As far as the vm changes discussions go (MAP_INHERIT, processor local
data, etc), two things are obvious:
* It would be a lot of work, else someone would just post a patch
the next day and say "try it, report results".
* It is not obvious that implementing these ideas would be a big
win, else there would be an enthusiastic chorus of "great idea" instead
of these multi-faceted expressions of skepticism about the wisdom
of such changes to things that mostly or entirely work in their
present implementations.
* These aren't bug fixes, these are basic design changes, else
performance issues would be secondary to getting semantically
correct behavior (the question would be not whether to change the
implementation but whether the proposed change really fixes the bug).
In sum, one is not going to persuade a group of developers unconvinced
of the wisdom of a large change to the kernel to actually do the work,
so the only recourse is to implement it as an off-line patch instead
of the core of the mainline kernel. Many such projects exist: clusters,
security-hardened linux, virtual (fault-tolerant) linux, rt linux,
mklinux, etc.
Put up a web page, ftp directory, mailing list, have at it. Or "it was
just an idea that I don't have time to implement myself", and on to
other things.
Regards,
Clayton Weaver
<mailto:cgweav@eskimo.com>
(Seattle)
"Everybody's ignorant, just in different subjects." Will Rogers
-
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/