On Thu, May 11, 2006 at 10:50:57PM +0100, Phil Blundell wrote: > On Thu, 2006-05-11 at 13:42 -0700, Erik Hovland wrote: > > So I thought maybe I would just augment initscripts to have a script > > which did an ipkg remove <package> if the architecture is not right by > > reading /proc/cpuinfo. That doesn't seem like the right place unless I > > play with bootmisc.sh. Anyhow, that is the jist. > > > > I want to do this, but I am not sure about the implementation that is > > 'right' or straightforward for distributions to do this. > > The most obvious solution is to patch the asic3 mmc driver so that it > won't initialize (i.e. returns an error code from init_module() on > non-asic3 machines).
It could be even easier then that. It seems that in the 2.4 repository the h3900 is the only supported device with an asic3 companion chip. I could just make the driver only work with h3900. And return -ENODEV otherwise. I can just check this into 2.4 cvs. But I can also use an OE patch. I think there is very little downside to making the driver h3900 only. But I wanted to give people a chance to pipe up and say whether that is something they think the 2.4 driver still needs. So pipe up if you develop the hh.org 2.4 kernel on a handheld that is not an h39xx that uses the asic3 companion chip. E -- Erik Hovland mail: erik AT hovland DOT org web: http://hovland.org/ PGP/GPG public key available on request _______________________________________________ Oe mailing list [email protected] https://www.handhelds.org/mailman/listinfo/oe
