linux-kernel - Re: [RFCv2 04/10] x86/mm: Introduce X86_THREAD_LAM_U48 and X86_THREAD_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: <Yn0bjINOR9d/2/27@hirez.programming.kicks-ass.net>
Date:   Thu, 12 May 2022 16:37:00 +0200
From:   Peter Zijlstra <peterz@...radead.org>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     "Kirill A. Shutemov" <kirill.shutemov@...ux.intel.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Andy Lutomirski <luto@...nel.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 04/10] x86/mm: Introduce X86_THREAD_LAM_U48 and
 X86_THREAD_LAM_U57

On Thu, May 12, 2022 at 02:24:22PM +0200, Thomas Gleixner wrote:
> On Wed, May 11 2022 at 09:02, Peter Zijlstra wrote:
> > On Wed, May 11, 2022 at 05:27:45AM +0300, Kirill A. Shutemov wrote:
> >
> >> +#define LAM_NONE	0
> >> +#define LAM_U57		1
> >> +#define LAM_U48		2
> >
> >> +#define X86_THREAD_LAM_U48		0x1
> >> +#define X86_THREAD_LAM_U57		0x2
> >
> > Seriously pick an order and stick with it. I would suggest keeping the
> > hardware order and then you can do:
> >
> >> +static inline unsigned long lam_to_cr3(u8 lam)
> >> +{
> >
> > 	return (lam & 0x3) << X86_CR3_LAM_U57;
> 
> This "works" because the hardware ignores LAM_U48 if LAM_U57 is set, but
> I'd rather make that exclusive in the prctl() as setting both does not
> make any sense.

Yes, I was very much assuming the interface would not allow setting
both, that would be daft.