On Thu, 11 Nov 1999 19:03:56 -0800, Dan Kegel <dank@alumni.caltech.edu>
said:
> How would it change the comparison if we had a vector version of
> sigwaitinfo that could return n siginfo's at once?
Not much. It would speed up by at most a constant factor: we'd have
fewer syscalls involved, but we still have to do a read() or write()
syscall for any fds returned, so we are still O(x) where x is the number
of live fds since the last check.
It is worthwhile doing this, definitely, but it won't actually change
the scalability: just the absolute performance.
--Stephen
-
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/