From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1946408AbWJ0L0G (ORCPT ); Fri, 27 Oct 2006 07:26:06 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1946410AbWJ0L0F (ORCPT ); Fri, 27 Oct 2006 07:26:05 -0400 Received: from scrub.xs4all.nl ([194.109.195.176]:48523 "EHLO scrub.xs4all.nl") by vger.kernel.org with ESMTP id S1946408AbWJ0L0D (ORCPT ); Fri, 27 Oct 2006 07:26:03 -0400 Date: Fri, 27 Oct 2006 13:25:53 +0200 (CEST) From: Roman Zippel X-X-Sender: roman@scrub.home To: "Robert P. J. Day" cc: Linux Kernel Mailing List Subject: Re: so what's so special about sema_init() for alpha? In-Reply-To: Message-ID: References: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, On Tue, 24 Oct 2006, Robert P. J. Day wrote: > i'm still curious as to why the implementation for sema_init() for > the alpha can't be simplified as (allegedly) could all of the other > architecture sema_init() calls. Did you even look at the code it generates? It's not specific to alpha at all. Unless the structure is small enough, gcc will first generate a copy on the stack and then copy it to its final location. bye, Roman