Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.36 as permitted sender) client-ip=23.128.96.36;
MIME-Version: 1.0
Date:   Tue, 12 Sep 2023 01:42:29 +0000
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From:   "Yajun Deng" <yajun.deng@linux.dev>
Message-ID: <5aa174ef5cd662595ffcefd3545ff07ed3063d6c@linux.dev>
TLS-Required: No
Subject: Re: [PATCH] net/core: Export dev_core_stats_rx_dropped_inc sets
To:     "Stephen Hemminger" <stephen@networkplumber.org>
Cc:     davem@davemloft.net, edumazet@google.com, kuba@kernel.org,
        pabeni@redhat.com, horms@kernel.org, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org
In-Reply-To: <20230911091513.4f2f2aed@fedora>
References: <20230911091513.4f2f2aed@fedora>
 <20230911082016.3694700-1-yajun.deng@linux.dev>
Precedence: bulk

September 12, 2023 at 12:15 AM, "Stephen Hemminger" <stephen@networkplumb=
er.org> wrote:


>=20
>=20On Mon, 11 Sep 2023 16:20:16 +0800
> Yajun Deng <yajun.deng@linux.dev> wrote:
>=20
>=20>=20
>=20> Although there is a kfree_skb_reason() helper function that can be
> >  used to find the reason for dropped packets, but most callers didn't
> >  increase one of rx_dropped, tx_dropped, rx_nohandler and
> >  rx_otherhost_dropped.
> >=20=20
>=20>  For the users, people are more concerned about why the dropped in
> >  ifconfig is increasing. So we can export
> >  dev_core_stats_rx_dropped_inc sets, which users would trace them kno=
w
> >  why rx_dropped is increasing.
> >=20
>=20
> ifconfig has been frozen for over 10 years, and is deprecated so there
> is no point in catering to legacy api's. There are better API's such as
> ethtool and netlink that can provide more info.
>
Yes, ifconfig is deprecated. but the dropped in ifconfig and ip is the sa=
me.
We're more concerned about the reason for dropped packets.
ip, ethtool and netlink couldn't show the reason.