On Fri, Oct 24, 2014 at 09:40:20AM -0400, Johannes Weiner wrote:
> While moving charges from one memcg to another, page stat updates must
> acquire the old memcg's move_lock to prevent double accounting. That
> situation is denoted by an increased memcg->move_accounting. However,
> the charge moving code declares this way too early for now, even
> before summing up the RSS and pre-allocating destination charges.
>
> Shorten this slowpath mode by increasing memcg->move_accounting only
> right before walking the task's address space with the intention of
> actually moving the pages.
>
> Signed-off-by: Johannes Weiner <hannes@xxxxxxxxxxx>
Reviewed-by: Vladimir Davydov <vdavydov@xxxxxxxxxxxxx>
> ---
> mm/memcontrol.c | 21 ++++++++-------------
> 1 file changed, 8 insertions(+), 13 deletions(-)
>
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index c50176429fa3..23cf27cca370 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -5263,8 +5263,6 @@ static void __mem_cgroup_clear_mc(void)
>
> static void mem_cgroup_clear_mc(void)
> {
> - struct mem_cgroup *from = mc.from;
> -
> /*
> * we must clear moving_task before waking up waiters at the end of
> * task migration.
> @@ -5275,8 +5273,6 @@ static void mem_cgroup_clear_mc(void)
> mc.from = NULL;
> mc.to = NULL;
> spin_unlock(&mc.lock);
> -
> - atomic_dec(&from->moving_account);
> }
>
> static int mem_cgroup_can_attach(struct cgroup_subsys_state *css,
> @@ -5310,15 +5306,6 @@ static int mem_cgroup_can_attach(struct cgroup_subsys_state *css,
> VM_BUG_ON(mc.moved_charge);
> VM_BUG_ON(mc.moved_swap);
>
> - /*
> - * Signal mem_cgroup_begin_page_stat() to take
> - * the memcg's move_lock while we're moving
> - * its pages to another memcg. Then wait for
> - * already started RCU-only updates to finish.
> - */
> - atomic_inc(&from->moving_account);
> - synchronize_rcu();
> -
> spin_lock(&mc.lock);
> mc.from = from;
> mc.to = memcg;
> @@ -5450,6 +5437,13 @@ static void mem_cgroup_move_charge(struct mm_struct *mm)
> struct vm_area_struct *vma;
>
> lru_add_drain_all();
> + /*
> + * Signal mem_cgroup_begin_page_stat() to take the memcg's
> + * move_lock while we're moving its pages to another memcg.
> + * Then wait for already started RCU-only updates to finish.
> + */
> + atomic_inc(&mc.from->moving_account);
> + synchronize_rcu();
> retry:
> if (unlikely(!down_read_trylock(&mm->mmap_sem))) {
> /*
> @@ -5482,6 +5476,7 @@ retry:
> break;
> }
> up_read(&mm->mmap_sem);
> + atomic_dec(&mc.from->moving_account);
> }
>
> static void mem_cgroup_move_task(struct cgroup_subsys_state *css,
> --
> 2.1.2
>
--
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>