From mboxrd@z Thu Jan  1 00:00:00 1970
From: sunny.s.zhang <sunny.s.zhang@oracle.com>
Date: Fri, 29 Mar 2019 11:24:27 +0800
Subject: [Ocfs2-devel] [PATCH] ocfs2: fix ocfs2 read inode data panic in
 ocfs2_iget
In-Reply-To: <f4a83860-8e81-3a2e-66e0-bebc0c8e3426@gmail.com>
References: <1553477217-2736-1-git-send-email-sunny.s.zhang@oracle.com>
	<01433584-44d5-5b27-a0c9-fff4bff74864@gmail.com>
	<ab0a6ca7-2f6a-b738-72ce-4ffb6787aad2@oracle.com>
	<5C9846C7020000F90005B029@prv1-mh.provo.novell.com>
	<4b7dee86-52fe-d383-296e-5ce1bb7ef178@oracle.com>
	<e224f7e3-ad4e-b348-08dd-cf64993879c2@oracle.com>
	<f4a83860-8e81-3a2e-66e0-bebc0c8e3426@gmail.com>
Message-ID: <1d449c4e-d4e0-88bb-4288-e53e925c8226@oracle.com>
List-Id: <ocfs2-devel.oss.oracle.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
To: ocfs2-devel@oss.oracle.com

Hi Joseph,


? 2019?03?29? 10:21, Joseph Qi ??:
>
> On 19/3/29 09:05, sunny.s.zhang wrote:
>> Hi Gang,? Jiang Qi,? Joseph,
>>
>> Is my description clear?
>> Are there any other suggestions?
>> In addition, it works fine on ext4.
> So this happens when using ocfs2 as NFSv3 backend, right?
> If so, please describe the use case in commit log as well as the
> sequence how it happens.
Thank you very much for the feedback.
OK, I will add it to the commit log.
>
> BTW, since the dinode has been deleted, the block may be used in other
> way. So I don't think we can use "di->i_dtime != 0" to identify this
> case.
Yes, Although the time window is very small, it is strictly possible to 
happen.
I will make a new patch and consider the above question.
Thanks,
Sunny
>
> Thanks,
> Joseph