From owner-cvs-src Thu Feb 13 18: 4:39 2003 Delivered-To: cvs-src@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 0EA2D37B401; Thu, 13 Feb 2003 18:04:37 -0800 (PST) Received: from haystack.lclark.edu (lclark.edu [149.175.1.2]) by mx1.FreeBSD.org (Postfix) with SMTP id E79A543F3F; Thu, 13 Feb 2003 18:04:27 -0800 (PST) (envelope-from eta@lclark.edu) Received: from [149.175.30.191] ([149.175.30.191]) by haystack.lclark.edu (SAVSMTP 3.0.0.44) with SMTP id M2003021318040029589 ; Thu, 13 Feb 2003 18:04:00 -0800 Subject: Re: cvs commit: src/sys/modules Makefile From: Eric Anholt To: Peter Wemm Cc: Dag-Erling Smorgrav , Marcel Moolenaar , src-committers@FreeBSD.org, cvs-src@FreeBSD.org, cvs-all@FreeBSD.org In-Reply-To: <20030213223058.769DA2A8C1@canning.wemm.org> References: <20030213223058.769DA2A8C1@canning.wemm.org> Content-Type: text/plain Organization: Message-Id: <1045185451.11981.17.camel@leguin> Mime-Version: 1.0 X-Mailer: Ximian Evolution 1.2.2 Date: 13 Feb 2003 18:06:15 -0800 Content-Transfer-Encoding: 7bit Sender: owner-cvs-src@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Thu, 2003-02-13 at 14:30, Peter Wemm wrote: > Eric Anholt wrote: > > On Thu, 2003-02-13 at 14:17, Peter Wemm wrote: > > > The lines in agpreg.h were what Eric added after the fact. This is an > > > i386-only chipset. It shouldn't have been in the sparc64, ia64, and alpha > > > builds. It certainly shouldn't be in powerpc whenever we start doing > > > builds for that. > > > > > > And IMHO, we certainly shouldn't be building/installing/releasing something > > > that we *know* hard locks the machine (Drew's report on agp_amd.c locking > > > up the UP1000). > > > > He said that he tried that when agp_amd first came out. agp_amd was > > broken on that chipset for all architectures for some time. > > > > I am working on a patch to only build the appropriate stuff for each > > arch. > > That would be much better. Please feel free to replace my hack. > But please don't enable it on platforms that it doesn't work on. > > For example, we have: > void > agp_flush_cache() > { > #ifdef __i386__ > wbinvd(); > #endif > } > > There's no alpha or ia64 support, so don't enable it there unless this > cache flushing isn't needed. Here's a patch to remove the arch-specific drivers from conf/files and only enable them for the appropriate architectures. As far as flush_cache on alpha, everyone else (NetBSD, linux) just does a memory barrier there, and linux has a note that it's probably wrong but they're not sure. http://people.freebsd.org/~anholt/files/agp-alphadiff Does this look appropriate? -- Eric Anholt eta@lclark.edu http://people.freebsd.org/~anholt/ anholt@FreeBSD.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-src" in the body of the message