David> Date: Tue, 9 Nov 1999 20:55:15 +0100 (CET) From: Andrea
David> Arcangeli <andrea@suse.de>
> Also the idle task will be less fast in rescheduling itself.
David> I don't think 200 or so cycles will even show up on the radar.
David> However, this mentioned cycle cost is arch dependant, and this
David> is why it should be an architecture level decision whether or
David> not to do something like this.
David> Oh yes, there are other options available btw. There are
David> machines where DMA controllers exist which can be put into a
David> zero'ing or loopback mode to do the zero'ing work. And thus
David> have no "cpu cost" except to write the address to the register
David> on the controller.
Hmmm good point, it is of course possible it is a win under certain
circumstances. On SMP boxes with some of the CPUs being idle in
particular.
Jes
-
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/