Re: [PATCH 3/3] x86: fix node_possible_map logic -v2

[Andi Kleen]

> Should it?  It seems to me that CPUs 0-1 should be antipreferentially 
> scheduled, 

You could do that, but the question is if it matters. It would
only make a difference on systems which are not fully loaded, 
and it's unclear how much.

And at some point you need to use these cores anyways; usually it's much
worse to not use a CPU and overload others than to use it with slower memory.

Is it worth adding a lot of fixes all over hte kernel? Not clear
to me.

Also it's a obscure situation and it affects a lot of code, so you
would likely have to continuously fight with regressions as Jack
has discovered.

-Andi
-- 
ak@xxxxxxxxxxxxxxx -- Speaking for myself only.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/