linux-kernel - Re: [RFC PATCH 01/14] userfaultfd: set dirty and young on writeprotect

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]

Message-ID: <18EDDBB1-255A-475F-812A-8CA947493EC8@vmware.com>
Date:   Wed, 20 Jul 2022 18:09:26 +0000
From:   Nadav Amit <namit@...are.com>
To:     David Hildenbrand <david@...hat.com>
CC:     Linux MM <linux-mm@...ck.org>, LKML <linux-kernel@...r.kernel.org>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Mike Rapoport <rppt@...ux.ibm.com>,
        Axel Rasmussen <axelrasmussen@...gle.com>,
        Andrea Arcangeli <aarcange@...hat.com>,
        Andrew Cooper <andrew.cooper3@...rix.com>,
        Andy Lutomirski <luto@...nel.org>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Peter Xu <peterx@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Thomas Gleixner <tglx@...utronix.de>,
        Will Deacon <will@...nel.org>, Yu Zhao <yuzhao@...gle.com>,
        Nick Piggin <npiggin@...il.com>
Subject: Re: [RFC PATCH 01/14] userfaultfd: set dirty and young on
 writeprotect

On Jul 20, 2022, at 11:00 AM, David Hildenbrand <david@...hat.com> wrote:

> My patch review skills have seen better days. I thought you'd be
> removing the pte_write() check ... :( Tired eyes ...
> 
>> Having said that, I do notice now that pte_mkdirty() should not be done
>> only this condition is fulfilled. Instead we should just have
>> something like:
>> 
>> if (will_need) {
>> ptent = pte_mkyoung(ptent);
>> if (pte_write(ptent))
>> ptent = pte_mkdirty(ptent);
>> }
> 
> As can_change_pte_writable() will fail if it stumbles over a !pte_dirty
> page in current code ... so I assume you would have that code before the
> actual pte_mkwrite() logic, correct?

No, I thought this should go after for 2 reasons:

1. You want to allow the PTE to be made writable following the
can_change_pte_writable().

2. You do not want to set a non-writable PTE as dirty, especially since it
might then be used to determine that the PTE can become writable. Doing so
would circumvent cases in which set_page_dirty() needs to be called and
break things down.