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;
MIME-Version: 1.0
References: <20211116012912.723980-1-longman@redhat.com> <CAD=FV=URCo5xv3k3jWbxV1uRkUU5k6bcnuB1puZhxayEyVc6-A@mail.gmail.com>
 <20220719104104.1634-1-hdanton@sina.com> <CAD=FV=Xng_Mcd-9SaK29XSbsthDWLTR6sz53VEktAQFng6a27A@mail.gmail.com>
 <20220722115510.2101-1-hdanton@sina.com>
In-Reply-To: <20220722115510.2101-1-hdanton@sina.com>
From:   Doug Anderson <dianders@chromium.org>
Date:   Fri, 22 Jul 2022 07:02:42 -0700
Message-ID: <CAD=FV=XS_AqMveo=a2zN+=b3DxM-yJQapd27Dq8dR+kSb2oL_A@mail.gmail.com>
Subject: Re: [PATCH v5] locking/rwsem: Make handoff bit handling more consistent
To:     Hillf Danton <hdanton@sina.com>
Cc:     Waiman Long <longman@redhat.com>,
        Peter Zijlstra <peterz@infradead.org>,
        Will Deacon <will@kernel.org>,
        Davidlohr Bueso <dave@stgolabs.net>, MM <linux-mm@kvack.org>,
        LKML <linux-kernel@vger.kernel.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

Hi,

On Fri, Jul 22, 2022 at 4:55 AM Hillf Danton <hdanton@sina.com> wrote:
>
> On Tue, 19 Jul 2022 08:30:02 -0700 Doug Anderson wrote:
> >
> > I haven't done any stress testing other than my test case, though, so
> > I can't speak to whether there might be any other unintended issues.
>
> The diff below is prepared for any regressions I can imagine in stress
> tests by adding changes to both read and write acquirer slow pathes.
>
> On the read side, make lock stealing more aggressive; on the other hand,
> write acquirers try to set HANDOFF after a RWSEM_WAIT_TIMEOUT nap to
> force the reader acquirers to take the slow path.
>
> Hillf
>
> --- a/kernel/locking/rwsem.c
> +++ b/kernel/locking/rwsem.c
> @@ -992,13 +992,7 @@ rwsem_down_read_slowpath(struct rw_semap
>         struct rwsem_waiter waiter;
>         DEFINE_WAKE_Q(wake_q);
>
> -       /*
> -        * To prevent a constant stream of readers from starving a sleeping
> -        * waiter, don't attempt optimistic lock stealing if the lock is
> -        * currently owned by readers.
> -        */
> -       if ((atomic_long_read(&sem->owner) & RWSEM_READER_OWNED) &&
> -           (rcnt > 1) && !(count & RWSEM_WRITER_LOCKED))
> +       if (WARN_ON_ONCE(count & RWSEM_FLAG_READFAIL))
>                 goto queue;
>
>         /*
> @@ -1169,7 +1163,11 @@ rwsem_down_write_slowpath(struct rw_sema
>                                 goto trylock_again;
>                 }
>
> -               schedule();
> +               if (RWSEM_FLAG_HANDOFF & atomic_long_read(&sem->count))
> +                       schedule();
> +               else
> +                       schedule_timeout(1 + RWSEM_WAIT_TIMEOUT);
> +
>                 lockevent_inc(rwsem_sleep_writer);
>                 set_current_state(state);
>  trylock_again:
> --

Thanks! I added this diff to your previous diff and my simple test
still passes and I don't see your WARN_ON triggered.

How do we move forward? Are you going to officially submit a patch
with both of your diffs squashed together? Are we waiting for
additional review from someone?

-Doug