mm: numa: Add fault driven placement and migration
authorPeter Zijlstra <a.p.zijlstra@chello.nl>
Thu, 25 Oct 2012 12:16:43 +0000 (14:16 +0200)
committerMel Gorman <mgorman@suse.de>
Tue, 11 Dec 2012 14:42:45 +0000 (14:42 +0000)
commitcbee9f88ec1b8dd6b58f25f54e4f52c82ed77690
treed4cfbcfa3e89742216cd792d4aa914356406b532
parenta720094ded8cbb303111035be91858011d2eac71
mm: numa: Add fault driven placement and migration

NOTE: This patch is based on "sched, numa, mm: Add fault driven
placement and migration policy" but as it throws away all the policy
to just leave a basic foundation I had to drop the signed-offs-by.

This patch creates a bare-bones method for setting PTEs pte_numa in the
context of the scheduler that when faulted later will be faulted onto the
node the CPU is running on.  In itself this does nothing useful but any
placement policy will fundamentally depend on receiving hints on placement
from fault context and doing something intelligent about it.

Signed-off-by: Mel Gorman <mgorman@suse.de>
Acked-by: Rik van Riel <riel@redhat.com>
arch/sh/mm/Kconfig
arch/x86/Kconfig
include/linux/mm_types.h
include/linux/sched.h
kernel/sched/core.c
kernel/sched/fair.c
kernel/sched/features.h
kernel/sched/sched.h
kernel/sysctl.c
mm/huge_memory.c
mm/memory.c