Date:   Tue, 7 Feb 2023 11:54:55 -0500
From:   Johannes Weiner <hannes@cmpxchg.org>
To:     Michal =?iso-8859-1?Q?Koutn=FD?= <mkoutny@suse.com>
Cc:     Yosry Ahmed <yosryahmed@google.com>,
        Michal Hocko <mhocko@suse.com>,
        Shakeel Butt <shakeelb@google.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>
Subject: Re: [RFC PATCH] mm: memcontrol: don't account swap failures not due
 to cgroup limits
Message-ID: <Y+KCX6XHj/lv9ul3@cmpxchg.org>
References: <20230202155626.1829121-1-hannes@cmpxchg.org>
 <CAJD7tkaCpD0LpzdA+NsZj2WK=iQCLn7RS9qc7K53Qonxhp4TgA@mail.gmail.com>
 <20230206161843.GD21332@blackbody.suse.cz>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <20230206161843.GD21332@blackbody.suse.cz>
Precedence: bulk

On Mon, Feb 06, 2023 at 05:18:43PM +0100, Michal Koutn? wrote:
> On Thu, Feb 02, 2023 at 10:30:40AM -0800, Yosry Ahmed <yosryahmed@google.com> wrote:
> > > 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.
> > 
> > This option makes sense to me, but I can't speak to the change of
> > documented behavior. However, looking at the code, it seems like if we do this
> > the "max" & "fail" counters become effectively the same. "fail" would
> > not provide much value then.
> > 
> > I wonder if it makes sense to have both, and clarify that "fail" -
> > "max" would be non-limit based failures (e.g. ran out of swap space),
> > or would this cause confusion as to whether those non-limit failures
> > were transient (THP fallback) or eventual?
> 
> I somewhat second this.
> 
> Perhaps, could the patch (and arguments) be split in two:
> 1) count .max events on respective limit's level (other limits consistency),

Okay, I'll split this one out. It's good to have regardless of what we
do with the fail counter.

Thanks