On Tue, 2011-06-28 at 07:45 -0700, Allison Henderson wrote:
> This patch adds a new test 255 that tests that a hole can be punched even when the
> disk is full. Reserved blocks should be used to allow a punch hole to proceed even
> when there is not enough blocks to further fragment the file. To test this, the
> file system is fragmented by punching holes in regular intervals and filling
> the file system between punches. This will eventually force the file system to use
> reserved blocks to proceed with the punch hole operation.
>
> The work in this patch is a continuation from a previous patch set that has been
> partially accepted.
>
> Signed-off-by: Allison Henderson <achender@xxxxxxxxxxxxxxxxxx>
First of all, I renumbered the test 256 (and will do
that for you if necessary before committing the change).
Second, I can confirm that this new test passes for
XFS on my setup. I don't know what the threshold for
inclusion in the "quick" group is, but it's under a
minute so that's fine with me.
Third, maybe we can create a new group "punch" or
something like that to tag tests that exercise
hole punching functionality (tests 252, 255, and
now this one at least; maybe 175 and 176 too).
Finally, I found a bug and a few other spots that
really ought to be fixed but aren't really serious
problems. I also have a number of other things that
I commented on, but tried to make it clear where
they're just remarks to consider, not requests to
make a change.
-Alex
> ---
> v5->v6
>
> Test moved out of 252 and put in its own test 255
>
> _fill_fs and _test_full_fs_punch have been moved from common.punch
> to test 255 and modified to use the _user_do routines in common.rc
>
> _fill_fs has been optimized to stop once files smaller than a block
> cannot be created.
>
> v6->v7
> Fixed bad file add
>
>
> :000000 100644 0000000... 3d39fdb... A 255
> :000000 100644 0000000... 3525403... A 255.out
> :100644 100644 1f86075... c045e70... M group
> 255 | 178 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
> 255.out | 1 +
> group | 1 +
> 3 files changed, 180 insertions(+), 0 deletions(-)
>
> diff --git a/255 b/255
> new file mode 100644
> index 0000000..3d39fdb
> --- /dev/null
> +++ b/255
> @@ -0,0 +1,178 @@
> +#! /bin/bash
> +# FS QA Test No. 255
> +#
> +# Test Full File System Hole Punching
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2011 IBM Corporation. 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
> +#
> +#-----------------------------------------------------------------------
> +#
> +# creator
> +owner=achender@xxxxxxxxxxxxxxxxxx
> +
> +seq=`basename $0`
> +echo "QA output created by $seq"
> +
> +here=`pwd`
> +tmp=/tmp/$$
> +status=1 # failure is the default!
> +
> +_cleanup()
> +{
> + rm -f $tmp.*
> +}
> +
> +trap "_cleanup ; exit \$status" 0 1 2 3 15
> +
> +# get standard environment, filters and checks
> +. ./common.rc
> +. ./common.filter
> +. ./common.punch
> +
> +# real QA test starts here
> +_supported_fs generic
> +_supported_os Linux
> +
> +_require_xfs_io_falloc_punch
> +_require_scratch
> +_require_user
> +
> +testfile=$TEST_DIR/255.$$
testfile=$TEST_DIR/$seq.$$
> +
> +
The _fill_fs function seems like it may be useful elsewhere,
so it may end up in common.rc eventually. (Easy enough to
move it there later though.)
> +# _fill_fs()
> +#
> +# Fills a file system by repeatedly creating files in the given folder
> +# starting with the given file size. Files are reduced in size when
> +# they can no longer fit untill no more files can be created.
> +#
> +# This routine is used by _test_full_fs_punch to test that a hole may
> +# still be punched when the disk is full by borrowing reserved blocks.
> +# All files are created as a non root user to prevent reserved blocks
> +# from being consumed.
> +#
> +_fill_fs() {
> + local file_size=$1
> + local dir=$2
> + local block_size=$3
> + local file_count=1
> + local bytes_written=0
> +
> + if [ $# -ne 3 ]
> + then
> + echo "USAGE: _fill_fs filesize dir block size"
> + exit 1
> + fi
> +
> + # Creation of files or folders
> + # must not be done as root or
> + # reserved blocks will be consumed
> + _user_do "mkdir -p $dir &> /dev/null"
I personally prefer seeing "> /dev/null 2>&1" rather
than "&> /dev/null", to make it that much more
explicit that both are being redirected. This is
purely a style thing, and I say this mainly to see
if anyone else has a preference one way or another.
(This appears several times in the file, so if you
do choose to change it, do so throughout.)
