Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Sun, 15 Jul 2001 21:30:29 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Sun, 15 Jul 2001 21:30:19 -0400 Received: from c008-h000.c008.snv.cp.net ([209.228.34.63]:4344 "HELO c008.snv.cp.net") by vger.kernel.org with SMTP id ; Sun, 15 Jul 2001 21:30:07 -0400 X-Sent: 16 Jul 2001 01:30:00 GMT Message-ID: <3B52438B.3CC6E1BC@acm.org> Date: Mon, 16 Jul 2001 11:29:47 +1000 From: Gareth Hughes X-Mailer: Mozilla 4.77 [en] (X11; U; Linux 2.4.6 i686) X-Accept-Language: en MIME-Version: 1.0 To: linux-kernel@vger.kernel.org Subject: 4.1.0 DRM (was Re: Linux 2.4.6-ac3) In-Reply-To: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Alan Cox wrote: > > Right but we cannot go around breaking support for older setups. A user > updating their 2.4.x stable kernel and finding X11 no longer works simply isnt > an acceptable situation for serious users. Agreed 100%. > Why was so much of it macroised rather than turned into library routines ? Customization is the big one. Some drivers need AGP, some need MTRRs, some need DMA, some need interrupts etc. With the templates, this is done with minimal code duplication, and allows the final routines to be space-efficient -- only the code that each driver actually needs is present. This is advantageous when building as modules, or building a single driver into the kernel. You could argue that it's less important when building many drivers into the kernel, but how many people do that? Plus, there were symbol clashing problems with building some drivers into the kernel and some as modules. Keith Owens could comment on that. The new code avoids this problem as well. > > ... New drivers are much > > easier to write as well, which is a nice side-effect. > > That I can believe Pity that no new drivers have been written to prove this... -- Gareth - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/