I really appreciate work done by you, and many other for the Linux development.
As you have indicated too, that the code size is getting gigantic, but yet
it has to be maintained. I think I understand your dilemma, but I don't see
a solution without backing out months of your hard work.
In Linus situation, he is as uncomfortable with large patches from you as you
are with the large 2.3.* kernel patches without knowing what changes fixes
which problem. Linus is looking for manageable fixes in smaller patches, and
want to know what a set of changes (as numerous as it is) fixes which set of
interrelated problems.
On the other hand, you want to be able to fix problem for people like us
who want to stay with a stable kernel (say 2.2.6 for instance) but would
like some of the network driver updated. This may be done because 2.2.10+
is unstable for SMP or for various other reasons. But we are a small group
of users compare to many who would want to stay with the kernel tree updates
regularly.
You find the best way to maintain your driver code is to isolate it from the
kernel tree for small but numerous changes and have some testing on the
side. When you feel comfortable enough, you ask to have the patches applied
but were rejected. You also see that applying incremental patches to
the still under development tree difficult to debug.
I not sure how this can be resolved nicely Donald, because right now changes
in the drivers a so numerous that no one know what changes fixes exactly what
problem anymore. Please give it some thought, take a deep breath and think
about it :)
We all want to have a great operating system here, and want to get problem
fixed. You are always the author, our net card guy, but we want you to be the
maintainer as well because you know your code better than anyone. Please find
a way to work together with Linus.
- Bernard Wei -o)
bwei@home.com o) o) /\\
________________________________/\__/\_____>V/__
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/