linux-kernel - Re: [PATCH] mm: add access/dirty bit on numa page fault

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 for Android: free password hash cracker in your pocket

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

Message-ID: <4bb3e4ef-29f9-448b-22ad-1ae7c2d008ff@loongson.cn>
Date:   Thu, 17 Mar 2022 09:09:16 +0800
From:   maobibo <maobibo@...ngson.cn>
To:     Anshuman Khandual <anshuman.khandual@....com>,
        Andrew Morton <akpm@...ux-foundation.org>
Cc:     linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH] mm: add access/dirty bit on numa page fault



On 03/16/2022 05:42 PM, Anshuman Khandual wrote:
> 
> 
> On 3/16/22 12:33, maobibo wrote:
>>
>> On 03/16/2022 02:43 PM, Anshuman Khandual wrote:
>>>
>>> On 3/16/22 06:38, Bibo Mao wrote:
>>>> During numa page fault, dirty bit can be added for old pte if
>>>> fail to migrate on write fault. And if it succeeds to migrate,
>>>> access bit can be added for migrated new pte, also dirty bit
>>>> can be added for write fault.
>>> The current code does not set the access and dirty bits when ever
>>> applicable i.e on FAULT_FLAG_WRITE, on the pte (old if migration
>>> fails, new if migration succeeds) ? Did not this cause any problem
>>> earlier ? I am wondering how this might have gone unnoticed.
> 
>> On arm/x86 platform hw will set access/dirty bits automatically,
>> however on MIPS platform access/dirty bits are set by software in next
>> page fault, it is relatively easier to watch on MIPS platform.
> 
> Could you please update this in the commit message as well ?

sure, will do. 

regards
bibo, mao