From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261329AbVFOIoM (ORCPT ); Wed, 15 Jun 2005 04:44:12 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261331AbVFOIoL (ORCPT ); Wed, 15 Jun 2005 04:44:11 -0400 Received: from e2.ny.us.ibm.com ([32.97.182.142]:13770 "EHLO e2.ny.us.ibm.com") by vger.kernel.org with ESMTP id S261329AbVFOIoJ (ORCPT ); Wed, 15 Jun 2005 04:44:09 -0400 From: Arnd Bergmann To: Roy Lee Subject: Re: One question about fork, vfork and clone Date: Wed, 15 Jun 2005 10:41:42 +0200 User-Agent: KMail/1.7.2 Cc: linux-kernel@vger.kernel.org References: <42AFE410.4020803@gmail.com> In-Reply-To: <42AFE410.4020803@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset="big5" Content-Transfer-Encoding: 7bit Content-Disposition: inline Message-Id: <200506151041.44365.arnd@arndb.de> Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org On Middeweken 15 Juni 2005 10:17, Roy Lee wrote: > is this means the library calls no more wrap the sys_clone() for the > three library call(fork,vfork,clone), but > call the corresponding syscall? or it never did that? The sys_vfork and sys_fork entry points are there so old libc versions can still work. Some architectures that were added after sys_clone was introduced wrongly copied this code, but new architectures that get added should just provide sys_clone and have the other calls implemented as wrappers in libc. Arnd <><