From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <linux-kernel-owner@vger.kernel.org>
Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand
	id S933309AbaDVRET (ORCPT <rfc822;w@1wt.eu>);
	Tue, 22 Apr 2014 13:04:19 -0400
Received: from mail-vc0-f169.google.com ([209.85.220.169]:45982 "EHLO
	mail-vc0-f169.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S932737AbaDVRER (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Tue, 22 Apr 2014 13:04:17 -0400
MIME-Version: 1.0
In-Reply-To: <CAObL_7HdWs2hoNYd0gKzh6iVJr293Z9p+Dg1C6u+5GYQiDfgnA@mail.gmail.com>
References: <CAObL_7EJi5+m-oDXRy4hu+-OTZ=9wZ9WEivTMsdDtccU00wfWA@mail.gmail.com>
	<5355A9E9.9070102@zytor.com>
	<CAObL_7EP+zPpx9TVgsSu2iFN+r0U8yy6UEZtdk=CPwowXUu=Qw@mail.gmail.com>
	<1dbe8155-58da-45c2-9dc0-d9f4b5a6e643@email.android.com>
	<CAObL_7FUDpV9md+UnDbXxWw=trrXLFLNNJMNegdezrQt7rm6TA@mail.gmail.com>
	<a035392c-f332-4b3f-b851-13b0c7a0fc68@email.android.com>
	<CAObL_7FMX9yaGVi19pVwsU5VwHqKLLWMEB7kwDF-fatsGnHvdQ@mail.gmail.com>
	<ee12ff5e-91fe-487b-bed9-4472f15f94fe@email.android.com>
	<CAObL_7HTDvN2zu2_CDnVR_ztZ-b7PfLYz0csuVX-ShQ7EHGEjg@mail.gmail.com>
	<20140422112312.GB15882@pd.tnic>
	<20140422144659.GF15882@pd.tnic>
	<CAObL_7FGs4n6zusbdwTLi5W5q2V81Sf7pOnOmHPFyv5d7jMfvA@mail.gmail.com>
	<53569467.1030809@zytor.com>
	<CAObL_7F9yxt=vXjbssYB5wjZ7HUyKcstG7KYaRWxDDK0n7_vQw@mail.gmail.com>
	<CA+55aFyg1n6=Lnp_qhqdGESoP3u-sv_+MbvSdT4MEutGQAJESg@mail.gmail.com>
	<CAObL_7HdWs2hoNYd0gKzh6iVJr293Z9p+Dg1C6u+5GYQiDfgnA@mail.gmail.com>
Date: Tue, 22 Apr 2014 10:04:16 -0700
X-Google-Sender-Auth: hKjtKjfTCtKu_-mWn5ijPEoMblk
Message-ID: <CA+55aFzRf2Dhh3Eea1E74cpD9DXijUHpsXa71AURy_n6F_JKbw@mail.gmail.com>
Subject: Re: [PATCH] x86-64: espfix for 64-bit mode *PROTOTYPE*
From: Linus Torvalds <torvalds@linux-foundation.org>
To: Andrew Lutomirski <amluto@gmail.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>, Borislav Petkov <bp@alien8.de>,
        "H. Peter Anvin" <hpa@linux.intel.com>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
        Ingo Molnar <mingo@kernel.org>,
        Alexander van Heukelum <heukelum@fastmail.fm>,
        Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
        Boris Ostrovsky <boris.ostrovsky@oracle.com>,
        Arjan van de Ven <arjan.van.de.ven@intel.com>,
        Brian Gerst <brgerst@gmail.com>,
        Alexandre Julliard <julliard@winehq.com>,
        Andi Kleen <andi@firstfloor.org>, Thomas Gleixner <tglx@linutronix.de>
Content-Type: text/plain; charset=UTF-8
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Tue, Apr 22, 2014 at 10:00 AM, Andrew Lutomirski <amluto@gmail.com> wrote:
>
> My point is that it may be safe to remove the special espfix fixup
> from #PF, which is probably the most performance-critical piece here,
> aside from iret itself.

Actually, even that is unsafe.

Why?

The segment table is shared for a process. So you can have one thread
doing a load_ldt() that invalidates a segment, while another thread is
busy taking a page fault. The segment was valid at page fault time and
is saved on the kernel stack, but by the time the page fault returns,
it is no longer valid and the iretq will fault.

Anyway, if done correctly, this whole espfix should be totally free
for normal processes, since it should only trigger if SS is a LDT
entry (bit #2 set in the segment descriptor). So the normal fast-path
should just have a simple test for that.

And if you have a SS that is a descriptor in the LDT, nobody cares
about performance any more.

             Linus