +
> > +		_scratch_mount >> $seqres.full
> > +	done
> > +
> > +	echo "Inject reduce_max_iextents error tag"
> > +	xfs_io -x -c 'inject reduce_max_iextents' $SCRATCH_MNT
> > +
> > +	echo "Remove xattr to trigger -EFBIG"
> > +	$SETFATTR_PROG -x "$last" $testfile >> $seqres.full 2>&1
> > +	if [[ $? == 0 ]]; then
> > +		echo "Xattr removal succeeded; Should have failed "
> 
> So at this point the user has a file for which he can't ever remove the
> xattrs for fear of overflowing naextents.  The only way to clear this is
> to delete the file, so shouldn't you be testing that this succeeds?

Ok. I will add that. However there is another issue here. I think it would better
suited to discuss that w.r.t XFS_IEXT_DIR_MANIP_CNT tests (i.e. patch 6).

> 
> --D
> 
> > +		exit 1
> > +	fi
> > +}
> > +
> > +attr_set
> > +attr_remove
> > +
> > +# success, all done
> > +status=0
> > +exit
> > diff --git a/tests/xfs/525.out b/tests/xfs/525.out
> > new file mode 100644
> > index 00000000..cc40e6e2
> > --- /dev/null
> > +++ b/tests/xfs/525.out
> > @@ -0,0 +1,16 @@
> > +QA output created by 525
> > +* Set xattrs
> > +Format and mount fs
> > +Consume free space
> > +Create fragmented filesystem
> > +Inject reduce_max_iextents error tag
> > +Inject bmap_alloc_minlen_extent error tag
> > +Create xattrs
> > +Verify uquota inode's extent count
> > +* Remove xattrs
> > +Format and mount fs
> > +Consume free space
> > +Create fragmented filesystem
> > +Create initial xattr extents
> > +Inject reduce_max_iextents error tag
> > +Remove xattr to trigger -EFBIG
> > diff --git a/tests/xfs/group b/tests/xfs/group
> > index 3fa38c36..bd38aff0 100644
> > --- a/tests/xfs/group
> > +++ b/tests/xfs/group
> > @@ -522,3 +522,4 @@
> >  522 auto quick quota
> >  523 auto quick realtime growfs
> >  524 auto quick punch zero insert collapse
> > +525 auto quick attr
> 


-- 
chandan