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: <20230403103440.2895683-1-vladimir.oltean@nxp.com>
 <20230403103440.2895683-7-vladimir.oltean@nxp.com> <CAM0EoMn9iwTBUW-OaK2sDtTS-PO2_nGLuvGmrqY5n8HYEdt7XQ@mail.gmail.com>
 <20230407164103.vstxn2fmswno3ker@skbuf>
In-Reply-To: <20230407164103.vstxn2fmswno3ker@skbuf>
From:   Jamal Hadi Salim <jhs@mojatatu.com>
Date:   Fri, 7 Apr 2023 14:49:01 -0400
Message-ID: <CAM0EoM=go4RNohHpt6Z9wFk0AU81gJY3puBTUOC6F0xMocJouQ@mail.gmail.com>
Subject: Re: [PATCH v4 net-next 6/9] net/sched: mqprio: allow per-TC user
 input of FP adminStatus
To:     Vladimir Oltean <vladimir.oltean@nxp.com>
Cc:     netdev@vger.kernel.org, "David S. Miller" <davem@davemloft.net>,
        Eric Dumazet <edumazet@google.com>,
        Jakub Kicinski <kuba@kernel.org>,
        Paolo Abeni <pabeni@redhat.com>,
        Cong Wang <xiyou.wangcong@gmail.com>,
        Jiri Pirko <jiri@resnulli.us>,
        Vinicius Costa Gomes <vinicius.gomes@intel.com>,
        Kurt Kanzenbach <kurt@linutronix.de>,
        Gerhard Engleder <gerhard@engleder-embedded.com>,
        Amritha Nambiar <amritha.nambiar@intel.com>,
        Ferenc Fejes <ferenc.fejes@ericsson.com>,
        Xiaoliang Yang <xiaoliang.yang_1@nxp.com>,
        Roger Quadros <rogerq@kernel.org>,
        Pranavi Somisetty <pranavi.somisetty@amd.com>,
        Harini Katakam <harini.katakam@amd.com>,
        Giuseppe Cavallaro <peppe.cavallaro@st.com>,
        Alexandre Torgue <alexandre.torgue@foss.st.com>,
        Michael Sit Wei Hong <michael.wei.hong.sit@intel.com>,
        Mohammad Athari Bin Ismail <mohammad.athari.ismail@intel.com>,
        Oleksij Rempel <linux@rempel-privat.de>,
        Jacob Keller <jacob.e.keller@intel.com>,
        linux-kernel@vger.kernel.org, Ferenc Fejes <fejes@inf.elte.hu>,
        Simon Horman <simon.horman@corigine.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Fri, Apr 7, 2023 at 12:41=E2=80=AFPM Vladimir Oltean <vladimir.oltean@nx=
p.com> wrote:
>
> On Fri, Apr 07, 2023 at 12:22:26PM -0400, Jamal Hadi Salim wrote:
> > > +enum {
> > > +       TC_FP_EXPRESS =3D 1,
> > > +       TC_FP_PREEMPTIBLE =3D 2,
> > > +};
> >
> > Suggestion: Add a MAX to this enum (as is traditionally done)..
>
> Max what? This doesn't count anything, it just expresses whether the
> quality of one traffic class, from the Frame Preemption standard's
> perspective, is express or preemptible...
>
> > > @@ -145,13 +149,94 @@ static int mqprio_parse_opt(struct net_device *=
dev, struct tc_mqprio_qopt *qopt,
> > >         return 0;
> > >  }
> > >
> > > +static const struct
> > > +nla_policy mqprio_tc_entry_policy[TCA_MQPRIO_TC_ENTRY_MAX + 1] =3D {
> > > +       [TCA_MQPRIO_TC_ENTRY_INDEX]     =3D NLA_POLICY_MAX(NLA_U32,
> > > +                                                        TC_QOPT_MAX_=
QUEUE),
> >
> > And use it here...
>
> Where? Above or below the comment? I think you mean below (for the
> policy of TCA_MQPRIO_TC_ENTRY_FP)?
>

That was what I meant. I misread that code thinking it was a nested
TLV range check. If it is only going to be those two specific values,
I understand - but then wondering why you
need a u32; wouldnt a u8 be sufficient? The only reason you would need
a MAX is if it is possible that new values greater than
TC_FP_PREEMPTIBLE showing up in the future.

> > Out of curiosity, could you have more that 16 queues in this spec? I
> > noticed 802.1p mentioned somewhere (typically 3 bits).
>
> "This spec" is IEEE 802.1Q :) It doesn't say how many "queues" (struct
> netdev_queue) there are, and this UAPI doesn't work with those, either.
> The standard defines 8 priority values, groupable in (potentially fewer)
> traffic classes. Linux liked to extend the number of traffic classes to
> 16 (and the skb->priority values are arbitrarily large IIUC) and this is
> where that number 16 came from. The number of 16 traffic classes still
> allows for more than 16 TXQs though.
>
> > Lead up question: if the max is 16 then can preemptible_tcs for example=
 be u32?
>
> I don't understand this question, sorry. preemptible_tcs is declared as
> "unsigned long", which IIUC is at least 32-bit.

I meant: if you only had 16 possible values, meaning 16 bits are
sufficient, (although i may be misunderstanding the goal of those
bits) why not be explicit and use the proper type/size?

cheers,
jamal

> >
> > > +       [TCA_MQPRIO_TC_ENTRY_FP]        =3D NLA_POLICY_RANGE(NLA_U32,
> > > +                                                          TC_FP_EXPR=
ESS,
> > > +                                                          TC_FP_PREE=
MPTIBLE),