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: <0dfee8c9d17c20f9a87c39dbc57f635d998b08d2.1648609552.git.jamie.bainbridge@gmail.com>
 <YkSzLJ72M5f5EL2L@t14s.localdomain>
In-Reply-To: <YkSzLJ72M5f5EL2L@t14s.localdomain>
From:   Jamie Bainbridge <jamie.bainbridge@gmail.com>
Date:   Fri, 1 Apr 2022 10:09:56 +1000
Message-ID: <CAAvyFNgL1_YsnkGdJM8t9L1zT60AEfUMeReVx=2DTtLZ_WLScQ@mail.gmail.com>
Subject: Re: [PATCH v3 net] sctp: count singleton chunks in assoc user stats
To:     Marcelo Ricardo Leitner <marcelo.leitner@gmail.com>
Cc:     Vlad Yasevich <vyasevich@gmail.com>,
        Neil Horman <nhorman@tuxdriver.com>,
        "David S. Miller" <davem@davemloft.net>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>, linux-sctp@vger.kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Thu, 31 Mar 2022 at 05:44, Marcelo Ricardo Leitner
<marcelo.leitner@gmail.com> wrote:
>
> On Wed, Mar 30, 2022 at 01:06:02PM +1000, Jamie Bainbridge wrote:
> > Singleton chunks (INIT, HEARTBEAT PMTU probes, and SHUTDOWN-
> > COMPLETE) are not counted in SCTP_GET_ASOC_STATS "sas_octrlchunks"
> > counter available to the assoc owner.
> >
> > These are all control chunks so they should be counted as such.
> >
> > Add counting of singleton chunks so they are properly accounted for.
> >
> > Fixes: 196d67593439 ("sctp: Add support to per-association statistics via a new SCTP_GET_ASSOC_STATS call")
> > Signed-off-by: Jamie Bainbridge <jamie.bainbridge@gmail.com>
> > ---
> >  net/sctp/outqueue.c | 2 ++
> >  1 file changed, 2 insertions(+)
> >
> > diff --git a/net/sctp/outqueue.c b/net/sctp/outqueue.c
> > index a18609f608fb786b2532a4febbd72a9737ab906c..bed34918b41f24810677adc0cd4fbd0859396a02 100644
> > --- a/net/sctp/outqueue.c
> > +++ b/net/sctp/outqueue.c
> > @@ -914,6 +914,7 @@ static void sctp_outq_flush_ctrl(struct sctp_flush_ctx *ctx)
> >                               ctx->asoc->base.sk->sk_err = -error;
> >                               return;
> >                       }
> > +                     ctx->asoc->stats.octrlchunks++;
> >                       break;
> >
> >               case SCTP_CID_ABORT:
> > @@ -939,6 +940,7 @@ static void sctp_outq_flush_ctrl(struct sctp_flush_ctx *ctx)
> >               case SCTP_CID_HEARTBEAT:
> >                       if (chunk->pmtu_probe) {
> >                               sctp_packet_singleton(ctx->transport, chunk, ctx->gfp);
> > +                             ctx->asoc->stats.octrlchunks++;
>
> sctp_packet_singleton can fail. It shouldn't be propagated to the
> socket but octrlchunks shouldn't be incremented then. Not too diferent
> from the one above.

Ah, thanks for the catch! Is this syntax assigning to error okay?

error = sctp_packet_singleton(ctx->transport, chunk, ctx->gfp);
if (!error)
          ctx->asoc->stats.octrlchunks++;
break;

> >                               break;
> >                       }
> >                       fallthrough;
> > --
> > 2.35.1
> >