Re: [PATCH] xfs/444: test log replay after XFS_IOC_SWAPEXT

[Brian Foster <bfoster@redhat.com>]

On Fri, Feb 23, 2018 at 12:33:41PM -0600, Eric Sandeen wrote:
> This is a mashup of xfs/042 and some of the log replay tests;
> it checks whether the log can be replayed if we crash immediately
> after an xfs_fsr / XFS_IOC_SWAPEXT.
> 
> Hint: it can't.  It fails because the temporary donor inode has
> been deleted and has invalid mode 0 when we try to replay its
> swapext operation.  Kernel patches to fix it will follow soon.
> 
> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
> ---
> 
> diff --git a/tests/xfs/444 b/tests/xfs/444
> new file mode 100755
> index 0000000..e88438a
> --- /dev/null
> +++ b/tests/xfs/444
> @@ -0,0 +1,144 @@
...
> +# Test performs several operations to produce a badly fragmented file, then
> +# create enough contiguous free space for xfs_fsr to defragment the fragmented
> +# file:
> +#
> +# - create fs with 3 minimum sized (16Mb) allocation groups
> +# - create 16x1MB contiguous files which will become large free space extents
> +#   when deleted
> +# - put a small "space" between each of the 16 contiuguous files to ensure we
> +#   have separated free space extents
> +# - fill the remaining free space with a "fill file"
> +# - mount/unmount/fill remaining free space with a pad file
> +# - punch alternate single block holes in the the "fill file" to create
> +#   fragmented free space.
> +# - use fill2 to generate a very large fragmented file
> +# - delete the 16 large contiguous files created initially
> +# - run xfs_fsr on the filesystem
> +# - check checksums for remaining files
> +

Without having dug into the core issue, I wonder whether this sequence
could be simplified a bit by using 'xfs_io -c swapext' followed by a
shutdown?

Brian

> +rm -f $seqres.full
> +_do_die_on_error=message_only
> +
> +echo -n "Make a 48 megabyte filesystem on SCRATCH_DEV and mount... "
> +_scratch_mkfs_xfs -dsize=48m,agcount=3 2>&1 >/dev/null || _fail "mkfs failed"
> +_scratch_mount || _fail "mount failed" 
> +
> +echo "done"
> +
> +echo -n "Reserve 16 1Mb unfragmented regions... "
> +for i in 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
> +do
> +	_do "$XFS_IO_PROG -f -c \"resvsp 0 1m\" $SCRATCH_MNT/hole$i"
> +	_do "$XFS_IO_PROG -f -c \"resvsp 0 4k\" $SCRATCH_MNT/space$i"
> +	_do "xfs_bmap -vp $SCRATCH_MNT/hole$i"
> +done
> +echo "done" 
> +
> +# set up filesystem
> +echo -n "Fill filesystem with fill file... "
> +for i in `seq 0 1 31`; do
> +	_do "$XFS_IO_PROG -f -c \"falloc ${i}m 1m\" $SCRATCH_MNT/fill"
> +done
> +_do "xfs_bmap -vp $SCRATCH_MNT/fill"
> +echo "done"
> +# flush the filesystem - make sure there is no space "lost" to pre-allocation
> +_do "_scratch_unmount"
> +_do "_scratch_mount"
> +echo -n "Use up any further available space... "
> +_do "$XFS_IO_PROG -f -c \"falloc 0 1m\" $SCRATCH_MNT/pad"
> +echo "done"
> +
> +# create fragmented file
> +#_do "Delete every second file" "_cull_files"
> +echo -n "Punch every second 4k block... "
> +for i in `seq 0 8 32768`; do
> +	# This generates excessive output that significantly slows down the
> +	# test. It's not necessary for debug, so just bin it.
> +	$XFS_IO_PROG -f -c "unresvsp ${i}k 4k" $SCRATCH_MNT/fill \
> +								> /dev/null 2>&1
> +done
> +_do "xfs_bmap -vp $SCRATCH_MNT/fill"
> +_do "sum $SCRATCH_MNT/fill >$tmp.fillsum1"
> +echo "done"
> +
> +echo -n "Create one very large file... "
> +_do "src/fill2 -d nbytes=16000000,file=$SCRATCH_MNT/fragmented"
> +echo "done"
> +_do "xfs_bmap -v $SCRATCH_MNT/fragmented"
> +_do "sum $SCRATCH_MNT/fragmented >$tmp.sum1"
> +_do "Remove other files" "rm -rf $SCRATCH_MNT/{pad,hole*}"
> +
> +# defragment
> +_do "Run xfs_fsr on filesystem" "$XFS_FSR_PROG -v $SCRATCH_MNT/fragmented"
> +_do "xfs_bmap -v $SCRATCH_MNT/fragmented"
> +
> +echo "godown"
> +src/godown -v -f $SCRATCH_MNT >> $seqres.full
> +
> +echo "unmount"
> +_scratch_unmount
> +
> +echo "mount with replay"
> +_scratch_mount $mnt >>$seqres.full 2>&1 \
> +    || _fail "mount failed: $mnt $MOUNT_OPTIONS"
> +
> +# success, all done
> +echo "xfs_fsr tests passed."
> +status=0 ; exit
> diff --git a/tests/xfs/444.out b/tests/xfs/444.out
> new file mode 100644
> index 0000000..a0e7cd5
> --- /dev/null
> +++ b/tests/xfs/444.out
> @@ -0,0 +1,13 @@
> +QA output created by 444
> +Make a 48 megabyte filesystem on SCRATCH_DEV and mount... done
> +Reserve 16 1Mb unfragmented regions... done
> +Fill filesystem with fill file... done
> +Use up any further available space... done
> +Punch every second 4k block... done
> +Create one very large file... done
> +Remove other files... done
> +Run xfs_fsr on filesystem... done
> +godown
> +unmount
> +mount with replay
> +xfs_fsr tests passed.
> diff --git a/tests/xfs/group b/tests/xfs/group
> index e2397fe..85033b5 100644
> --- a/tests/xfs/group
> +++ b/tests/xfs/group
> @@ -441,3 +441,4 @@
>  441 auto quick clone quota
>  442 auto stress clone quota
>  443 auto quick ioctl fsr
> +444 auto quick fsr log
> 
> --
> To unsubscribe from this list: send the line "unsubscribe fstests" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html