Date: Tue, 15 Dec 2009 08:26:07 -0800 From: Marcel Moolenaar <xcllnt@mac.com> To: Anton Shterenlikht <mexas@bristol.ac.uk> Cc: freebsd-ia64@freebsd.org Subject: Re: ia64 kernel config Message-ID: <3DDCCE88-C254-4154-B8AC-69CF4A8A8C60@mac.com> In-Reply-To: <20091215150254.GA21739@mech-cluster241.men.bris.ac.uk> References: <20091215011146.GA14252@mech-cluster241.men.bris.ac.uk> <7AFD99EA-5641-4B4B-B906-C4B16EDF196F@mac.com> <20091215150254.GA21739@mech-cluster241.men.bris.ac.uk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Dec 15, 2009, at 7:02 AM, Anton Shterenlikht wrote: >> BTW: I have mine here: >> http://ns1.xcllnt.net/~marcel/machines/ia64/hob/HOB > > Marcel, why do you build GPT support as a module, > and not straight in the kernel? I thought you need > GPT support for ia64 in any case. Am I wrong? I'm a developer. I wrote GEOM_PART. Building them all by default and having them as modules helps me test the various partitioning schemes. If you look closely, you'll see "device scc" in my config as well. There's no need for it to be there, because I don't have the hardware in my ia64 box. But, I wrote the driver and this way I keep it included when I change something in uart (which I wrote too). > In particular, if one uses GEOM_MIRROR for root > partition, I imagine GPT module must be loaded > prior to mirror creation. Am I right? Yes. Remember that modules can be loaded from the loader, prior to booting the kernel. In that sense there's virtually no difference from having them compiled-in. FYI, -- Marcel Moolenaar xcllnt@mac.com
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?3DDCCE88-C254-4154-B8AC-69CF4A8A8C60>