From: Eryu Guan <guan@eryu.me>
To: Sun Ke <sunke32@huawei.com>
Cc: fstests@vger.kernel.org, yuchao0@huawei.com
Subject: Re: [PATCH] generic/103: leave more space for f2fs
Date: Sun, 13 Jun 2021 22:44:57 +0800 [thread overview]
Message-ID: <YMYZ6dW7w7p4liLC@desktop> (raw)
In-Reply-To: <20210611112211.1408767-2-sunke32@huawei.com>
On Fri, Jun 11, 2021 at 07:22:10PM +0800, Sun Ke wrote:
> It failed on f2fs:
> QA output created by 103
> +fallocate: No space left on device
> Silence is golden.
> ...
>
> Try to leave ~512KB, but during the fallocate, f2fs will be filled.
> Provide error prompt: "fallocate: No space left on device".
>
> Leave more space for f2fs.
And this also could be a bug in f2fs, that couldn't fallocate enough
space when free space is larger than request size.
I think the whole patchset needs review from f2fs folks.
Thanks,
Eryu
>
> Signed-off-by: Sun Ke <sunke32@huawei.com>
> ---
> I add some test code
> + df
> avail=`_get_available_space $SCRATCH_MNT`
> + echo "avail = ${avail}"
> filesizekb=$((avail / 1024 - 512))
> + echo "filesizekb = ${filesizekb}"
> $XFS_IO_PROG -fc "falloc 0 ${filesizekb}k" $file
> + df
> }
>
> f2fs:
> QA output created by 103
> +Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> +/dev/sdb 12580864 553292 12027572 5% /mnt/scratch
> +avail = 12316233728
> +filesizekb = 12027060
> +fallocate: No space left on device
> +Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> +/dev/sdb 12580864 12580864 0 100% /mnt/scratch
> Silence is golden.
>
> xfs:
> QA output created by 103
> +Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> +/dev/sdb 12572672 120828 12451844 1% /mnt/scratch
> +avail = 12750688256
> +filesizekb = 12451332
> +Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> +/dev/sdb 12572672 12572160 512 100% /mnt/scratch
> Silence is golden.
>
> f2fs:
> QA output created by 103
> +Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> +/dev/sdb 12319880 41048 11633304 1% /mnt/scratch
> +avail = 11912503296
> +filesizekb = 11632792
> +Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> +/dev/sdb 12319880 11673844 508 100% /mnt/scratch
> Silence is golden.
>
> after change (f2fs)
> QA output created by 103
> Filesystem 1K-blocks Used Available Use% Mounted on
> ...
> /dev/sdb 12580864 12580396 468 100% /tmp/scratch
> Silence is golden.
>
> tests/generic/103 | 10 +++++++++-
> 1 file changed, 9 insertions(+), 1 deletion(-)
>
> diff --git a/tests/generic/103 b/tests/generic/103
> index b22b0159623e..dfce3870781c 100755
> --- a/tests/generic/103
> +++ b/tests/generic/103
> @@ -40,10 +40,18 @@ rm -f $seqres.full
> _consume_freesp()
> {
> file=$1
> + leave=512
> +
> + # Try to leave ~512KB, but during the fallocate, f2fs will be filled.
> + # Provide error prompt: "fallocate: No space left on device". Leave
> + # more space for f2fs.
> + if [ $FSTYP == "f2fs" ]; then
> + leave=12288
> + fi
>
> # consume nearly all available space (leave ~512kB)
> avail=`_get_available_space $SCRATCH_MNT`
> - filesizekb=$((avail / 1024 - 512))
> + filesizekb=$((avail / 1024 - $leave))
> $XFS_IO_PROG -fc "falloc 0 ${filesizekb}k" $file
> }
>
> --
> 2.25.4
next prev parent reply other threads:[~2021-06-13 14:45 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-11 11:22 [PATCH] common/rc: f2fs do not support metadata journaling Sun Ke
2021-06-11 11:22 ` [PATCH] generic/103: leave more space for f2fs Sun Ke
2021-06-13 14:44 ` Eryu Guan [this message]
2021-06-15 3:36 ` Sun Ke
2021-06-15 3:28 ` Sun Ke
2021-06-16 13:15 ` [f2fs-dev] " Chao Yu
2021-06-17 6:22 ` Sun Ke
2021-06-11 11:22 ` [PATCH] generic/260: f2fs is also special Sun Ke
2021-06-15 3:29 ` Sun Ke
2021-06-16 13:43 ` [f2fs-dev] " Chao Yu
2021-06-17 6:38 ` Sun Ke
2021-06-15 3:25 ` [PATCH] common/rc: f2fs do not support metadata journaling Sun Ke
2021-06-16 12:36 ` [f2fs-dev] " Chao Yu
2021-06-17 3:45 ` Sun Ke
2021-06-17 23:46 ` Chao Yu
2021-06-18 2:03 ` Sun Ke
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=YMYZ6dW7w7p4liLC@desktop \
--to=guan@eryu.me \
--cc=fstests@vger.kernel.org \
--cc=sunke32@huawei.com \
--cc=yuchao0@huawei.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