[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <3eca92d0-5a57-3fff-63e1-edd3217341d5@windriver.com>
Date: Wed, 4 Sep 2019 10:36:00 +0800
From: "Hongzhi, Song" <hongzhi.song@...driver.com>
To: <linux-btrfs@...r.kernel.org>, <linux-kernel@...r.kernel.org>
CC: <josef@...icpanda.com>, <dsterba@...e.com>
Subject: Re: Bug Report: Btrfs can't allocate space for delete when block size
arounds 512M
Anybody notice this?
--Hongzhi
On 7/17/19 4:34 PM, Hongzhi, Song wrote:
> Hi friends,
>
> *Description:*
>
>
> One LTP testcase, fs_fill.c, fails on btrfs with kernel error when
> unlink files on Btrfs device:
>
> "BTRFS warning (device loop0): could not allocate space for a
> delete; will truncate on mount".
>
>
> I found the loop block device formatted with btrfs roughly rangs
> from 460M to 560M will cause the error.
>
> 256M and 1G all pass.
>
>
> The fs_fill.c source code:
>
> [https://github.com/linux-test-project/ltp/blob/master/testcases/kernel/fs/fs_fill/fs_fill.c]
>
>
> The fs_fill.c calls unlink which triggers the error.
>
> [https://github.com/linux-test-project/ltp/blob/e3457e42c1b93f54bb81da746eba314fd34ad40e/testcases/kernel/fs/fs_fill/fs_fill.c#L55]
>
>
> [https://github.com/linux-test-project/ltp/blob/e3457e42c1b93f54bb81da746eba314fd34ad40e/lib/safe_macros.c#L358]
>
>
>
> *Error info:*
>
> The issue maybe not reproduced everytime but four fifths chance.
>
> fs_fill.c:53: INFO: Unlinking mntpoint/thread5/file0
> safe_macros.c:360: BROK: fs_fill.c:55:
> unlink(mntpoint/thread10/file0) failed: ENOSPC
> safe_macros.c:360: BROK: fs_fill.c:55:
> unlink(mntpoint/thread11/file0) failed: ENOSPC
> [62477.378848] BTRFS warning (device loop0): could not allocate
> space for a delete; will truncate on mount
> [62477.378905] BTRFS warning (device loop0): could not allocate
> space for a delete; will truncate on mount
>
>
>
> *Kernel:*
>
> After v5.2-rc1, qemux86-64
>
> # make -j40 ARCH=x86_64 CROSS_COMPILE=x86-64-gcc
> use qemu to bootup kernel
>
>
> *LTP:*
>
> master branch: I tested on 20190625
> Reproduce:
>
> // build Ltp
> # cd Ltp-source
> # ./build.sh
>
> // copy files to qemu
> # cp runltp testcases/kernel/fs/fs_fill/fs_fill to qemu
>
> // login to qemu:
> // adjust block device size to 512M
> # vi runltp
> in function: create_block()
> dd if=/dev/zero of=${TMP}/test.img bs=1024 count=262144
> --->
> dd if=/dev/zero of=${TMP}/test.img bs=1024 count=524288
>
> // execute testcase
> # runltp -f fs -s fs_fill
>
>
> *Analysis:*
>
> One new kernel commit contained in v5.2-rc1 introduces the issue.
>
> commit c8eaeac7b734347c3afba7008b7af62f37b9c140
> Author: Josef Bacik <josef@...icpanda.com>
> Date: Wed Apr 10 15:56:10 2019 -0400
>
> btrfs: reserve delalloc metadata differently
> ...
>
>
> Anyone's reply will be appreciated.
>
> --Hongzhi
>
>
>
>
>
>
Powered by blists - more mailing lists