Re: [PATCH v2] mm: memcontrol: protect the memory in cgroup from being oom killed

At 2022-12-08 16:14:10, "Michal Hocko" [off-list ref] wrote:

On Thu 08-12-22 07:59:27, 程垲涛 Chengkaitao Cheng wrote:

quoted

At 2022-12-08 15:33:07, "Michal Hocko" [off-list ref] wrote:

quoted

On Thu 08-12-22 11:46:44, chengkaitao wrote:

quoted

From: chengkaitao <redacted>

We created a new interface <memory.oom.protect> for memory, If there is
the OOM killer under parent memory cgroup, and the memory usage of a
child cgroup is within its effective oom.protect boundary, the cgroup's
tasks won't be OOM killed unless there is no unprotected tasks in other
children cgroups. It draws on the logic of <memory.min/low> in the
inheritance relationship.

It has the following advantages,
1. We have the ability to protect more important processes, when there
is a memcg's OOM killer. The oom.protect only takes effect local memcg,
and does not affect the OOM killer of the host.
2. Historically, we can often use oom_score_adj to control a group of
processes, It requires that all processes in the cgroup must have a
common parent processes, we have to set the common parent process's
oom_score_adj, before it forks all children processes. So that it is
very difficult to apply it in other situations. Now oom.protect has no
such restrictions, we can protect a cgroup of processes more easily. The
cgroup can keep some memory, even if the OOM killer has to be called.

Signed-off-by: chengkaitao <redacted>
---
v2: Modify the formula of the process request memcg protection quota.

The new formula doesn't really address concerns expressed previously.
Please read my feedback carefully again and follow up with questions if
something is not clear.

The previous discussion was quite scattered. Can you help me summarize
your concerns again?

The most important part is http://lkml.kernel.org/r/Y4jFnY7kMdB8ReSW@dhcp22.suse.cz
: Let me just emphasise that we are talking about fundamental disconnect.
: Rss based accounting has been used for the OOM killer selection because
: the memory gets unmapped and _potentially_ freed when the process goes
: away. Memcg changes are bound to the object life time and as said in
: many cases there is no direct relation with any process life time.

We need to discuss the relationship between memcg's mem and process's mem, 

task_usage = task_anon(rss_anon) + task_mapped_file(rss_file) 
	 + task_mapped_share(rss_share) + task_pgtables + task_swapents

memcg_usage	= memcg_anon + memcg_file + memcg_pgtables + memcg_share
	= all_task_anon + all_task_mapped_file + all_task_mapped_share 
	 + all_task_pgtables + unmapped_file + unmapped_share
	= all_task_usage + unmapped_file + unmapped_share - all_task_swapents

Memcg is directly related to processes for most memory. On the other hand, 
unmapped_file pages and unmapped_share pages aren't charged into the 
process, but these memories can not be released by the oom killer. Therefore, 
they should not apply to cgroup for protection quota. They can be excluded 
during calculation.

       memcg A
    /     |     \
task-x  task-y  common-cache
   2G     2G        2G

eoom.protect(memcg A) = 3G;
usage(memcg A) = 6G
usage(task x) = 2G
usage(task y) = 2G
common-cache = 2G

After calculation,
actual-protection(task x) = 1G
actual-protection(task y) = 1G

This formula is more fair for groups with fewer common-caches (unmapped_
file pages and unmapped_share pages).
In extreme environments, unmapped_file pages and unmapped_share pages 
may lock a large share of protection quota, but it is expected.

That is to the per-process discount based on rss or any per-process
memory metrics.

Another really important question is the actual configurability. The
hierarchical protection has to be enforced and that means that same as
memory reclaim protection it has to be enforced top-to-bottom in the
cgroup hierarchy. That makes the oom protection rather non-trivial to
configure without having a good picture of a larger part of the cgroup
hierarchy as it cannot be tuned based on a reclaim feedback.

There is an essential difference between reclaim and oom killer. The reclaim 
cannot be directly perceived by users, so memcg need to count indicators 
similar to pgscan_(kswapd/direct). However, when the user process is killed 
by oom killer, users can clearly perceive and count (such as the number of 
restarts of a certain type of process). At the same time, the kernel also has 
memory.events to count some information about the oom killer, which can 
also be used for feedback adjustment. Of course, I can also add some 
indicators similar to the accumulated memory released by the oom killer 
to help users better grasp the dynamics of the oom killer. Do you think it 
is valuable?
-- 
Thanks for your comment!
chengkaitao