linux-kernel - Re: [f2fs-dev] [PATCH] f2fs: introduce sysfs/data_io_flag to attach REQ

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Thu, 9 Apr 2020 10:31:48 +0800
From:   Chao Yu <yuchao0@...wei.com>
To:     Jaegeuk Kim <jaegeuk@...nel.org>
CC:     <linux-kernel@...r.kernel.org>,
        <linux-f2fs-devel@...ts.sourceforge.net>, <kernel-team@...roid.com>
Subject: Re: [f2fs-dev] [PATCH] f2fs: introduce sysfs/data_io_flag to attach
 REQ_META/FUA

On 2020/4/9 10:28, Jaegeuk Kim wrote:
> On 04/09, Chao Yu wrote:
>> On 2020/4/9 10:20, Jaegeuk Kim wrote:
>>> On 04/07, Chao Yu wrote:
>>>> On 2020/4/7 10:59, Jaegeuk Kim wrote:
>>>>> On 04/07, Chao Yu wrote:
>>>>>> On 2020/4/4 0:12, Jaegeuk Kim wrote:
>>>>>>> This patch introduces a way to attach REQ_META/FUA explicitly
>>>>>>> to all the data writes given temperature.
>>>>>>>
>>>>>>> -> attach REQ_FUA to Hot Data writes
>>>>>>>
>>>>>>> -> attach REQ_FUA to Hot|Warm Data writes
>>>>>>>
>>>>>>> -> attach REQ_FUA to Hot|Warm|Cold Data writes
>>>>>>>
>>>>>>> -> attach REQ_FUA to Hot|Warm|Cold Data writes as well as
>>>>>>>           REQ_META to Hot Data writes
>>>>>>
>>>>>> Out of curiosity, what scenario it is used for?
>>>>>
>>>>> It's testing purpose to compare the bandwidths per different IO flags.
>>>>
>>>> Thanks for the hint. :)
>>>>
>>>> As nobarrier was set in Android, so REQ_PREFLUSH will not be considered in
>>>> this sysfs interface?
>>>
>>> I don't see any diff on performance, so not interesting. :)
>>
>> I doubt it may has diff on non-ufs/emmc device? just guess.
> 
> I don't have any data on emmc, so maybe. Which case do we send REQ_PREFLUSH?

fsync w/ barrier mount option.

> 
>>
>> Thanks,
>>
>>>
>>>>
>>>> Thanks,
>>>>
>>>>>
>>>>>>
>>>>>> Thanks,
>>>>> .
>>>>>
>>> .
>>>
> .
>