From owner-freebsd-alpha Fri Aug 17 9:43: 7 2001 Delivered-To: freebsd-alpha@freebsd.org Received: from mail.rpi.edu (mail.rpi.edu [128.113.22.40]) by hub.freebsd.org (Postfix) with ESMTP id B070237B407; Fri, 17 Aug 2001 09:43:01 -0700 (PDT) (envelope-from drosih@rpi.edu) Received: from [128.113.24.47] (gilead.acs.rpi.edu [128.113.24.47]) by mail.rpi.edu (8.11.3/8.11.3) with ESMTP id f7HGgvk74952; Fri, 17 Aug 2001 12:42:58 -0400 Mime-Version: 1.0 X-Sender: drosih@mail.rpi.edu Message-Id: In-Reply-To: <20010817081147.DCEEA3811@overcee.netplex.com.au> References: <20010817081147.DCEEA3811@overcee.netplex.com.au> Date: Fri, 17 Aug 2001 12:42:54 -0400 To: Peter Wemm , Wilko Bulte From: Garance A Drosihn Subject: Re: dropping compiler support in -current Cc: "David O'Brien" , alpha@FreeBSD.org, developers@FreeBSD.org Content-Type: text/plain; charset="us-ascii" ; format="flowed" Sender: owner-freebsd-alpha@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.org At 1:11 AM -0700 8/17/01, Peter Wemm wrote: >Lets go over a couple of key things.. > >- There are going to be alpha boxes around for another year so so. The >hardware is viable for a while more yet. The problem is developer time. What is the future of alpha boxes? (the hardware itself, even ignoring freebsd's support for them). Isn't that all sold off to Intel now? I was toying with the idea of buying an alpha box this past summer, just for testing freebsd changes on it, but that didn't seem such a good idea once it seemed that the hardware was heading for a dead end. Besides (speaking solely on a personal preference), I'd be more interested in PPC (I also do Mac support at RPI), and Sparc-64 platforms. This is not meant as statement of what freebsd should work on, it just indicates the hardware that I have the most access to. >Don't forget, we're in the middle of major architectural turmoil and >the core infrastructure is bumpy as hell right now. I'm not sure that >its a bad thing that the pace of the alpha port is cooling off a bit, >otherwise there would be a lot of energy spent on moving targets. Would >it be such a bad thing if the real effort to stabilize alpha didn't >happen till 5.1 time frame when the core kernel services have stabilized? To some degree that makes sense, but I think it's a risky approach. If we don't have enough active Alpha developers now, why do we expect to see more of them six months from now? And if we wait until i386 support is "solid", then isn't there a risk that some architectural decisions will be made which will be less-than-optimal for alpha? Won't it be even harder to do the "right thing" for Alpha at that time, if all the work has already been settled on and done for i386? And if we go hands-off for Alpha, that will just mean that there are more loose-ends which won't even get tested on Alpha until six months from now. Various "new things" will depend on other changes to the system, but those other changes will not be happening on alpha so there will be no way to test the "new things" on Alpha. If we don't have enough Alpha users who really are keeping current with -current now, then it certainly is reasonable to sideline that port for while. But if it happens, I suspect it will be that much harder to resurrect the port six months from now. In my mind, it's pretty much up to the people using freebsd on alpha. -- Garance Alistair Drosehn = gad@eclipse.acs.rpi.edu Senior Systems Programmer or gad@freebsd.org Rensselaer Polytechnic Institute or drosih@rpi.edu To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-alpha" in the body of the message