mm: unify new_node_page and alloc_migrate_target
authorMichal Hocko <mhocko@suse.com>
Mon, 10 Jul 2017 22:48:47 +0000 (15:48 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Mon, 10 Jul 2017 23:32:31 +0000 (16:32 -0700)
commit8b9132388964df2cfe151a88fd1dd8219dabf23c
tree4b4fb6c14c4cc44bcf5be7a4827ba2681f6f7a3e
parent4db9b2efe94967be34e3b136a93251a3c1736dd5
mm: unify new_node_page and alloc_migrate_target

Commit 394e31d2ceb4 ("mem-hotplug: alloc new page from a nearest
neighbor node when mem-offline") has duplicated a large part of
alloc_migrate_target with some hotplug specific special casing.

To be more precise it tried to enfore the allocation from a different
node than the original page.  As a result the two function diverged in
their shared logic, e.g.  the hugetlb allocation strategy.

Let's unify the two and express different NUMA requirements by the given
nodemask.  new_node_page will simply exclude the node it doesn't care
about and alloc_migrate_target will use all the available nodes.
alloc_migrate_target will then learn to migrate hugetlb pages more
sanely and use preallocated pool when possible.

Please note that alloc_migrate_target used to call alloc_page resp.
alloc_pages_current so the memory policy of the current context which is
quite strange when we consider that it is used in the context of
alloc_contig_range which just tries to migrate pages which stand in the
way.

Link: http://lkml.kernel.org/r/20170608074553.22152-4-mhocko@kernel.org
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Cc: Xishi Qiu <qiuxishi@huawei.com>
Cc: zhong jiang <zhongjiang@huawei.com>
Cc: Joonsoo Kim <js1304@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
include/linux/migrate.h
mm/memory_hotplug.c
mm/page_isolation.c