Re: [2.6.24-rc8-mm1][regression?] numactl --interleave=all doesn'tworks on memoryless node.

[David Rientjes]

On Tue, 5 Feb 2008, Paul Jackson wrote:

> David wrote:
> > The more alarming result of these remaps is in the MPOL_BIND case, as 
> > we've talked about before.  The language in set_mempolicy(2):
> 
> You're diving into the middle of a rather involved discussion
> we had on the other various patches proposed to extend the
> interaction of mempolicy's with cpusets and hotplug.
> 

I've simply identified that MPOL_BIND mempolicy interactions with a task's 
changing mems_allowed as a result of a cpuset move or mems change is also 
an issue that can be addressed at the same time as the interleave problem.  

And it can be done with the addition of a single MPOL_F_* flag.

> I choose not to hijack this current thread with my rebuttal,
> which you've seen before, of your points here.
> 

The issues of mempolicies working over memoryless nodes and supporting 
changing cpusets are very closely related and can be addressed in the same 
way.  It would be disappointing to see a lot of work done to fix the 
memoryless node issue or the changing cpuset mems issue and then realize 
both could have been fixed quite simply with a relatively small set of 
changes.

		David
--
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/