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 S1752527AbbCZAF4 (ORCPT <rfc822;w@1wt.eu>);
	Wed, 25 Mar 2015 20:05:56 -0400
Received: from mail-ie0-f177.google.com ([209.85.223.177]:36389 "EHLO
	mail-ie0-f177.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1752498AbbCZAFv (ORCPT
	<rfc822;linux-kernel@vger.kernel.org>);
	Wed, 25 Mar 2015 20:05:51 -0400
MIME-Version: 1.0
In-Reply-To: <BD648FB0-CBC0-4DB5-893A-4D4986585413@zytor.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>
	<BD648FB0-CBC0-4DB5-893A-4D4986585413@zytor.com>
Date: Wed, 25 Mar 2015 17:05:50 -0700
X-Google-Sender-Auth: uAxpqLTwFe8W0DmPOf0tu8czBHM
Message-ID: <CA+55aFyJg1248Df6tj6KMA2o2KCQbZJ7i6POCPureVwzX91Dzg@mail.gmail.com>
Subject: Re: [PATCH 3/4] x86/asm/entry/64: use smaller insns
From: Linus Torvalds <torvalds@linux-foundation.org>
To: "H. Peter Anvin" <hpa@zytor.com>
Cc: Denys Vlasenko <dvlasenk@redhat.com>, 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>
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 Wed, Mar 25, 2015 at 4:56 PM, H. Peter Anvin <hpa@zytor.com> wrote:
> No, movabs is yet another instruction (with a 64-bit absolute address.) But movq can mean 10 or 7 bytes...

I mentioned movabs because that is literally what as generates at
least for me (or then objdump is confused):

  [torvalds@i7 ~]$ as -v
  GNU assembler version 2.24 (x86_64-redhat-linux) using BFD version
version 2.24

  [torvalds@i7 ~]$ cat t.s
  main:
        movq $0x12, %rdi
        movq $0x1234, %rdi
        movq $0x123456, %rdi
        movq $0x12345678, %rdi
        movq $0x123456789ab, %rdi

  [torvalds@i7 ~]$ as t.s
  [torvalds@i7 ~]$ objdump -d a.out
  ...
     0: 48 c7 c7 12 00 00 00 mov    $0x12,%rdi
     7: 48 c7 c7 34 12 00 00 mov    $0x1234,%rdi
     e: 48 c7 c7 56 34 12 00 mov    $0x123456,%rdi
    15: 48 c7 c7 78 56 34 12 mov    $0x12345678,%rdi
    1c: 48 bf ab 89 67 45 23 movabs $0x123456789ab,%rdi
    23: 01 00 00

so 'as' is clearly just stupid. It already takes the size of the
constant into account and generates different instructions. Why not
for the common 32-bit case too?

Oh well.

                       Linus