Agreed. You won't hear a complaint from me here... I was only pointing
out that the behavior was expected and can be accounted for, so that we
don't hear criticisms based on the bonnie figures.
> I'll look into the bonnie #s; but were I to guess it's lmbench
> doing a _lot_ of syscalls - the patch i posted didn't penalize
> the idle process for doing that (even though both the original
> (unpublished) patch, and my current one does - i removed this,
> and later added it back).
Okay, yeah that makes sense, about the abundant syscalls. Some of the
lmbench tests syscall a LOT, approaching about 50% system time for the
CPU. I don't see that penalizing the idle task matters, though, because
when there's idle CPU time it will be run anyway, regardless of penalties.
Really, the right thing to do here seems to be ignoring the problem. The
patch does what it's designed to do; it doesn't magically fix some
unfixable problems, but I don't think anyone expected it to do so. :)
> Thanks for the numbers (these were for UP, right?).
Yeah, sorry. UP machine, UP kernel. Otherwise two tasks wouldn't have
made much sense. <g>
Chris Smith <cd_smith@ou.edu>
-
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/