[Techtalk] Theory vs. practice

[Jeff Dike]

val at nmt.edu said:
> I'm sorry - I realize that many of the best kernel developers don't
> even post on l-k.  

I think almost everyone reads it, for some definition of "read" :-)

> When you see people who are pushing for (and have
> written the patches for) a preemptive Linux kernel, and these selfsame
> people don't understand priority inversion or think it is a problem...

Robert Love hasn't been impressing me with the quality of his arguments,
but I think he understands what priority inversion is.  And it's possible
that's he's right about it not being a problem (although it's not clear that
he's thought about it very much, and people who have thought about it disagree
with him vehemently).

In any case, rml isn't in charge of the kernel.  He's done the Linux world
good by implementing preemption in the first place.  And even if that
is all that he's capable of doing and has no understanding of the issues
with getting preemption into the mainline kernel, that's OK (although it
could obviously be better).  There are people who do who can take what he's
done and make it workable, assuming that's possible.

				Jeff