On Fri, Dec 01, 2000 at 12:41:40PM -0600, Geoffrey Wossum wrote:
> Hello all,
> Is anyone working/planning to create 2.2 or 2.4 series uClinux kernels? I
> don't think there's any compelling technology reason to go to newer
> kernels. However, it's difficult to explain that to someone outside the
> uClinux development community. When you tell someone that your little box
> is running 2.0.38, they suddenly don't think as much of it.
I'm all for it, but this is not exactly the best time. We could go with
2.2 and be obsolete RSN (should've been last year, IIRC), or we can go
to 2.4 as soon as that comes out and hope there aren't a lot of changes
to early 2.4 versions that affect us out of the gate. If the latter, we
could, of course, be working on it pending the release of 2.4. I have
been running 2.4 on my desktop box for awhile.
Being obsolete (referenced in above paragraph) hasn't hurt us materially
in the past.
I, personally, would like to go to 2.4 for the frame buffer support and
modular (programming style, not loadable) keyboard code. 2.4 is alleged
to be smaller in footprint than 2.2 for a change. 2.2 has a notably
larger footprint than 2.0, and that has been one of the arguments for
not going with it. Most of our targets don't have a lot of memory.
I'm told that the changes for being MMUless are trivial, but I,
personally, do not have the wherewithall to tackle an entire, separate
uClinux flavor. Every port would need work, too. Of course, that could
be handled by simply keeping the latest of 2.0 up and adding the ports
to the 2.4 version as people do the re-porting.
This message resent by the email@example.com list server http://www.uClinux.com/
This archive was generated by hypermail 2b30 : Sun Apr 07 2002 - 00:01:39 EST