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 S1754362AbbJNQ5l (ORCPT <rfc822;w@1wt.eu>);
	Wed, 14 Oct 2015 12:57:41 -0400
Received: from mail-lf0-f45.google.com ([209.85.215.45]:35079 "EHLO
	mail-lf0-f45.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753479AbbJNQ5j (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Wed, 14 Oct 2015 12:57:39 -0400
Date: Wed, 14 Oct 2015 19:57:36 +0300
From: Cyrill Gorcunov <gorcunov@gmail.com>
To: Andy Lutomirski <luto@amacapital.net>
Cc: Andy Lutomirski <luto@kernel.org>, Stas Sergeev <stsp@list.ru>,
        X86 ML <x86@kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        Brian Gerst <brgerst@gmail.com>, Denys Vlasenko <dvlasenk@redhat.com>,
        Linus Torvalds <torvalds@linux-foundation.org>,
        Borislav Petkov <bp@alien8.de>, Pavel Emelyanov <xemul@parallels.com>
Subject: Re: [RFC 3/4] x86/signal/64: Re-add support for SS in the 64-bit
 signal context
Message-ID: <20151014165736.GT2048@uranus>
References: <cover.1444697927.git.luto@kernel.org>
 <6e14e2f7ce1c38fc1996bd4d5d3d9dc70b8bd94d.1444697927.git.luto@kernel.org>
 <20151014164021.GS2048@uranus>
 <CALCETrXo7+QvkRcsDDMM+XCZ=hyysDbK1zixVN2Wa1T-0AZwqg@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CALCETrXo7+QvkRcsDDMM+XCZ=hyysDbK1zixVN2Wa1T-0AZwqg@mail.gmail.com>
User-Agent: Mutt/1.5.24 (2015-08-30)
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

On Wed, Oct 14, 2015 at 09:42:58AM -0700, Andy Lutomirski wrote:
> 
> That's the intent.
> 
> If you write __pad = 0, don't set UC_STRICT_RESTORE_SS, and leave cs
> set to a 64-bit value, then the kernel will detect that 0 is not a
> valid SS and will fix it for you.
> 
> If you do write UC_STRICT_RESTORE_SS (e.g. if you saved on a new
> kernel and you restore the saved uc_flags), then you'll get a new
> signal delivered.

Yeah, I just wanna be sure I understood the patch correctly. Thanks!

> If you're restoring a 32-bit or 16-bit context, then none of the above
> applies, but I doubt that CRIU supports that anyway.

True. I didn't implemented compat mode for criu yet, it's in todo.