What is the nature and description of the request?

They want to decrease the overhead of cgroups. Today, the kernel uses  a lot of
2MB slab objects (8 objects per 1GB of RAM on x86_64 with SPARSEMEM). It turned
out that this memory is allocated for per memory sections page_cgroup arrays. 

    Why does the customer need this? (List the business requirements here)

More available memory is always a good thing.

    How would the customer like to achieve this? (List the functional
requirements here)

We can reduce this internal fragmentation by splitting the single page_cgroup
array into more arrays where each one is well kmalloc aligned. This patch
	http://thread.gmane.org/gmane.linux.kernel.mm/58765/focus=58940 
implements this idea.

    For each functional requirement listed in question 4, specify how Red Hat
and the customer can test to confirm the requirement is successfully
implemented.

Look at slab statistics to see that there are fewer 2 mb slab objects.

    Is there already an existing RFE upstream or in Red Hat bugzilla?

Not yet.

    How quickly does this need resolved? (desired target release)

minor

    Does this request meet the RHEL Inclusion criteria (please review)

yes

    List the affected packages

kernel

    Would the customer be able to assist in testing this functionality if
implemented?

yes