Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Subject: Re: [f2fs-dev] [f2fs bug] infinite loop in
 f2fs_get_meta_page_nofail()
To:     <jaegeuk@kernel.org>
CC:     Eric Biggers <ebiggers@kernel.org>,
        <syzbot+ee250ac8137be41d7b13@syzkaller.appspotmail.com>,
        <syzkaller-bugs@googlegroups.com>, <linux-kernel@vger.kernel.org>,
        <linux-f2fs-devel@lists.sourceforge.net>
References: <000000000000432c5405b1113296@google.com>
 <20201007213253.GD1530638@gmail.com> <20201007215305.GA714500@google.com>
 <c7baef0d-d459-114f-7146-627f0c4159ad@huawei.com>
 <20201009015015.GA1931838@google.com>
 <8fa4f9fe-5ca5-f3a3-c8f4-e800373c1e46@huawei.com>
 <20201009043237.GB1973455@google.com>
 <a842eec8-2dbb-aa01-cc64-c513d59cad24@huawei.com>
 <20201009145626.GA2186792@google.com>
From:   Chao Yu <yuchao0@huawei.com>
Message-ID: <70faa161-bcd7-64a3-4a6c-04963c0784b6@huawei.com>
Date:   Tue, 13 Oct 2020 10:30:36 +0800
User-Agent: Mozilla/5.0 (Windows NT 10.0; WOW64; rv:52.0) Gecko/20100101
 Thunderbird/52.9.1
MIME-Version: 1.0
In-Reply-To: <20201009145626.GA2186792@google.com>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Language: en-US
Content-Transfer-Encoding: 7bit
Precedence: bulk

Jaegeuk,

I guess you missed sending last applied patch to mailing list?

Thanks,

On 2020/10/9 22:56, jaegeuk@kernel.org wrote:
> On 10/09, Chao Yu wrote:
>> On 2020/10/9 12:32, jaegeuk@kernel.org wrote:
>>> On 10/09, Chao Yu wrote:
>>>> On 2020/10/9 9:50, jaegeuk@kernel.org wrote:
>>>>> On 10/09, Chao Yu wrote:
>>>>>> On 2020/10/8 5:53, jaegeuk@kernel.org wrote:
>>>>>>> On 10/07, Eric Biggers wrote:
>>>>>>>> [moved linux-fsdevel to Bcc]
>>>>>>>>
>>>>>>>> On Wed, Oct 07, 2020 at 02:18:19AM -0700, syzbot wrote:
>>>>>>>>> Hello,
>>>>>>>>>
>>>>>>>>> syzbot found the following issue on:
>>>>>>>>>
>>>>>>>>> HEAD commit:    a804ab08 Add linux-next specific files for 20201006
>>>>>>>>> git tree:       linux-next
>>>>>>>>> console output: https://syzkaller.appspot.com/x/log.txt?x=17fe30bf900000
>>>>>>>>> kernel config:  https://syzkaller.appspot.com/x/.config?x=26c1b4cc4a62ccb
>>>>>>>>> dashboard link: https://syzkaller.appspot.com/bug?extid=ee250ac8137be41d7b13
>>>>>>>>> compiler:       gcc (GCC) 10.1.0-syz 20200507
>>>>>>>>> syz repro:      https://syzkaller.appspot.com/x/repro.syz?x=1336413b900000
>>>>>>>>> C reproducer:   https://syzkaller.appspot.com/x/repro.c?x=12f7392b900000
>>>>>>>>>
>>>>>>>>> The issue was bisected to:
>>>>>>>>>
>>>>>>>>> commit eede846af512572b1f30b34f9889d7df64c017d4
>>>>>>>>> Author: Jaegeuk Kim <jaegeuk@kernel.org>
>>>>>>>>> Date:   Fri Oct 2 21:17:35 2020 +0000
>>>>>>>>>
>>>>>>>>>         f2fs: f2fs_get_meta_page_nofail should not be failed
>>>>>>>>>
>>>>>>>>
>>>>>>>> Jaegeuk, it looks like the loop you added in the above commit doesn't terminate
>>>>>>>> if the requested page is beyond the end of the device.
>>>>>>>
>>>>>>> Yes, that will go infinite loop. Otherwise, it will trigger a panic during
>>>>>>> the device reboot. Let me think how to avoid that before trying to get the
>>>>>>> wrong lba access.
>>>>>>
>>>>>> Delivering f2fs_get_sum_page()'s return value needs a lot of codes change, I think
>>>>>> we can just zeroing sum_page in error case, as we have already shutdown f2fs via
>>>>>> calling f2fs_stop_checkpoint(), then f2fs_cp_error() will stop all updates to
>>>>>> filesystem data including summary pages.
>>>>>
>>>>> That sounds like one solution tho, I'm afraid of getting another panic by
>>>>> wrong zero'ed summary page.
>>>>
>>>> What case do you mean? maybe I missed some corner cases?
>>>
>>> I sent v2 to fix syzbot issue, which fixes wrong use of
>>> f2fs_get_meta_page_nofail.
>>
>> I agreed to fix that case, however we may encounter deadloop in other
>> places where we call f2fs_get_meta_page_nofail()? like the case that
>> filesystem will always see EIO after we shutdown device via dmflakey?
> 
> We may need another option to deal with this. At least, however, it's literally
> _nofail function which should guarantee no error, instead of hiding the error
> with zero'ed page.
> 
>>
>> Thanks,
>>
>>>
>>>>
>>>> Thanks,
>>>>
>>>>>
>>>>>>
>>>>>> Thoughts?
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>>>
>>>>>>>>
>>>>>>>> - Eric
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> Linux-f2fs-devel mailing list
>>>>>>> Linux-f2fs-devel@lists.sourceforge.net
>>>>>>> https://lists.sourceforge.net/lists/listinfo/linux-f2fs-devel
>>>>>>> .
>>>>>>>
>>>>> .
>>>>>
>>> .
>>>
> .
>