linux-kernel - Re: [RFCv2 08/10] x86/mm: Make LAM_U48 and mappings above 47-bits mutually exclusive

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: <20220513232242.7ac7herw7iyfkd2h@black.fi.intel.com>
Date:   Sat, 14 May 2022 02:22:42 +0300
From:   "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Dave Hansen <dave.hansen@...ux.intel.com>,
        Andy Lutomirski <luto@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>, x86@...nel.org,
        Andrey Ryabinin <aryabinin@...tuozzo.com>,
        Alexander Potapenko <glider@...gle.com>,
        Dmitry Vyukov <dvyukov@...gle.com>,
        "H . J . Lu" <hjl.tools@...il.com>,
        Andi Kleen <ak@...ux.intel.com>,
        Rick Edgecombe <rick.p.edgecombe@...el.com>,
        linux-mm@...ck.org, linux-kernel@...r.kernel.org
Subject: Re: [RFCv2 08/10] x86/mm: Make LAM_U48 and mappings above 47-bits
 mutually exclusive

On Thu, May 12, 2022 at 03:36:31PM +0200, Thomas Gleixner wrote:
> On Wed, May 11 2022 at 05:27, Kirill A. Shutemov wrote:
> > LAM_U48 steals bits above 47-bit for tags and makes it impossible for
> > userspace to use full address space on 5-level paging machine.
> 
> > Make these features mutually exclusive: whichever gets enabled first
> > blocks the othe one.
> 
> So this patch prevents a mapping above 47bit when LAM48 is enabled, but
> I fail to spot how an already existing mapping above 47bit would prevent
> LAM48 from being enabled.
> 
> Maybe I'm missing something which makes this magically mutually
> exclusive.

It is in 09/10. See lam_u48_allowed()

-- 
 Kirill A. Shutemov