Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Sender: Tejun Heo <htejun@gmail.com>
Date:   Mon, 17 Jul 2023 09:52:24 -1000
From:   Tejun Heo <tj@kernel.org>
To:     Hao Jia <jiahao.os@bytedance.com>
Cc:     lizefan.x@bytedance.com, hannes@cmpxchg.org, mkoutny@suse.com,
        cgroups@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: Re: [PATCH] cgroup/rstat: record the cumulative per-cpu time of
 cgroup and its descendants
Message-ID: <ZLWb-LsBD041hMvr@slm.duckdns.org>
References: <20230717093612.40846-1-jiahao.os@bytedance.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230717093612.40846-1-jiahao.os@bytedance.com>
Precedence: bulk

Hello,

On Mon, Jul 17, 2023 at 05:36:12PM +0800, Hao Jia wrote:
> Now the member variable bstat of the structure cgroup_rstat_cpu

You said "now" indicating that the behavior has changed recently but I don't
see what changed there. Can you elaborate?

> records the per-cpu time of the cgroup itself, but does not
> include the per-cpu time of its descendants. The per-cpu time

It does. The per-cpu delta is added to its parent and then that will in turn
be used to propagate to its parent.

> including descendants is very useful for calculating the
> per-cpu usage of cgroups.
> 
> Although we can indirectly obtain the total per-cpu time
> of the cgroup and its descendants by accumulating the per-cpu
> bstat of each descendant of the cgroup. But after a child cgroup
> is removed, we will lose its bstat information. This will cause
> the cumulative value to be non-monotonic, thus affecting
> the accuracy of cgroup per-cpu usage.
> 
> So we add the cumul_bstat variable to record the total
> per-cpu time of this cgroup and its descendants, which is
> similar to "cpuacct.usage*" in cgroup v1. And this is
> also helpful for the migration from cgroup v1 to cgroup v2.
> After adding this variable, we can obtain the per-cpu time of
> cgroup and its descendants in user mode through eBPF, etc.

I think you're misunderstanding how the code works. Can you please double
check?

Thanks.

-- 
tejun