Received-SPF: softfail (google.com: domain of transitioning linux-kernel-owner@vger.kernel.org does not designate 23.128.96.19 as permitted sender) client-ip=23.128.96.19;
Message-ID: <0d0fb94a-ff66-ac31-e126-0eaf4dca0d6a@redhat.com>
Date:   Mon, 21 Mar 2022 21:09:46 -0600
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.5.0
Subject: Re: [PATCH v5] mm/oom_kill.c: futex: Close a race between do_exit and
 the oom_reaper
Content-Language: en-US
To:     Michal Hocko <mhocko@suse.com>, Davidlohr Bueso <dave@stgolabs.net>
References: <20220318033621.626006-1-npache@redhat.com>
 <Yjg9ncgep58gFLiN@dhcp22.suse.cz> <20220322004231.rwmnbjpq4ms6fnbi@offworld>
 <c8bb0b6d-981c-8591-d5b6-17414c934758@redhat.com>
 <20220322025724.j3japdo5qocwgchz@offworld>
From:   Nico Pache <npache@redhat.com>
Cc:     linux-mm@kvack.org, Andrea Arcangeli <aarcange@redhat.com>,
        Joel Savitz <jsavitz@redhat.com>,
        Andrew Morton <akpm@linux-foundation.org>,
        linux-kernel@vger.kernel.org, Rafael Aquini <aquini@redhat.com>,
        Waiman Long <longman@redhat.com>, Baoquan He <bhe@redhat.com>,
        Christoph von Recklinghausen <crecklin@redhat.com>,
        Don Dutile <ddutile@redhat.com>,
        "Herton R . Krzesinski" <herton@redhat.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Darren Hart <dvhart@infradead.org>,
        Andre Almeida <andrealmeid@collabora.com>,
        David Rientjes <rientjes@google.com>
In-Reply-To: <20220322025724.j3japdo5qocwgchz@offworld>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk


On 3/21/22 20:57, Davidlohr Bueso wrote:
> On Mon, 21 Mar 2022, Nico Pache wrote:
> 
>> We could proceed with the V3 approach; however if we are able to find a complete
>> solution that keeps both functionalities (Concurrent OOM Reaping & Robust Futex)
>> working, I dont see why we wouldnt go for it.
> 
> Because semantically killing the process is, imo, the wrong thing to do. My
> performance argument before however is bogus as the overhead of robust futexes
> is pretty negligible within the lifetime of a lock. That said, the users still
> have good(?) reasons for not wanting the lock holder to crash on them.

From my understanding, the whole point of the robust futex is to allow forward
progress in an application in which the lock holder CAN crash/exit/oom. So
semantically nothing is wrong with killing the futex holder... the whole point
of the robustness is to handle these cases. We just have a case were the oom
killer is racing with said handling of the futex, invalidating the memory before
the exit path (handle_futex_death) can awake one of the other waiters.

-- Nico
> 
> Thanks,
> Davidlohr
>