linux-ext4 - Re: how to disable readahead

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>] [day] [month] [year] [list]

Date:   Thu, 2 Aug 2018 16:11:58 +0000
From:   "Bean Huo (beanhuo)" <beanhuo@...ron.com>
To:     Reindl Harald <h.reindl@...lounge.net>,
        Jack Wang <jack.wang.usish@...il.com>
CC:     "tytso@....edu" <tytso@....edu>,
        "linux-ext4@...r.kernel.org" <linux-ext4@...r.kernel.org>
Subject: Re: how to disable readahead

>
>Am 02.08.2018 um 17:57 schrieb Bean Huo (beanhuo):
>>>>> I guess you're doing buffered IO, have you try to use direct IO?
>>>>>
>>>> Direct I/O is another case, it is true that will bypass readahead.
>>>> But I want to disable readahead with page cache.
>>>
>>> don't get me wrong but when you state "I am newbie on ext4" this all
>>> sounds like "having solution, seeking for problem"
>>
>> No solution, I only want to disable ext4 readahead with keeping page cache.
>> I am curious about that flash based SDD/UFS/eMMC already has readahead
>in the FW, what is benefit of readahead in the FS under this condition?
>
>you still miss to point out the *problem* you are trying to solve and "i want"
>is no reasoning - as newbie you really should trust the developers that they
>know what they are doing and not play around with something for no good
>reason

Please focus on my question: how to disable readahead
What's problem I found. I am not sure it is related to readahead. I need to validate between conditions.
If you know how to disable readahead, please help.
Or we cannot disable that so far.