On Wed, 28 Aug 2002, Rusty Russell wrote:
> > - HT-aware affinity.
> >
> > Tasks should attempt to 'stick' to physical CPUs, not logical CPUs.
>
> Linus disagreed with this before when I discussed it with him, and with
> the current (stupid, non-portable, broken) set_affinity syscall he's
> right.
actually, affinity still works just fine, users can bind tasks to logical
CPUs as well. What i meant was the affinity logic of the scheduler (ie.
affinity decisions done by the scheduler), not the externally visible
affinity API.
> You don't know if someone said "schedule me on cpu 0" because they
> really want to be scheduled on CPU 0, or because they really *don't*
> want to be scheduled on CPU 1 (where something else is running). You
> can't just assume they are equivalent if they are the same physical CPU.
i dont assume that. There's also a fair amount of code in the kernel that
relies on binding threads to particular CPUs, the patch does not break
that in any way.
Ingo
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Sat Aug 31 2002 - 22:00:23 EST