Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Tue, 23 Jan 2018 20:45:52 +0000
From:   Alan Cox <gnomes@lxorguk.ukuu.org.uk>
To:     Andrew Cooper <andrew.cooper3@citrix.com>
Cc:     David Woodhouse <dwmw@amazon.co.uk>, <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH v2 5/5] x86/pti: Do not enable PTI on fixed Intel
 processors
Message-ID: <20180123204552.14040f98@alans-desktop>
In-Reply-To: <bb364dbb-4f16-56ab-c306-d96edea94dad@citrix.com>
References: <1516726375-25168-1-git-send-email-dwmw@amazon.co.uk>
        <1516726375-25168-6-git-send-email-dwmw@amazon.co.uk>
        <20180123173312.1d8cf02f@alans-desktop>
        <bb364dbb-4f16-56ab-c306-d96edea94dad@citrix.com>
Organization: Intel Corporation
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

> > static int in_order_cpu(void)
> > {
> > 	/* Processors with CPU id etc */
> > 	if (x86_match_cpu(cpu_in_order))
> > 		return 1;
> > 	/* Other rules here */
> > 	return 0;
> > }  
> 
> Why does in-order vs out-of-order matter?
> 
> There are leaky SP3 gadgets which satisfy in-order requirements, so long
> as the processor is capable of speculating 3 instructions past an
> unresolved branch.
> 
> What would (at a guess) save an in-order speculative processor from
> being vulnerable is if memory reads are issued and resolve in program
> order, but in that case, it is not the in-order property of the
> processor which makes it safe.

Fair point - I should rename it cpu_speculates(). The atoms in that
list don't speculate.

Alan
[My Cyrix 6x86 had a different kind of meltdown problem....]