Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Mon, 17 Jan 2022 12:33:33 +0100
From:   Michal Hocko <mhocko@suse.com>
To:     Joel Savitz <jsavitz@redhat.com>
Cc:     Andrew Morton <akpm@linux-foundation.org>,
        linux-kernel <linux-kernel@vger.kernel.org>,
        Waiman Long <longman@redhat.com>, linux-mm@kvack.org,
        Nico Pache <npache@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        Darren Hart <dvhart@infradead.org>,
        Davidlohr Bueso <dave@stgolabs.net>,
        =?iso-8859-1?Q?Andr=E9?= Almeida <andrealmeid@collabora.com>
Subject: Re: [PATCH] mm/oom_kill: wake futex waiters before annihilating
 victim shared mutex
Message-ID: <YeVUDVV4Z5ur9Flh@dhcp22.suse.cz>
References: <20211207214902.772614-1-jsavitz@redhat.com>
 <20211207154759.3f3fe272349c77e0c4aca36f@linux-foundation.org>
 <YbB0d6T8RbHW48sZ@dhcp22.suse.cz>
 <YbDX16LAkvzgYHpH@dhcp22.suse.cz>
 <CAL1p7m4ka1v-Zoi-RpDy5ME-bMikGPX5V_4Hod-Y0KHOq_G8zA@mail.gmail.com>
 <YbG1mu0CLONo+Z7l@dhcp22.suse.cz>
 <CAL1p7m7mWxLE-7Qf_QjmREJ2AvfSexPvybPyHvxTUugxsPPxjQ@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <CAL1p7m7mWxLE-7Qf_QjmREJ2AvfSexPvybPyHvxTUugxsPPxjQ@mail.gmail.com>
Precedence: bulk

I have only noticed your email now after replying to v3 so our emails
have crossed.

On Fri 14-01-22 09:39:55, Joel Savitz wrote:
> > What has happened to the oom victim and why it has never exited?
> 
> What appears to happen is that the oom victim is sent SIGKILL by the
> process that triggers the oom while also being marked as an oom
> victim.
> 
> As you mention in your patchset introducing the oom reaper in commit
> aac4536355496 ("mm, oom: introduce oom reaper"), the purpose the the
> oom reaper is to try and free more memory more quickly than it
> otherwise would have been by assuming anonymous or swapped out pages
> won't be needed in the exit path as the owner is already dying.
> However, this assumption is violated by the futex_cleanup() path,
> which needs access to userspace in fetch_robust_entry() when it is
> called in exit_robust_list(). Trace_printk()s in this failure path
> reveal an apparent race between the oom reaper thread reaping the
> victim's mm and the futex_cleanup() path. There may be other ways that
> this race manifests but we have been most consistently able to trace
> that one.

Please let's continue the discussion in the v3 email thread:
http://lkml.kernel.org/r/20220114180135.83308-1-npache@redhat.com
-- 
Michal Hocko
SUSE Labs