From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (qmail 32027 invoked by alias); 7 Dec 2007 21:48:06 -0000 Received: (qmail 32000 invoked by uid 22791); 7 Dec 2007 21:48:05 -0000 X-Spam-Check-By: sourceware.org Received: from pfepa.post.tele.dk (HELO pfepa.post.tele.dk) (195.41.46.235) by sourceware.org (qpsmtpd/0.31) with ESMTP; Fri, 07 Dec 2007 21:47:53 +0000 Received: from x1-6-00-0f-9f-c6-3e-90 (unknown [80.197.1.215]) by pfepa.post.tele.dk (Postfix) with ESMTP id 8073FFAC029; Fri, 7 Dec 2007 22:47:50 +0100 (CET) Received: from x1-6-00-0f-9f-c6-3e-90 (localhost.localdomain [127.0.0.1]) by x1-6-00-0f-9f-c6-3e-90 (8.14.0/8.14.0) with ESMTP id lB7Llns0002943; Fri, 7 Dec 2007 22:47:49 +0100 Received: (from rask@localhost) by x1-6-00-0f-9f-c6-3e-90 (8.14.0/8.14.0/Submit) id lB7Lllg2002942; Fri, 7 Dec 2007 22:47:47 +0100 Date: Fri, 07 Dec 2007 21:48:00 -0000 From: Rask Ingemann Lambertsen To: Mark Mitchell Cc: GCC Patches , rsandifo@nildram.co.uk, fortran@gcc.gnu.org, java-patches@gcc.gnu.org Subject: Re: Link tests after GCC_NO_EXECUTABLES Message-ID: <20071207214746.GB17368@sygehus.dk> References: <20071201120251.GT17368@sygehus.dk> <20071201223447.GU17368@sygehus.dk> <47531F54.6010802@codesourcery.com> <20071205172224.GM17368@sygehus.dk> <47574456.1070108@codesourcery.com> <20071206175819.GO17368@sygehus.dk> <4758A3BD.5050102@codesourcery.com> <20071207153101.GR17368@sygehus.dk> <47599307.30409@codesourcery.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <47599307.30409@codesourcery.com> User-Agent: Mutt/1.5.14 (2007-02-12) Mailing-List: contact gcc-patches-help@gcc.gnu.org; run by ezmlm Precedence: bulk List-Id: List-Archive: List-Post: List-Help: Sender: gcc-patches-owner@gcc.gnu.org X-SW-Source: 2007-12/txt/msg00364.txt.bz2 On Fri, Dec 07, 2007 at 10:37:59AM -0800, Mark Mitchell wrote: > > I have to admit that I'm getting concerned. The issue Richard has > raised about Cygwin, and the various corner-cases here are making me > think that we're at risk of introducing instability. I think so too. If the Cygwin example had been raised a week ago I wouldn't have used ${with_newlib} to enable use of the cache file. Status is that I would still hate to loose the ability to build the simulator targets in a straight-forward way. At the same time, I can easily live with adding one more option to the configure command to get the simulator targets to build. So my suggestion is this: 1) Make the cached link test results require an explicit option to enable them. I can code that up tomorrow. 2) Put in the libstdc++ patch. 3) Revert the top level changes for the -B and -L options. Does this sound safe enough? -- Rask Ingemann Lambertsen Danish law requires addresses in e-mail to be logged and stored for a year