From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from list by monty-python.gnu.org with tmda-scanned (Exim 4.24) id 1AMDh6-0007el-Sy for qemu-devel@nongnu.org; Tue, 18 Nov 2003 16:47:32 -0500 Received: from mail by monty-python.gnu.org with spam-scanned (Exim 4.24) id 1AMDga-0007Uh-HN for qemu-devel@nongnu.org; Tue, 18 Nov 2003 16:47:31 -0500 Received: from [62.210.158.45] (helo=quito.magic.fr) by monty-python.gnu.org with esmtp (Exim 4.24) id 1AMDgZ-0007To-Ol for qemu-devel@nongnu.org; Tue, 18 Nov 2003 16:46:59 -0500 Received: from jma1.dev.netgem.com (gw.netgem.com [195.68.2.34]) by quito.magic.fr (8.11.6/8.11.2) with ESMTP id hAIKjRW26574 for ; Tue, 18 Nov 2003 21:45:27 +0100 (CET) Subject: Re: [Qemu-devel] [ADD] PPC processor emulation From: Jocelyn Mayer In-Reply-To: <339AF9FC-1A05-11D8-A9B2-000A95E9828E@raylucke.com> References: <20031117105133.7e856e56.Jens.Arm@gmx.de> <1069140512.14646.2174.camel@rapid> <1069151842.13659.2338.camel@rapid> <1069157594.14646.2350.camel@rapid> <339AF9FC-1A05-11D8-A9B2-000A95E9828E@raylucke.com> Content-Type: text/plain Message-Id: <1069188287.2755.231.camel@jma1.dev.netgem.com> Mime-Version: 1.0 Date: 18 Nov 2003 21:44:47 +0100 Content-Transfer-Encoding: 7bit Reply-To: qemu-devel@nongnu.org List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu mailing list On Tue, 2003-11-18 at 21:24, Raymond W.Lucke IV wrote: > Hmm... > > My goal is not necessarily to run Linux binaries directly on Darwin, > that would truly mean doing Linux syscall emulation. But from what I > understand, trying to make vl would actually be in a number of ways > much simpler, since it does not translate syscalls one by one, but > instead the operating system being executed handles it's own, and calls > virtual hardware provided by vl. > > Ray Well, you're right, in some way: there is no syscall emulation to do for vl. But you have to emulate a machine that looks like a real one. I think most of the softmmu code for ix86 could be re-used for PPC and that peripheral emulation could be separated into hardware emulation, independant from the target, and bus-glue, which defines how the device will be accessed for a given target. I think that trying to have some code to emulate a CHRP or a PREP machine would be a good start, but we _need_ PCI for those targets. But, if we get one of those hardware emulated, with improvements for the CPU emulation to handle supervisor instructions and exceptions, we would be able to try to boot AIX, Linux, maybe AUX & old MacOSes. It would be great to do this ! In my opinion, the hardest points are: PPC emulation improvement, cleaning the current vl code to separate x86 dedicated parts from generic ones and PCI. But I don't think we have to worry about PCI: x86 emulation will need it too :=) Regards. -- Jocelyn Mayer Never organized