From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from pigeon.gentoo.org ([69.77.167.62] helo=lists.gentoo.org) by finch.gentoo.org with esmtp (Exim 4.60) (envelope-from <gentoo-embedded+bounces-2686-garchives=archives.gentoo.org@lists.gentoo.org>) id 1Kbepf-0002RQ-7E for garchives@archives.gentoo.org; Fri, 05 Sep 2008 17:10:51 +0000 Received: from pigeon.gentoo.org (localhost [127.0.0.1]) by pigeon.gentoo.org (Postfix) with SMTP id 06130E0469; Fri, 5 Sep 2008 17:10:51 +0000 (UTC) Received: from homeless.linbsd.net (homeless.linbsd.net [64.127.112.66]) by pigeon.gentoo.org (Postfix) with ESMTP id CA58AE0469 for <gentoo-embedded@lists.gentoo.org>; Fri, 5 Sep 2008 17:10:50 +0000 (UTC) Received: from [192.168.1.2] (dsl092-011-131.sfo1.dsl.speakeasy.net [66.92.11.131]) by homeless.linbsd.net (Postfix) with ESMTPA id B4BF1585C8 for <gentoo-embedded@lists.gentoo.org>; Fri, 5 Sep 2008 10:10:49 -0700 (PDT) Subject: Re: [gentoo-embedded] openmoko overlay? From: Ned Ludd <solar@gentoo.org> To: gentoo-embedded@lists.gentoo.org In-Reply-To: <48C13357.8050004@gmail.com> References: <48BEC33C.30501@gmail.com> <20080903224428.54750e89.miknix@gmail.com> <20080903214610.GA5820@mbp> <48C13357.8050004@gmail.com> Content-Type: text/plain Date: Fri, 05 Sep 2008 10:10:49 -0700 Message-Id: <1220634649.21033.28.camel@media> Precedence: bulk List-Post: <mailto:gentoo-embedded@lists.gentoo.org> List-Help: <mailto:gentoo-embedded+help@lists.gentoo.org> List-Unsubscribe: <mailto:gentoo-embedded+unsubscribe@lists.gentoo.org> List-Subscribe: <mailto:gentoo-embedded+subscribe@lists.gentoo.org> List-Id: Gentoo Linux mail <gentoo-embedded.gentoo.org> X-BeenThere: gentoo-embedded@lists.gentoo.org Reply-to: gentoo-embedded@lists.gentoo.org Mime-Version: 1.0 X-Mailer: Evolution 2.12.3 Content-Transfer-Encoding: 7bit X-Archives-Salt: 0cd0f6c6-3dbe-448b-8f6b-01f966cc1187 X-Archives-Hash: 92473bc39acd197fe6eeac51bce823c1 On Fri, 2008-09-05 at 09:25 -0400, Christopher Friedt wrote: > Cool - I'll check it out. > > The moko is an arm920t core (armv4t arch), but I believe that the > official packages aren't even compiled with the EABI. The core itself is > exactly the same as that of the TS-72xx boards, which I've worked with > extensively. The packaging, pinouts, and on-chip peripherals are really > all that are different, so creating a toolchain is a non-issue. > > The /proc/cpuinfo component of the FreeRunner's kernel is slightly > bjorked - through some odd patches they have mashed board info into the > cpuinfo structure somehow instead of the cleaner /proc/SBC/info structure. > > I should have no problem taking my current instructions from the > gentoo-wiki and editing them to provide info for the openmoko. U-Boot, > and the usb-loader utility (dfu-util) are the only components that I > have little experience with. Although it was reasonably easy enough to > load the om2008.8 image. > > So I guess, in the long run, I'll be aiming to provide a kernel patchset > specifically for the freerunner hardware in the portage tree, as well as > a toolchain patchset for the samsung sc324XX SoC that the openmoko folks > are using. It should be possible to build everything that the openmoko > folks provide in any of their images, given a list of packages & the > patchsets, but using the (far better IMHO) ebuild build system as > opposed to bitbake. > > Nedd & Angelo - with the PDA-centric overlay ( & binary packages? ) that > you guys are creating, are there any plans in the works for making a > (graphical?) package manager app for mobile devices? I'm using > portage-utils pretty extensively these days for a push-ing distributed > package manager, which my boss has me unfortunately implementing as a > grails web-app, but I'd be very interested in collaborating on something > like that, gtk-based, ncurses-based, written in Java for all I care, etc. One of the problems that miknix has run into so far is that the openmoko-appmanager depends on ipkg and friends. There was some talk of recoding/hacking on that one to make it portage-utils aware. Feel free to discuss that further with him. In the freerunner they are using the sc32440 for this generation of the phone right? It should only be minor kernel patching required. But I'd rather us not have an explicit openmoko-sources as I'm using the sc324xx chips myself. I'm not sure why you would have to provide any toolchain patches. Everything works as of today the the misc CHOST's the device can use. armv4tl-softfloat-linux-gnueabi / arm-softfloat-linux-gnueabi / arm-softfloat-linux-uclibc etc....