linux-kernel - Re: [RFC PATCH] tools/x86: add kcpuid tool to show raw CPU features

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: <20200902165950.GP1362448@hirez.programming.kicks-ass.net>
Date:   Wed, 2 Sep 2020 18:59:50 +0200
From:   peterz@...radead.org
To:     Dave Hansen <dave.hansen@...el.com>
Cc:     Feng Tang <feng.tang@...el.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>,
        "H . Peter Anvin" <hpa@...or.com>, Borislav Petkov <bp@...en8.de>,
        x86@...nel.org, linux-kernel@...r.kernel.org
Subject: Re: [RFC PATCH] tools/x86: add kcpuid tool to show raw CPU features

On Wed, Sep 02, 2020 at 09:52:33AM -0700, Dave Hansen wrote:
> On 9/2/20 9:45 AM, peterz@...radead.org wrote:
> > On Thu, Aug 27, 2020 at 03:49:03PM +0800, Feng Tang wrote:
> >> End users frequently want to know what features their processor
> >> supports, independent of what the kernel supports.
> >>
> >> /proc/cpuinfo is great. It is omnipresent and since it is provided by
> >> the kernel it is always as up to date as the kernel. But, it can be
> >> ambiguous about processor features which can be disabled by the kernel
> >> at boot-time or compile-time.
> > 
> > Let me once again suggest we use CPUID faulting to 'fix' this.
> > 
> > We really should clear the CPUID bits when the kernel explicitly
> > disables things.
> > 
> > https://lkml.kernel.org/r/1552431636-31511-18-git-send-email-fenghua.yu@intel.com
> 
> Wouldn't we also have to turn on faulting for all the bits that are
> enumerated, but of which the kernel is unaware?
> 
> Right now, a bit that we clearcpuid= and one about which the kernel is
> totally unaware look the same.

I mostly care about the things the kernel disables (and thus knows
about), the rest is a pass-through IIRC.