lkml.org 
[lkml]   [2018]   [Nov]   [26]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [patch V2 27/28] x86/speculation: Add seccomp Spectre v2 user space protection mode

* Thomas Gleixner <tglx@linutronix.de> wrote:

> On Sun, 25 Nov 2018, Linus Torvalds wrote:
>
> > [ You forgot to fix your quilt setup.. ]
>
> Duh. Should have pinned that package.
>
> > On Sun, 25 Nov 2018, Thomas Gleixner wrote:
> > >
> > > The mitigation guide documents how STIPB works:
> > >
> > > Setting bit 1 (STIBP) of the IA32_SPEC_CTRL MSR on a logical processor
> > > prevents the predicted targets of indirect branches on any logical
> > > processor of that core from being controlled by software that executes
> > > (or executed previously) on another logical processor of the same core.
> >
> > Can we please just fix this stupid lie?
>
> Well, it's not a lie. The above is correct, it just does not tell WHY this
> works.

Well, it's a "technically correct but misleading" phrase, which has much
more of the effects of an actual "lie" than that of a true description of
it.

I.e. in terms of what effects it's likely going to have on readers not
aware of the underlying mechanics it's much more correct to call it a
"lie" than to call it "truth" - which I think is at the core of Linus's
argument.

> > Yes, Intel calls it "STIBP" and tries to make it out to be about the
> > indirect branch predictor being per-SMT thread.
> >
> > But the reason it is unacceptable is apparently because in reality it
> > just disables indirect branch prediction entirely. So yes,
> > *technically* it's true that that limits indirect branch prediction
> > to just a single SMT core, but in reality it is just a "go really
> > slow" mode.
>
> Indeed. Just checked the documentation again, it's also not clear
> whether IBPB is required if STIPB is in use.

So I think we should clarify all this.

Thanks,

Ingo

\
 
 \ /
  Last update: 2018-11-26 14:31    [W:0.148 / U:1.632 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site