linux-kernel - Re: [PATCHv3 OPTIONAL 8/8] x86/mm: Extend LAM to support to LAM

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: <Yqr/WdgRb1IEDMVk@hirez.programming.kicks-ass.net>
Date:   Thu, 16 Jun 2022 12:00:57 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>
Cc:     Dave Hansen <dave.hansen@...ux.intel.com>,
        Andy Lutomirski <luto@...nel.org>, x86@...nel.org,
        Kostya Serebryany <kcc@...gle.com>,
        Andrey Ryabinin <ryabinin.a.a@...il.com>,
        Andrey Konovalov <andreyknvl@...il.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: [PATCHv3 OPTIONAL 8/8] x86/mm: Extend LAM to support to LAM_U48

On Fri, Jun 10, 2022 at 05:35:27PM +0300, Kirill A. Shutemov wrote:
> LAM_U48 allows to encode 15 bits of tags into address.
> 
> 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 other one.

This patch is broken in that it doesn't fix untag_pointer()

*If* you really want to continue down this road; you'll need something
like:

#define untagged_addr(mm, addr)        ({                              \
       u64 __addr = (__force u64)(addr);                               \
       s64 sign = (s64)__addr >> 63;                                   \
       __addr ^= sign;                                                 \
       __addr &= (mm)->context.untag_mask[sign & 1];                   \
       __addr ^= sign;                                                 \
       (__force __typeof__(addr))__addr;                               \
})

Which uses a different mask for kernel and user pointers.

Anyway, without this U48 patch on, the mask could be a constant, no need
to keep this variable, we can unconditionally unmask U57.

Let me go reply to that other mail too.