Alan Watson wrote:
> [SNIP...] It seems to be
> that under certain circumstances, the amonut of data being
> passed and the amount of computation being performed
> conspire to keep both processes on the same processor.
If you think this is what's happening (may not be of course) then you
might want to try changing PROC_CHANGE_PENALTY (in asm/smp.h) to 0, just
for the hell of it. Might not make any difference, but if it does then
it's another nail in the coffin for that particular technique of
processor biasing.
Neil
-
Linux SMP list: FIRST see FAQ at http://www.irisa.fr/prive/mentre/smp-faq/
To Unsubscribe: send "unsubscribe linux-smp" to [EMAIL PROTECTED]
