MIME-Version: 1.0
References: <20230202155626.1829121-1-hannes@cmpxchg.org>
In-Reply-To: <20230202155626.1829121-1-hannes@cmpxchg.org>
From:   Shakeel Butt <shakeelb@google.com>
Date:   Thu, 2 Feb 2023 10:27:16 -0800
Message-ID: <CALvZod6m-XOb6LvjAD0wXv1doP_LJb1yUJ-QRpJt3H+G=xyQXg@mail.gmail.com>
Subject: Re: [RFC PATCH] mm: memcontrol: don't account swap failures not due
 to cgroup limits
To:     Johannes Weiner <hannes@cmpxchg.org>
Cc:     Michal Hocko <mhocko@suse.com>,
        Roman Gushchin <roman.gushchin@linux.dev>,
        Tejun Heo <tj@kernel.org>, linux-mm@kvack.org,
        cgroups@vger.kernel.org, linux-kernel@vger.kernel.org,
        Christian Brauner <brauner@kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, Feb 2, 2023 at 7:56 AM Johannes Weiner <hannes@cmpxchg.org> wrote:
>
> Christian reports the following situation in a cgroup that doesn't
> have memory.swap.max configured:
>
>   $ cat memory.swap.events
>   high 0
>   max 0
>   fail 6218
>
> Upon closer examination, this is an ARM64 machine that doesn't support
> swapping out THPs. In that case, the first get_swap_page() fails, and
> the kernel falls back to splitting the THP and swapping the 4k
> constituents one by one. /proc/vmstat confirms this with a high rate
> of thp_swpout_fallback events.
>
> While the behavior can ultimately be explained, it's unexpected and
> confusing. I see three choices how to address this:
>
> a) Specifically exlude THP fallbacks from being counted, as the
>    failure is transient and the memory is ultimately swapped.
>
>    Arguably, though, the user would like to know if their cgroup's
>    swap limit is causing high rates of THP splitting during swapout.
>
> b) Only count cgroup swap events when they are actually due to a
>    cgroup's own limit. Exclude failures that are due to physical swap
>    shortage or other system-level conditions (like !THP_SWAP). Also
>    count them at the level where the limit is configured, which may be
>    above the local cgroup that holds the page-to-be-swapped.
>
>    This is in line with how memory.swap.high, memory.high and
>    memory.max events are counted.
>
>    However, it's a change in documented behavior.
>
> c) Leave it as is. The documentation says system-level events are
>    counted, so stick to that.
>
>    This is the conservative option, but isn't very user friendly.
>    Cgroup events are usually due to a local control choice made by the
>    user. Mixing in events that are beyond the user's control makes it
>    difficult to id root causes and configure the system properly.
>
> Implement option b).

I prefer option b too.

>
> Reported-by: Christian Brauner <brauner@kernel.org>
> Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>

Acked-by: Shakeel Butt <shakeelb@google.com>

I think we should CC stable as well for early exposure.