From owner-freebsd-mips@FreeBSD.ORG Thu May 8 17:14:10 2003 Return-Path: Delivered-To: freebsd-mips@freebsd.org Received: by hub.freebsd.org (Postfix, from userid 931) id 5C9C737B401; Thu, 8 May 2003 17:14:10 -0700 (PDT) Date: Thu, 8 May 2003 19:14:10 -0500 From: Juli Mallett To: "Steffen M. Boelaars" Message-ID: <20030508191410.A51169@FreeBSD.org> References: <200305081434.12830.freebsd.org@graphicscave.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i In-Reply-To: <200305081434.12830.freebsd.org@graphicscave.com>; from freebsd.org@graphicscave.com on Thu, May 08, 2003 at 02:34:12PM +0200 X-Title: Code Maven X-Towel: Yes X-Negacore: Yes X-Authentication-Warning: localhost: juli pwned teh intarweb cc: freebsd-mips@freebsd.org Subject: Re: What's going on here. Can I help? Test? X-BeenThere: freebsd-mips@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Porting FreeBSD to MIPS List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Fri, 09 May 2003 00:14:10 -0000 * "Steffen M. Boelaars" [ Date: 2003-05-08 ] [ w.r.t. What's going on here. Can I help? Test? ] > And apart from this I am also myself just very interested in getting another > OS than Irix to run on one of my many SGI systems. > I have an Indigo2 r10000 195 MHz here that is dieing to be used. robert@ is working on an R10K machine as well (an Octane?). I'm working on IP22s. > Sooo.... after all this blablabla: how's things with the project, where can I > do something/anything at all to make things work a bit more? Nothing really works now. A ton of kernel code is still needed, currently I'm trying to get exceptions working properly (the NetBSD code leaves a bit to be desired and doesn't work right @64-bits). Also I am seeing a lot of unexpected access (deterministic) to bogus bits of memory, possibly as a result of stack corruption, or a stack size issue, I'm not sure. Using random printfs seems to make these things better/worse non-deterministically, but a printf in a given location deterministically changes how things work. This may be evidence of a compiler bug, but I'm not sure. Certainly there are bugs in the compiler. With -O the compiler ICEs, I seem to recall. imp@ suggests maybe it's a cache issue, but I've tried with L1 cache disabled, with the kernel running in KSEG1, and so on, to no avail. Thanx, juli. -- juli mallett. email: jmallett@freebsd.org; efnet: juli;