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 S1753949AbbCQOgW (ORCPT <rfc822;w@1wt.eu>);
	Tue, 17 Mar 2015 10:36:22 -0400
Received: from mail-we0-f174.google.com ([74.125.82.174]:34665 "EHLO
	mail-we0-f174.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752814AbbCQOgV (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Tue, 17 Mar 2015 10:36:21 -0400
Date: Tue, 17 Mar 2015 15:36:14 +0100
From: Ingo Molnar <mingo@kernel.org>
To: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
        Steven Rostedt <rostedt@goodmis.org>, Borislav Petkov <bp@alien8.de>,
        "H. Peter Anvin" <hpa@zytor.com>,
        Andy Lutomirski <luto@amacapital.net>, Oleg Nesterov <oleg@redhat.com>,
        Frederic Weisbecker <fweisbec@gmail.com>,
        Alexei Starovoitov <ast@plumgrid.com>, Will Drewry <wad@chromium.org>,
        Kees Cook <keescook@chromium.org>, x86@kernel.org,
        linux-kernel@vger.kernel.org
Subject: Re: [PATCH 1/2 v2] x86/asm/entry/64: Enable interrupts *after* we
 fetch PER_CPU_VAR(old_rsp)
Message-ID: <20150317143614.GA18462@gmail.com>
References: <1426600344-8254-1-git-send-email-dvlasenk@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1426600344-8254-1-git-send-email-dvlasenk@redhat.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org


* Denys Vlasenko <dvlasenk@redhat.com> wrote:

> Without this change, it is still not possible to get rid of
> PER_CPU_VAR(old_rsp) usage in switch_to: if preemption happens
> while we did not fetch PER_CPU_VAR(old_rsp) and stored it in pt_regs->sp,
> PER_CPU_VAR(old_rsp) gets corrupted by other task's user sp.

Well, wouldn't this be a much clearer explanation:

"We want to use PER_CPU_VAR(old_rsp) as a simple temporary register, 
 to shuffle user-space RSP into (and from) when we set up the system 
 call stack frame. At that point we cannot shuffle values into general 
 purpose registers, because we have not saved them yet.

 To be able to do this shuffling into a memory location, we must be 
 atomic and must not be preempted while we do the shuffling, otherwise 
 the 'temporary' register gets overwritten by some other task's 
 temporary register contents ..."

Agreed?

Thanks,

	Ingo