On Tuesday 13 April 2010 21:16:44 Benjamin Herrenschmidt wrote:

On Tue, 2010-04-13 at 16:02 +1000, Mark Nelson wrote:

quoted

That's a good question, and one that I probably should have added to
the
commit message.
But, following through, it looks like we end up calling into
__remove_section() from mm/memory_hotplug.c and if
CONFIG_SPARSEMEM_VMEMMAP is enabled we just return EBUSY as freeing
memmap with vmemmap isn't implemented yet.

So for the moment, I'm not sure we have to worry about it.

We probably don't. IE. The vmemmap will remain for those struct pages,
which means they won't need to be allocated again if some memory is
plugged back there. If not, we just waste a bit of memory. Not a big
deal.

Excellent! That makes sense.

Thanks Ben!

Mark.