[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <Z0+mseLf3mn9mFxz@pop-os.localdomain>
Date: Tue, 3 Dec 2024 16:47:45 -0800
From: Cong Wang <xiyou.wangcong@...il.com>
To: Eric Dumazet <edumazet@...gle.com>
Cc: "David S . Miller" <davem@...emloft.net>,
Jakub Kicinski <kuba@...nel.org>, Paolo Abeni <pabeni@...hat.com>,
Jamal Hadi Salim <jhs@...atatu.com>,
Victor Nogueira <victor@...atatu.com>,
Jiri Pirko <jiri@...nulli.us>, netdev@...r.kernel.org,
eric.dumazet@...il.com
Subject: Re: [PATCH net-next] net_sched: sch_fq: add three drop_reason
On Tue, Dec 03, 2024 at 09:09:29PM +0000, Eric Dumazet wrote:
> Add three new drop_reason, more precise than generic QDISC_DROP:
>
> "tc -s qd" show aggregate counters, it might be more useful
> to use drop_reason infrastructure for bug hunting.
>
> 1) SKB_DROP_REASON_FQ_DROP_BAND_LIMIT
> Whenever a packet is added while its band limit is hit.
> Corresponding value in "tc -s qd" is bandX_drops XXXX
>
> 2) SKB_DROP_REASON_FQ_DROP_HORIZON_LIMIT
> Whenever a packet has a timestamp too far in the future.
> Corresponding value in "tc -s qd" is horizon_drops XXXX
>
> 3) SKB_DROP_REASON_FQ_DROP_FLOW_LIMIT
> Whenever a flow has reached its limit.
> Corresponding value in "tc -s qd" is flows_plimit XXXX
>
Just a nit: maybe remove the second "DROP" in these long names?
Reviewed-by: Cong Wang <cong.wang@...edance.com>
Thanks.
Powered by blists - more mailing lists