OLPC kernel patches not upstream yet...

[Jim Gettys]

Jeremy Katz asked the reasonable question of the feasibility of Fedora
carrying patches in advance of kernel.org, for stuff destined to
eventual inclusion. I quickly checked with Andres on the current state
of what isn't upstream in kernel.org.  

There are two big items left (for this round changes; if we can really
get to things like fixing USB resume, there are likely more such in the
future...).

o The device tree patch; the hold up here is that there are multiple
implementations of this on Sparc, PPC, and now x86, and no current
agreement among the guilty on which should become the "standard"
version. Until/unless the kernel community can come to consensus, this
seems something not reasonable for a fedora patch.  Andres would like to
hear from davem in particular on this topic. And we *might* get more
insight week after next after the kernel summit.

o the touchpad driver for our ALPS touchpad.  We're finally likely to be
able to come up with something reasonable now that the EC code problem
has been found, that won't be full of so many one off hack work-arounds.
As this is a separate driver for unique hardware, this seems like
something that could go into Fedora without likely headaches, if someone
wants to keep it synchronized.  But I think the standard touchpad driver
does at least "work".

Note that this presumes we're still using at least a OLPC .config file
going forward as we don't have VESA support for booting in our firmware
(we use fbdev), so a generic x86 Fedora kernel won't boot.  We also
don't want/need to carry the over whelming ton of device drivers, mostly
for hardware that can never be plugged in, just for space reasons.  It
would certainly be nice someday if a generic x86 kernel could boot (and
a standard Fedora desktop spin); I'm not exactly sure what that would
entail at this point, but for now, this seems like a more speculative
question and beyond the initial time frame for G1G1.  I'm personally
also by far mos comfortable using a kernel that is as close as possible
to what we've been testing for our 8.2.0 release at this date, for
support reasons.

Jeremy also noted a bit of version skew between the fedora and olpc rpm
spec file that could/should be resolved; there were a couple of trivial
config file changes that enables something build from the OLPC kernel to
be used in a Fedora spin (the squashfs patch, which all distros carry
but isn't in kernel.org, and one other item that I forget).

Andres, Deepak, did I miss anything?
                            - Jim

-- 
Jim Gettys <jg at laptop.org>
One Laptop Per Child