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 S1752187AbbCYX5M (ORCPT <rfc822;w@1wt.eu>);
	Wed, 25 Mar 2015 19:57:12 -0400
Received: from terminus.zytor.com ([198.137.202.10]:45703 "EHLO mail.zytor.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1751413AbbCYX5K (ORCPT <rfc822;linux-kernel@vger.kernel.org>);
	Wed, 25 Mar 2015 19:57:10 -0400
User-Agent: K-9 Mail for Android
In-Reply-To: <CA+55aFwHuaj80NaRNZW3Jfzh7_=nsx5p5=07PXUpaYwLzH-T=Q@mail.gmail.com>
References: <1427303896-24023-1-git-send-email-dvlasenk@redhat.com> <1427303896-24023-3-git-send-email-dvlasenk@redhat.com> <CA+55aFwHuaj80NaRNZW3Jfzh7_=nsx5p5=07PXUpaYwLzH-T=Q@mail.gmail.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit
Content-Type: text/plain;
 charset=UTF-8
Subject: Re: [PATCH 3/4] x86/asm/entry/64: use smaller insns
From: "H. Peter Anvin" <hpa@zytor.com>
Date: Wed, 25 Mar 2015 16:56:00 -0700
To: Linus Torvalds <torvalds@linux-foundation.org>,
        Denys Vlasenko <dvlasenk@redhat.com>
CC: Ingo Molnar <mingo@kernel.org>, Steven Rostedt <rostedt@goodmis.org>,
        Borislav Petkov <bp@alien8.de>, 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>,
        the arch/x86 maintainers <x86@kernel.org>,
        Linux Kernel Mailing List <linux-kernel@vger.kernel.org>
Message-ID: <BD648FB0-CBC0-4DB5-893A-4D4986585413@zytor.com>
Sender: linux-kernel-owner@vger.kernel.org
List-ID: <linux-kernel.vger.kernel.org>
X-Mailing-List: linux-kernel@vger.kernel.org

No, movabs is yet another instruction (with a 64-bit absolute address.) But movq can mean 10 or 7 bytes...

On March 25, 2015 4:51:50 PM PDT, Linus Torvalds <torvalds@linux-foundation.org> wrote:
>On Wed, Mar 25, 2015 at 10:18 AM, Denys Vlasenko <dvlasenk@redhat.com>
>wrote:
>> The $AUDIT_ARCH_X86_64 parameter to syscall_trace_enter_phase1/2
>> is a 32-bit constant, loading it with 32-bit MOV produces 5-byte insn
>> instead of 10-byte one.
>
>Side note: has anybody talked to the assembler people? This would seem
>to be very much something that the assembler could have noticed and
>done on its own. It's a bit sad that we need to overspecify these
>things..
>
>If it had actually been a 64-bit constant, the assembler would have
>ended up silently using a different instruction encoding *anyway*
>("movabs"), so it's not like the "movq" in any way specifies one
>particular instruction representation, and the assembler already picks
>different instruction versions for different constant values. Why not
>this one?
>
>                       Linus

-- 
Sent from my mobile phone.  Please pardon brevity and lack of formatting.