on 2022/07/22 0:43, Zorro Lang wrote:
> On Wed, Jul 20, 2022 at 10:21:19AM -0700, Darrick J. Wong wrote:
>> On Wed, Jul 20, 2022 at 01:45:37PM +0800, Yang Xu wrote:
>>> When I run xfs/144 manually, the step as below:
>>> mkfs.xfs -f -d agcount=3200,size=6366g -d file,n
>>>
>>> I accidentally pressed the Enter key, it triggers Segmentation fault.
>>> Then I found Darrick has fixed this one weeks ago. So add this test.
>>>
>>> Signed-off-by: Yang Xu <xuyang2018.jy@xxxxxxxxxxx>
>>> ---
>>> tests/xfs/549 | 33 +++++++++++++++++++++++++++++++++
>>> tests/xfs/549.out | 2 ++
>>> 2 files changed, 35 insertions(+)
>>> create mode 100755 tests/xfs/549
>>> create mode 100644 tests/xfs/549.out
>>>
>>> diff --git a/tests/xfs/549 b/tests/xfs/549
>>> new file mode 100755
>>> index 00000000..31ead3c7
>>> --- /dev/null
>>> +++ b/tests/xfs/549
>>> @@ -0,0 +1,33 @@
>>> +#! /bin/bash
>>> +# SPDX-License-Identifier: GPL-2.0
>>> +# Copyright (c) 2022 FUJITSU LIMITED. All rights reserved.
>>> +#
>>> +# FS QA Test 549
>>> +#
>>> +# Regression test for xfsprogs commit
>>> +# 50dba8189b1f ("mkfs: terminate getsubopt arrays properly")
>>> +#
>>> +# This case test mkfs.xfs whether can terminate getsubopt arrays properly.
>>> +# If not, it will trigger segmentation fault.
>>> +#
>>> +
>>> +. ./common/preamble
>>> +_begin_fstest auto quick mkfs
>>> +
>>> +# real QA test starts here
>>> +_supported_fs xfs
>>> +_fixed_by_git_commit xfsprogs 50dba8189b1f \
>>> + "mkfs: terminate getsubopt arrays properly"
>>> +_require_test
>>> +
>>> +testfile=$TEST_DIR/a
>>> +rm -rf $testfile
>>> +
>>> +$MKFS_XFS_PROG -f -d agcount=4 -d file,name=$testfile,nrext64=0 \
>>> + >> $seqres.full 2>&1
>>
>> Heh, you don't even need $testfile --
>>
>> "$MKFS_XFS_PROG -d agcount=3 -d garbagegarbagegarbage=0"
>>
>> is enough to cause the crash. But this works just as well, so
>
> Thanks Darrick.
>
> If it's not needed, let's remove those meaningless lines, likes
>
> testfile=$TEST_DIR/a
> rm -rf $testfile
> $MKFS_XFS_PROG -f -d agcount=4 -d file,name=$testfile,nrext64=0
>
> That makes me feel this test need "EEXIST" error, to trigger this bug :)
The reason is that I use the above commad to trigger this bug in xfs/144.
But this confuse you, so I will use ""$MKFS_XFS_PROG -d agcount=3 -d
garbage=0" on v2.
Best Regards
Yang Xu
>
> Thanks,
> Zorro
>
>
>>
>> Reviewed-by: Darrick J. Wong <djwong@xxxxxxxxxx>
>>
>> --D
>>
>>> +
>>> +echo "Silence is golden"
>>> +
>>> +# success, all done
>>> +status=0
>>> +exit
>>> diff --git a/tests/xfs/549.out b/tests/xfs/549.out
>>> new file mode 100644
>>> index 00000000..4e3acd3f
>>> --- /dev/null
>>> +++ b/tests/xfs/549.out
>>> @@ -0,0 +1,2 @@
>>> +QA output created by 549
>>> +Silence is golden
>>> --
>>> 2.23.0
>>>
>>
>
