On Fri, 2005-10-21 at 22:19 +0200, ness wrote:
Jonathan S. Shapiro wrote:
What *I* am saying is: the kernel must not include *any* operation that
would *ever*, under *any* circumstances, exceed log(n) time (n = number
of in-memory objects).
OK, this is your opinion. But you say this is only necessary for
real-time capabilities. In my eyes, the system should decide whether it
wants those rt-caps or not.
A *very* minor clarification:
We need to be careful about the word "capabilities". In the discussion
above, Ness appears (to me) to be confused, but I cannot be sure. I
believe that he us using the phrase "real time capabilities" here to
mean "the ability to support real-time guarantees", but he later goes on
to write "rt-caps", which makes matters confusing.
I suggest that we reserve the word "capability" in our discussions to
refer to the thing that is invoked, and use other terms elsewhere to
avoid confusion.
shap