From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753532Ab3EFOaK (ORCPT ); Mon, 6 May 2013 10:30:10 -0400 Received: from mail-ve0-f179.google.com ([209.85.128.179]:42045 "EHLO mail-ve0-f179.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751950Ab3EFOaI (ORCPT ); Mon, 6 May 2013 10:30:08 -0400 MIME-Version: 1.0 In-Reply-To: References: Date: Mon, 6 May 2013 07:30:08 -0700 X-Google-Sender-Auth: de1Yl6kV7XMYbso9r37msofVJDk Message-ID: Subject: Re: h8300 SYMBOL_NAME() breakage From: Linus Torvalds To: Geert Uytterhoeven Cc: Rusty Russell , Al Viro , James Hogan , "linux-kernel@vger.kernel.org" Content-Type: text/plain; charset=UTF-8 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Mon, May 6, 2013 at 5:31 AM, Geert Uytterhoeven wrote: > > How should this be fixed? > - Re-add SYMBOL_NAME() to arch/h8300/include/asm/linkage.h? No. The whole point was to *not* have these random private definitions. I guess you could use VMLINUX_SYMBOL() and include and then use VMLINUX_SYMBOL(). Although.. > - Kill SYMBOL_NAME() and SYMBOL_NAME_LABEL(), and hardcode the > underscores in the h8300 asm sources? .. I think this is what everybody else does. Although sometimes that underscore depends on what the linkage format is (ie historically a.out vs elf etc). We pretty much expect ELF these days, but maybe other conditional reasons exist - so there may be a reason you don't want to do the hardcoding (although on h8300 it seems to be unconditional at least for now). Linus