> + if [ $? -ne 0 ] ; then
> + return 0
> + fi
> +
> + if [ $file_size -lt $block_size ]
> + then
> + $file_size = $block_size
file_size=$block_size
(This one is a bug.)
> + fi
> +
> + while [ $file_size -ge $block_size ]
> + do
> + bytes_written=0
> + _user_do "$XFS_IO_PROG -F -f -c \"pwrite 0 $file_size\" $dir/$file_count.bin &> /dev/null"
> +
Try to split these long lines, e.g.:
_user_do "$XFS_IO_PROG -F -f -c \
\"pwrite 0 $file_size\" \
$dir/$file_count.bin &> /dev/null"
> + if [ -f $dir/$file_count.bin ]
> + then
> + bytes_written=`$XFS_IO_PROG -F -c "stat" $dir/$file_count.bin | grep size | cut -d ' ' -f3`
Personal preference again, but as long as we're using bash
and not sh, Use $(...) rather than `...` for sub-commands.
> + fi
> +
> + # If there was no room to make the file,
> + # then divide it in half, and keep going
> + if [ $bytes_written -lt $file_size ]
> + then
> + file_size=$(( $file_size / 2 ))
> + fi
> + file_count=$(( $file_count + 1 ))
> +
> + done
> +}
> +
> +# _test_full_fs_punch()
> +#
> +# This function will test that a hole may be punched
> +# even when the file system is full. Reserved blocks
> +# should be used to allow a punch hole to proceed even
> +# when there is not enough blocks to further fragment the
> +# file. To test this, this function will fragment the file
> +# system by punching holes in regular intervals and filling
> +# the file system between punches.
> +#
> +_test_full_fs_punch()
> +{
> + local hole_len=$1 # The length of the holes to punch
> + local hole_interval=$2 # The interval between the holes
> + local iterations=$3 # The number of holes to punch
> + local file_name=$4 # File to punch holes in
> + local block_size=$5 # File system block size
> + local file_len=$(( $(( $hole_len + $hole_interval )) * $iterations ))
> + local path=`dirname $file_name`
> + local hole_offset=0
> +
> + if [ $# -ne 5 ]
> + then
> + echo "USAGE: _test_full_fs_punch hole_len hole_interval iterations file_name block_size"
> + exit 1
> + fi
> +
> + rm -f $file_name &> /dev/null
> +
> + $XFS_IO_PROG -F -f -c "pwrite 0 $file_len" \
> + -c "fsync" $file_name &> /dev/null
> + chmod 666 $file_name
> +
> + _fill_fs $(( 1024 * 1024 * 1024 )) $path/fill $block_size
You are specifying a file size that's equal to the filesystem
size, which I think is guaranteed to fail its first pass. Maybe
start with half of the filesystem (or even 90% of it). (I
don't know whether this is a a real issue or not though.)
Also, you could maybe use a shell variable to represent the
filesystem size, since you use it twice in this script.
> +
> + for (( i=0; i<$iterations; i++ ))
> + do
> + # This part must not be done as root in order to
> + # test that reserved blocks are used when needed
> + _user_do "$XFS_IO_PROG -F -f -c \"fpunch $hole_offset $hole_len\" $file_name"
> + rc=$?
> + if [ $? -ne 0 ] ; then
> + echo Punch hole failed
> + break
> + fi
> +
> + hole_offset=$(( $hole_offset + $hole_len + $hole_interval ))
> +
> + _fill_fs $hole_len $path/fill.$i $block_size
> +
> + done
> +}
> +
> +# Make a small file system to fillkk
> +umount $SCRATCH_DEV &> /dev/null
> +_scratch_mkfs_sized $(( 1024 * 1024 * 1024 )) &> /dev/null
> +_scratch_mount
> +# Test must be able to write files with non-root permissions
> +chmod 777 $SCRATCH_MNT
> +
> +block_size=`stat -f $SCRATCH_DEV | grep "Block size" | cut -d " " -f3`
> +_test_full_fs_punch $(( $block_size * 2 )) $block_size 500 $SCRATCH_MNT/252.$$ $block_size
Use $SCRATCH_MNT/$seq.$$
...and possibly define this path in a variable
at the top of the file.
> +
> +status=0 ; exit
> diff --git a/255.out b/255.out
> new file mode 100644
> index 0000000..3525403
> --- /dev/null
> +++ b/255.out
> @@ -0,0 +1 @@
> +QA output created by 255
> diff --git a/group b/group
> index 1f86075..c045e70 100644
> --- a/group
> +++ b/group
> @@ -368,3 +368,4 @@ deprecated
> 252 auto quick prealloc
> 253 auto quick
> 254 auto quick
> +255 auto quick
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
