Re: [PATCH 10/12] Add /sys trigger for per-node memory compaction

[Mel Gorman <mel@csn.ul.ie>]

On Fri, Feb 19, 2010 at 07:31:42AM -0800, Greg KH wrote:
> On Fri, Feb 19, 2010 at 03:28:30PM +0000, Mel Gorman wrote:
> > On Fri, Feb 19, 2010 at 06:53:59AM -0800, Greg KH wrote:
> > > On Thu, Feb 18, 2010 at 06:02:40PM +0000, Mel Gorman wrote:
> > > > This patch adds a per-node sysfs file called compact. When the file is
> > > > written to, each zone in that node is compacted. The intention that this
> > > > would be used by something like a job scheduler in a batch system before
> > > > a job starts so that the job can allocate the maximum number of
> > > > hugepages without significant start-up cost.
> > > 
> > > As you are adding sysfs files, can you please also add documentation for
> > > the file in Documentation/ABI/ ?
> > > 
> > 
> > I looked at this before and hit a wall and then forgot about it. I couldn't
> > find *where* I should document it at the time. There isn't a sysfs-devices-node
> > file to add to and much (all?) of what is in that branch appears undocumented.
> 
> Well, you can always just document what you add, or you can document the
> existing stuff as well.  It's your choice :)
> 

Fair point!

I've taken note to document what's in there over time. For the moment,
is this a reasonable start? I'll split it into two patches but the end
result will be the same.

diff --git a/Documentation/ABI/testing/sysfs-devices-node b/Documentation/ABI/testing/sysfs-devices-node
new file mode 100644
index 0000000..1ee348b
--- /dev/null
+++ b/Documentation/ABI/testing/sysfs-devices-node
@@ -0,0 +1,15 @@
+What:		/sys/devices/system/node/nodeX
+Date:		October 2002
+Contact:	Linux Memory Management list <linux-mm@kvack.org>
+Description:
+		When CONFIG_NUMA is enabled, this is a directory containing
+		information on node X such as what CPUs are local to the
+		node.
+
+What:		/sys/devices/system/node/nodeX/compact
+Date:		February 2010
+Contact:	Mel Gorman <mel@csn.ul.ie>
+Description:
+		When this file is written to, all memory within that node
+		will be compacted. When it completes, memory will be free
+		in as contiguous blocks as possible.