Andrew, As you requested, here are some fixups and clarifications for the kmemcg series. It also handles one bug reported by Sasha. Please note that I didn't touch kmem_cache_shrink(): I believe that deserves a deeper and more thoughtful solution that will take time to brew. I plan to address that eventually in the scope of per-memcg kernel memory reclaim. I did, however, remove the delayed_work in favor of a normal worker. Memory will stay around for longer, but it will be reclaimed eventually, and given your objections I believe this is a more desirable trade off. Please let me know if there is anything you would like to see different, and sorry for not handling this earlier. Glauber Costa (7): memcg: simplify ida initialization move include of workqueue.h to top of slab.h file memcg: remove test for current->mm in memcg_stop/resume_kmem_account memcg: replace __always_inline with plain inline memcg: get rid of once-per-second cache shrinking for dead memcgs memcg: add comments clarifying aspects of cache attribute propagation slub: drop mutex before deleting sysfs entry include/linux/memcontrol.h | 12 +++++++++--- include/linux/slab.h | 6 +++--- mm/memcontrol.c | 34 ++++++++++------------------------ mm/slab.c | 1 + mm/slub.c | 34 +++++++++++++++++++++++++++++----- 5 files changed, 52 insertions(+), 35 deletions(-) -- 1.7.11.7 -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>