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: <20221218191310.130904-1-joel@joelfernandes.org>
In-Reply-To: <20221218191310.130904-1-joel@joelfernandes.org>
From:   Joel Fernandes <joel@joelfernandes.org>
Date:   Mon, 19 Dec 2022 23:07:17 -0500
Message-ID: <CAEXW_YRjAsx0HCnmjvth+yi0COTiynPRvjyT2sf1utMw5bTgiw@mail.gmail.com>
Subject: Re: [RFC 0/2] srcu: Remove pre-flip memory barrier
To:     linux-kernel@vger.kernel.org
Cc:     Josh Triplett <josh@joshtriplett.org>,
        Lai Jiangshan <jiangshanlai@gmail.com>,
        Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
        "Paul E. McKenney" <paulmck@kernel.org>, rcu@vger.kernel.org,
        Steven Rostedt <rostedt@goodmis.org>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Sun, Dec 18, 2022 at 2:13 PM Joel Fernandes (Google)
<joel@joelfernandes.org> wrote:
>
> Hello, I believe the pre-flip memory barrier is not required. The only reason I
> can say to remove it, other than the possibility that it is unnecessary, is to
> not have extra code that does not help. However, since we are issuing a fully
> memory-barrier after the flip, I cannot say that it hurts to do it anyway.
>
> For this reason, please consider these patches as "informational", than a
> "please merge". :-) Though, feel free to consider merging if you agree!
>
> All SRCU scenarios pass with these, with 6 hours of testing.
>
> thanks,
>
>  - Joel
>
> Joel Fernandes (Google) (2):
> srcu: Remove comment about prior read lock counts
> srcu: Remove memory barrier "E" as it is not required

And litmus tests confirm that "E" does not really do what the comments
say, PTAL:
Test 1:
C mbe
(*
 * Result: sometimes
 * Does previous scan see old reader's lock count, if a new reader saw
the new srcu_idx?
 *)

{}

P0(int *lockcount, int *srcu_idx) // updater
{
        int r0;
        r0 = READ_ONCE(*lockcount);
        smp_mb();       // E
        WRITE_ONCE(*srcu_idx, 1);
}

P1(int *lockcount, int *srcu_idx) // reader
{
        int r0;
        WRITE_ONCE(*lockcount, 1); // previous reader
        smp_mb();       // B+C
        r0 = READ_ONCE(*srcu_idx); // new reader
}
exists (0:r0=0 /\ 1:r0=1) (* Bad outcome. *)

Test 2:
C mbe2

(*
 * Result: sometimes
 * If updater saw reader's lock count, was that reader using the old idx?
 *)

{}

P0(int *lockcount, int *srcu_idx) // updater
{
        int r0;
        r0 = READ_ONCE(*lockcount);
        smp_mb();       // E
        WRITE_ONCE(*srcu_idx, 1);
}

P1(int *lockcount, int *srcu_idx) // reader
{
        int r0;
        int r1;
        r1 = READ_ONCE(*srcu_idx); // previous reader
        WRITE_ONCE(*lockcount, 1); // previous reader
        smp_mb();       // B+C
        r0 = READ_ONCE(*srcu_idx); // new reader
}
exists (0:r0=1 /\ 1:r1=1) (* Bad outcome. *)

thanks,

 - Joel


>
> kernel/rcu/srcutree.c | 10 ----------
> 1 file changed, 10 deletions(-)
>
> --
> 2.39.0.314.g84b9a713c41-goog
>