Re: [PATCH/RFC] Add node states sysfs class attributeS - V5

[Andrew Morton <akpm@linux-foundation.org>]

On Tue, 11 Sep 2007 09:56:15 -0400 Lee Schermerhorn <Lee.Schermerhorn@hp.com> wrote:

> Should be about ready to go...
> 
> Lee
> 
> 
> PATCH Add node 'states' sysfs class attributes v5
> 
> Against:  2.6.23-rc4-mm1
> 
> V4 -> V5:
> + further cleanup of print_nodes_state() suggested by Chirstoph.
> 
> V3 -> V4:
> + drop the annotations -- not needed with one value per file.
> + this simplifies print_nodes_state()
> + fix "function return type on separate line" style glitch
> 
> V2 -> V3:
> + changed to per state sysfs file -- "one value per file"
> 
> V1 -> V2:
> + style cleanup
> + drop 'len' variable in print_node_states();  compute from
>   final size.
> 
> Add a per node state sysfs class attribute file to
> /sys/devices/system/node to display node state masks.
> 
> E.g., on a 4-cell HP ia64 NUMA platform, we have 5 nodes:
> 4 representing the actual hardware cells and one memory-only
> pseudo-node representing a small amount [512MB] of "hardware
> interleaved" memory.  With this patch, in /sys/devices/system/node
> we see:
> 
> #ls -1F /sys/devices/system/node
> has_cpu
> has_normal_memory
> node0/
> node1/
> node2/
> node3/
> node4/
> online
> possible
> #cat /sys/devices/system/node/possible
> 0-255
> #cat /sys/devices/system/node/online
> 0-4
> #cat /sys/devices/system/node/has_normal_memory
> 0-4
> #cat /sys/devices/system/node/has_cpu
> 0-3
> 
> N.B., NOT TESTED with CONFIG_HIGHMEM=y.
> 

So how do we get it tested with CONFIG_HIGHMEM=y?  Needs an i386
numa machine, yes?  Perhaps Andy or Martin can remember to do this
sometime, but they'll need a test plan ;)

Thanks.

--
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>