From mboxrd@z Thu Jan 1 00:00:00 1970 From: Marcelo Ricardo Leitner Subject: Re: [PATCH] net: sched: tbf: handle GSO_BY_FRAGS case in enqueue Date: Thu, 15 Feb 2018 11:29:00 -0200 Message-ID: <20180215132900.GB4716@localhost.localdomain> References: <20180205033433.23647-1-dja@axtens.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: netdev@vger.kernel.org To: Daniel Axtens Return-path: Received: from mail-qk0-f195.google.com ([209.85.220.195]:35216 "EHLO mail-qk0-f195.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1032301AbeBON3G (ORCPT ); Thu, 15 Feb 2018 08:29:06 -0500 Received: by mail-qk0-f195.google.com with SMTP id c4so28572762qkm.2 for ; Thu, 15 Feb 2018 05:29:05 -0800 (PST) Content-Disposition: inline In-Reply-To: <20180205033433.23647-1-dja@axtens.net> Sender: netdev-owner@vger.kernel.org List-ID: On Mon, Feb 05, 2018 at 02:34:33PM +1100, Daniel Axtens wrote: > tbf_enqueue() checks the size of a packet before enqueuing it. > However, the GSO size check does not consider the GSO_BY_FRAGS > case, and so will drop GSO SCTP packets, causing a massive drop > in throughput. > > Use skb_gso_validate_mac_len() instead, as it does consider that > case. > > --- > > skb_gso_validate_mac_len() is an out-of-line call, but so is > skb_gso_mac_seglen(), so this is slower but not much slower. I > will send a patch to make the skb_gso_validate_* functions > inline-able shortly. > > Also, GSO_BY_FRAGS considered harmful - I'm pretty sure this is > not the only place it causes issues. Thanks for spotting these issues and fixing them. > > Signed-off-by: Daniel Axtens If this block was meant to be an out-of-band/changelog comment, your SOB line should be above the first --- marker. Anyhow, Reviewed-by: Marcelo Ricardo Leitner > --- > net/sched/sch_tbf.c | 3 ++- > 1 file changed, 2 insertions(+), 1 deletion(-) > > diff --git a/net/sched/sch_tbf.c b/net/sched/sch_tbf.c > index 229172d509cc..03225a8df973 100644 > --- a/net/sched/sch_tbf.c > +++ b/net/sched/sch_tbf.c > @@ -188,7 +188,8 @@ static int tbf_enqueue(struct sk_buff *skb, struct Qdisc *sch, > int ret; > > if (qdisc_pkt_len(skb) > q->max_size) { > - if (skb_is_gso(skb) && skb_gso_mac_seglen(skb) <= q->max_size) > + if (skb_is_gso(skb) && > + skb_gso_validate_mac_len(skb, q->max_size)) > return tbf_segment(skb, sch, to_free); > return qdisc_drop(skb, sch, to_free); > } > -- > 2.14.1 >