linux-kernel - Re: [PATCH] mm/userfaultfd: fix memory corruption due to 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: <CALCETrUUkGj00Z0HRuYOpjP8uGgbbs539EwG8tc71+PJR_=z_Q@mail.gmail.com>
Date:   Mon, 21 Dec 2020 10:22:38 -0800
From:   Andy Lutomirski <luto@...nel.org>
To:     Andrea Arcangeli <aarcange@...hat.com>
Cc:     Andy Lutomirski <luto@...nel.org>,
        Nadav Amit <nadav.amit@...il.com>,
        Dave Hansen <dave.hansen@...el.com>,
        linux-mm <linux-mm@...ck.org>, Peter Xu <peterx@...hat.com>,
        lkml <linux-kernel@...r.kernel.org>,
        Pavel Emelyanov <xemul@...nvz.org>,
        Mike Kravetz <mike.kravetz@...cle.com>,
        Mike Rapoport <rppt@...ux.vnet.ibm.com>,
        stable <stable@...r.kernel.org>,
        Minchan Kim <minchan@...nel.org>, Yu Zhao <yuzhao@...gle.com>,
        Will Deacon <will@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>
Subject: Re: [PATCH] mm/userfaultfd: fix memory corruption due to writeprotect

On Mon, Dec 21, 2020 at 10:04 AM Andrea Arcangeli <aarcange@...hat.com> wrote:
>
> Hello,
>
> On Sat, Dec 19, 2020 at 09:08:55PM -0800, Andy Lutomirski wrote:
> > On Sat, Dec 19, 2020 at 6:49 PM Andrea Arcangeli <aarcange@...hat.com> wrote:
> > > The ptes are changed always with the PT lock, in fact there's no
> > > problem with the PTE updates. The only difference with mprotect
> > > runtime is that the mmap_lock is taken for reading. And the effect
> > > contested for this change doesn't affect the PTE, but supposedly the
> > > tlb flushing deferral.
> >
> > Can you point me at where the lock ends up being taken in this path?
>
> pte_offset_map_lock in change_pte_range, as in mprotect, no difference.
>
> As I suspected on my follow up, the bug described wasn't there, but
> I'll look at the new theory posted.

Indeed.