Hi,
this series gets rid of charge reparenting at cgroup deletion, which
is possible now that the css can outlive the user-visible cgroup. Any
cache charges left after cgroup deletion simply remain with their css,
where they continue to get reclaimed during pressure on the parent.
Version 2:
- remove memcg->dead_count [vladimir]
- restore iterator generations [vladimir]
- restore memcg-initialized test [michal]
- document shared walk lockless magic [michal]
- split out sync stock draining removal [michal]
include/linux/cgroup.h | 26 ++
include/linux/page_counter.h | 4 +-
include/linux/percpu-refcount.h | 47 ++-
mm/memcontrol.c | 593 ++++++--------------------------------
mm/page_counter.c | 23 +-
5 files changed, 163 insertions(+), 530 deletions(-)
--
To unsubscribe from this list: send the line "unsubscribe cgroups" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
