[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200902170140.GQ1362448@hirez.programming.kicks-ass.net>
Date: Wed, 2 Sep 2020 19:01:40 +0200
From: peterz@...radead.org
To: Borislav Petkov <bp@...en8.de>
Cc: Feng Tang <feng.tang@...el.com>,
Thomas Gleixner <tglx@...utronix.de>,
Ingo Molnar <mingo@...hat.com>,
"H . Peter Anvin" <hpa@...or.com>,
Dave Hansen <dave.hansen@...el.com>, 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 06:55:01PM +0200, Borislav Petkov wrote:
> On Wed, Sep 02, 2020 at 06:45:38PM +0200, peterz@...radead.org wrote:
> > We really should clear the CPUID bits when the kernel explicitly
> > disables things.
>
> Actually, you want to *disable* the functionality behind it by clearing
> a bit in CR4 - and yes, not all features have CR4 bits - so that
> luserspace doesn't "probe" the existence of certain instructions.
>
> Example: you can still try to run RDRAND and succeed even if the
> corresponding CPUID bit is clear.
Well yes, but as you say, we don't have that :/ Clearing it in CPUID is
the best we can do.
Powered by blists - more mailing lists