From: Robert L. <rm...@te...> - 2002-02-20 19:44:19
|
On Wed, 2002-02-20 at 12:57, Erich Focht wrote: > The patch is for 2.5.4-K3. I'm actually developing on IA-64 and tested it > on Itanium systems based on 2.4.17 kernels where it survived my > tests. I hope this works for i386 and is helpful to someone. I was working on the same thing myself. I don't have a working solution, so you beat me, and thus good job. I think we need this, for various reasons, especially to implement a method of setting task affinity that we can export to userspace. I am a little surprised by how much code it took, though. Do we need the function to act asynchronously? In other words, is it a requirement that the task reschedule immediately, or only that when it next reschedules it obeys its affinity? Also, what is the reason for allowing multiple calls to set_cpus_allowed? How often would that even occur? Robert Love |