From mboxrd@z Thu Jan 1 00:00:00 1970 From: John Fastabend Subject: Re: [PATCH] net: pktgen: support injecting packets for qdisc testing Date: Fri, 08 Jan 2016 13:55:13 -0800 Message-ID: <56903041.8090900@gmail.com> References: <20160108020908.5940.89612.stgit@john-Precision-Tower-5810> <568FD5B7.2060602@gmail.com> <568FD7FC.8020005@gmail.com> <1452288964.8255.234.camel@edumazet-glaptop2.roam.corp.google.com> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit Cc: netdev@vger.kernel.org, alexei.starovoitov@gmail.com, davem@davemloft.net, jhs@mojatatu.com To: Eric Dumazet Return-path: Received: from mail-pf0-f172.google.com ([209.85.192.172]:34609 "EHLO mail-pf0-f172.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752105AbcAHVzZ (ORCPT ); Fri, 8 Jan 2016 16:55:25 -0500 Received: by mail-pf0-f172.google.com with SMTP id q63so15550251pfb.1 for ; Fri, 08 Jan 2016 13:55:25 -0800 (PST) In-Reply-To: <1452288964.8255.234.camel@edumazet-glaptop2.roam.corp.google.com> Sender: netdev-owner@vger.kernel.org List-ID: On 16-01-08 01:36 PM, Eric Dumazet wrote: > On Fri, 2016-01-08 at 07:38 -0800, John Fastabend wrote: > >> Also I'll respin this with a change so xmit_{drop|cn|policed} are not >> counted as tx_bytes and pkts sent (idea from Jamal). However I won't >> abort the burst because I do like this stressing the entry point of my >> qdisc. > > This reminds me that bonding/team drivers always return NETDEV_TX_OK > status, even if the chosen slave could not queue or send the packet. > > > hmm this could potentially break protocols that are handling the return codes like fcoe which puts a packet back on its queue if it can't be sent. .John