Re: [PATCH] Cpuset: alloc_pages_node overrides cpuset constraints

[Christoph Lameter]

On Mon, 20 Mar 2006, Paul Jackson wrote:

> The patch below just adds the __GFP_NOCPUSET flag on the memory
> migration code path, where we need it.  That code path is not
> performance critical.
> 
> We can deal with the long standing bug in cpusets, where it overrides
> all alloc_pages_node() calls, some other day.
> 
> What think you of this, Christoph?  Should we send it to Andrew?

It does not fix the general problem that alloc_pages_node must not respect 
cpusets. Slab, device drivers etc etc are still screwed up with this 
patch. 

We noticed this first in page migration but I think the other areas are 
much more problematic.

The earlier patch is correct. What Andrew wanted is an optimization 
not a general change in the patch. And

#ifdef CPUSETS

#endif

around the ORing of __GFP_NOCPUSET may suffice.
Maybe you got a better idea?
-
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/