Strange scheduling behavoir in SMP (kernel 2.2.14)

[Mark Hahn]

NOTE: I suggested to the author that he run his benchmark on the 
same hardware under UP and SMP.  result: no significant difference.
the conclusion from the test is that his compiler produces code
that runs faster on a K6-2 than a PIII; the scheduler is irrelevant.


On Fri, 28 Jan 2000, Peter Waltenberg wrote:
..
> No, it's still broken. Andrea has some patches for 2.2.
> 2.3 is still worse SMP than UP for anything CPU intensive.

where are your numbers?

> I'm amazed that people are whinging about a phantom 5% in the scheduler when
> there's a real 100%+ to be gained by fixing this one.

are you talking about ongoing silly-fest about long runqueues?
please don't confuse that with reality.

> It's a real bug. The current scheduler kicks running processes even when
> there's an idle CPU available. That means that heavy CPU use processes get
> ping-ponged across CPU's. 

where are your numbers?

regards, mark hahn.


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo en vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/