Re: [PATCH 0/5] Huge Pages Nodes Allowed

[David Rientjes <rientjes@google.com>]

On Thu, 18 Jun 2009, David Rientjes wrote:

> Manipulating hugepages via a nodemask seems less ideal than, as you 
> mentioned, per-node hugepage controls, probably via 
> /sys/kernel/system/node/node*/nr_hugepages.  This type of interface 
> provides all the functionality that this patchset does, including hugepage 
> allocation and freeing, but with more power to explicitly allocate and 
> free on targeted nodes.  /proc/sys/vm/nr_hugepages would remain to 
> round-robin the allocation (and freeing, with your patch 1/5 which I 
> ack'd).
> 
> Such an interface would also automatically deal with all memory 
> hotplug/remove issues without storing or keeping a nodemask updated.
> 

Expanding this proposal out a little bit, we'd want all the power of the 
/sys/kernel/mm/hugepages tunables for each node.  The best way of doing 
that is probably to keep the current /sys/kernel/mm/hugepages directory as 
is (already published Documentation/ABI/testing/sysfs-kernel-mm-hugepages) 
for the system-wide hugepage state and then add individual 
`hugepages-<size>kB' directories to each /sys/devices/system/node/node* to 
target allocations and freeing for the per-node hugepage state.  
Otherwise, we lack node targeted support for multiple hugepagesz= users.

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>