Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::3:7 as permitted sender) client-ip=2620:137:e000::3:7;
MIME-Version: 1.0
References: <20230911082016.3694700-1-yajun.deng@linux.dev>
 <CANn89i+W1iAQmOhunLbqpvHu8EUO6uawv6Uvx7qimyBa_PBNCg@mail.gmail.com>
 <f3e84a37-3218-0d52-e7ed-2d215fed58e3@intel.com> <CANn89i+AwmpjM-bNuYRS26v-GRrVoucewxgmkvv25PNM4VWPGA@mail.gmail.com>
 <39c906f6-910d-01c7-404a-8fe6a161ef2e@intel.com> <CANn89i+QSPoXphaLzfKCqCHxjsD20ifr8YPJM_fZ_H5kFZ7dwQ@mail.gmail.com>
 <8bc6c1cd-50bb-44ef-5979-3bb50a70afcb@intel.com> <CANn89iL6HVvRegORfP49prJV4EJU2-AbD4YXB-eo_vwU1JG1ew@mail.gmail.com>
In-Reply-To: <CANn89iL6HVvRegORfP49prJV4EJU2-AbD4YXB-eo_vwU1JG1ew@mail.gmail.com>
From:   Eric Dumazet <edumazet@google.com>
Date:   Tue, 12 Sep 2023 20:05:42 +0200
Message-ID: <CANn89iKbn97Rbjc+3uZMpUi0tqCuhj88UdFZhhnqpC6nJRLC=A@mail.gmail.com>
Subject: Re: [PATCH] net/core: Export dev_core_stats_rx_dropped_inc sets
To:     Alexander Lobakin <aleksander.lobakin@intel.com>
Cc:     Yajun Deng <yajun.deng@linux.dev>, davem@davemloft.net,
        kuba@kernel.org, pabeni@redhat.com, horms@kernel.org,
        netdev@vger.kernel.org, linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Sep 12, 2023 at 8:03=E2=80=AFPM Eric Dumazet <edumazet@google.com> =
wrote:

> Sure, this was what was suggested (perhaps not _very_ precisely, but
> the general idea was pretty clear).
> v2 seems ok, right ?
>
> It seems we are all on the same page.
>
> +static __cold struct net_device_core_stats __percpu
> *dev_core_stats(struct net_device *dev)
> +{
> +       /* This READ_ONCE() pairs with the write in netdev_core_stats_all=
oc() */
> +       struct net_device_core_stats __percpu *p =3D READ_ONCE(dev->core_=
stats);
> +
> +       if (likely(p))
> +               return p;
> +
> +       return netdev_core_stats_alloc(dev);
> +}
> +
> +#define DEV_CORE_STATS_INC(FIELD)                              \
> +void dev_core_stats_##FIELD##_inc(struct net_device *dev)      \
> +{                                                              \
> +       struct net_device_core_stats __percpu *p;               \
> +                                                               \
> +       p =3D dev_core_stats(dev);                                \
> +       if (p)                                                  \
> +               this_cpu_inc(p->FIELD);                         \
> +}                                                              \
> +EXPORT_SYMBOL(dev_core_stats_##FIELD##_inc)

Oh well, I just read the patch, and it seems wrong indeed.

netdev_core_stats_alloc() is the one that can be cold.