Re: [PATCH v1] generic/275: use free block counts to check if fs is filled sufficiently.

[Eryu Guan <guan@eryu.me>]

On Wed, Dec 02, 2020 at 06:27:31PM +0800, Xiaoli Feng wrote:
> From: Xiaoli Feng <xifeng@redhat.com>
> 
> For large block size, such as 64k, there're maybe leave more than 768k
> space when disk is full. I met one. When I test 64k for dax. It leaves
> about 800k space when disk is full. So change to jude the free block
> for >4k fs.

I'm not sure if that's a bug in the filesystem, and the test is designed
to find such bugs. Would you please make sure if that's a bug or not
first? Maybe by asking in proper mailing list.

Thanks,
Eryu

> ---
>  tests/generic/275 | 13 ++++++++++---
>  1 file changed, 10 insertions(+), 3 deletions(-)
> 
> diff --git a/tests/generic/275 b/tests/generic/275
> index 3a016037..2c5a807f 100755
> --- a/tests/generic/275
> +++ b/tests/generic/275
> @@ -62,9 +62,16 @@ rm -f $SCRATCH_MNT/tmp1
>  sync
>  echo "Post rm space:" >> $seqres.full
>  $DF_PROG $SCRATCH_MNT >>$seqres.full 2>&1
> -_freespace=`$DF_PROG -k $SCRATCH_MNT | tail -n 1 | awk '{print $5}'`
> -[ $_freespace -gt 1024 ] && _fail "could not sufficiently fill filesystem"
> -
> +bsize=$(_get_file_block_size $SCRATCH_MNT)
> +bsize=$((bsize / 1024))
> +_freespace=`$DF_PROG -k  $SCRATCH_MNT | tail -n 1 | awk '{print $5}'`
> +# When block size is >4k, use the free counts of block to judge
> +if [ "$bsize" -gt 4 ]; then
> +	_freeblock=$((_freespace / bsize))
> +	[ "$_freeblock" -gt 256 ] && _fail "could not sufficiently fill filesystem"
> +else
> +	[ "$_freespace" -gt 1024 ] && _fail "could not sufficiently fill filesystem"
> +fi
>  # Try to write more than available space in chunks that will allow at least one
>  # full write to succeed.
>  dd if=/dev/zero of=$SCRATCH_MNT/tmp1 bs=128k count=8 >>$seqres.full 2>&1
> -- 
> 2.18.1