MIME-Version: 1.0
References: <20230314131427.85135-1-kerneljasonxing@gmail.com> <20230314131427.85135-2-kerneljasonxing@gmail.com>
In-Reply-To: <20230314131427.85135-2-kerneljasonxing@gmail.com>
From:   Eric Dumazet <edumazet@google.com>
Date:   Tue, 14 Mar 2023 08:15:19 -0700
Message-ID: <CANn89iJzVjht5L1zxwCMTPXXoXdRMtRmzbL5UzHodhBJziCxYg@mail.gmail.com>
Subject: Re: [PATCH v3 net-next 1/2] net-sysfs: display two backlog queue len separately
To:     Jason Xing <kerneljasonxing@gmail.com>
Cc:     davem@davemloft.net, kuba@kernel.org, pabeni@redhat.com,
        ast@kernel.org, daniel@iogearbox.net, hawk@kernel.org,
        john.fastabend@gmail.com, stephen@networkplumber.org,
        simon.horman@corigine.com, sinquersw@gmail.com,
        bpf@vger.kernel.org, netdev@vger.kernel.org,
        linux-kernel@vger.kernel.org, Jason Xing <kernelxing@tencent.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
Precedence: bulk

On Tue, Mar 14, 2023 at 6:14=E2=80=AFAM Jason Xing <kerneljasonxing@gmail.c=
om> wrote:
>
> From: Jason Xing <kernelxing@tencent.com>
>
> Sometimes we need to know which one of backlog queue can be exactly
> long enough to cause some latency when debugging this part is needed.
> Thus, we can then separate the display of both.
>
> Signed-off-by: Jason Xing <kernelxing@tencent.com>
> Reviewed-by: Simon Horman <simon.horman@corigine.com>
> ---
> v3: drop the comment suggested by Simon
> Link: https://lore.kernel.org/lkml/20230314030532.9238-2-kerneljasonxing@=
gmail.com/
>
> v2: keep the total len of backlog queues untouched as Eric said
> Link: https://lore.kernel.org/lkml/20230311151756.83302-1-kerneljasonxing=
@gmail.com/
> ---
>  net/core/net-procfs.c | 19 +++++++++++++++----
>  1 file changed, 15 insertions(+), 4 deletions(-)
>
> diff --git a/net/core/net-procfs.c b/net/core/net-procfs.c
> index 1ec23bf8b05c..8056f39da8a1 100644
> --- a/net/core/net-procfs.c
> +++ b/net/core/net-procfs.c
> @@ -115,10 +115,19 @@ static int dev_seq_show(struct seq_file *seq, void =
*v)
>         return 0;
>  }
>
> +static u32 softnet_input_pkt_queue_len(struct softnet_data *sd)
> +{
> +       return skb_queue_len_lockless(&sd->input_pkt_queue);
> +}
> +
> +static u32 softnet_process_queue_len(struct softnet_data *sd)
> +{
> +       return skb_queue_len_lockless(&sd->process_queue);
> +}
> +
>  static u32 softnet_backlog_len(struct softnet_data *sd)
>  {
> -       return skb_queue_len_lockless(&sd->input_pkt_queue) +
> -              skb_queue_len_lockless(&sd->process_queue);
> +       return softnet_input_pkt_queue_len(sd) + softnet_process_queue_le=
n(sd);
>  }
>
>  static struct softnet_data *softnet_get_online(loff_t *pos)
> @@ -169,12 +178,14 @@ static int softnet_seq_show(struct seq_file *seq, v=
oid *v)
>          * mapping the data a specific CPU
>          */
>         seq_printf(seq,
> -                  "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08=
x %08x %08x\n",
> +                  "%08x %08x %08x %08x %08x %08x %08x %08x %08x %08x %08=
x %08x %08x "
> +                  "%08x %08x\n",
>                    sd->processed, sd->dropped, sd->time_squeeze, 0,
>                    0, 0, 0, 0, /* was fastroute */
>                    0,   /* was cpu_collision */
>                    sd->received_rps, flow_limit_count,
> -                  softnet_backlog_len(sd), (int)seq->index);
> +                  softnet_backlog_len(sd), (int)seq->index,
> +                  softnet_input_pkt_queue_len(sd), softnet_process_queue=
_len(sd));
>         return 0;


It is customary to wait ~24 hours between each version, so that
everybody gets a chance to comment,
and to avoid polluting mailing lists with too many messages/day.

(I see you are including lkml@, which seems unnecessary for this kind of pa=
tch)

Please address the feedback I gave for v2.

Thanks.