From owner-freebsd-emulation@FreeBSD.ORG Fri Jun 3 19:52:30 2005 Return-Path: X-Original-To: freebsd-emulation@freebsd.org Delivered-To: freebsd-emulation@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id C7D6516A41C for ; Fri, 3 Jun 2005 19:52:30 +0000 (GMT) (envelope-from sean@noc-phone.gigave.com) Received: from mailhost.gigave.com (mailhost.gigave.com [38.113.228.14]) by mx1.FreeBSD.org (Postfix) with ESMTP id AB2E343D1F for ; Fri, 3 Jun 2005 19:52:28 +0000 (GMT) (envelope-from sean@noc-phone.gigave.com) Date: Fri, 3 Jun 2005 12:52:39 -0700 From: Sean Chittenden To: freebsd-emulation@freebsd.org Message-ID: <20050603195239.GD974@sean.gigave.com> References: <20050603190658.GR8248@sean.gigave.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20050603190658.GR8248@sean.gigave.com> Subject: Re: [patch] Coping with setfs[gu]id16() stupidness... X-BeenThere: freebsd-emulation@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Development of Emulators of other operating systems List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 03 Jun 2005 19:52:30 -0000 > Is there any necessary symbol table magic that I need to work in order > to get the following patch to work? It compiles, but link_elf() > returns an error with "symbol linux_syscall_warn_enable not found". > At which point kldload(1) returns ENOENT (but that's another bug/story > all together). > > The intent of the patch is to quiet warnings when programs make use of > stupid linux syscalls (setfsuid16() and setfsgid16() come to mind). > In the patch, warnings are enabled with the patch they can be disabled > by setting compat.linux.syscall_warn_enable=0. > > Thanks in advance. -sc > > > http://people.FreeBSD.org/~seanc/patches/#compat_linux_syscall_warning > > -- > Sean Chittenden > Charter member of the "FilePro sucks" fan club I'm glad it's Friday. Andrew Gallatin aptly pointed out an #ifdef DEBUG that was wrapping missing symbol. Originally this was in linux_util.h, but I moved it to linux_mib.c for correctness when including in the actual tree. In the process, I mentally missed the transition from a .h to a .c where in a .c file you don't need/want an #endif. *sigh* The above URL has an updated patch for review that now works as expected. Any objections to me committing this and MFC'ing it after a week or two? -sc -- Sean Chittenden