From mboxrd@z Thu Jan 1 00:00:00 1970 From: Eric Dumazet Subject: Re: [PATCH net-next 2/2] bnx2x: use the default NAPI weight Date: Mon, 11 Mar 2013 07:14:57 +0100 Message-ID: <1362982497.4051.37.camel@edumazet-glaptop> References: <1362540509.15793.158.camel@edumazet-glaptop> <20130305.233744.498353801662910535.davem@davemloft.net> <1362553398.15793.168.camel@edumazet-glaptop> <20130306.145947.1445904594058156164.davem@davemloft.net> <1362606777.15793.198.camel@edumazet-glaptop> <20130310083800.GC10448@gondor.apana.org.au> Mime-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: 7bit Cc: David Miller , netdev@vger.kernel.org, eilong@broadcom.com, jhs@mojatatu.com, Tom Herbert To: Herbert Xu Return-path: Received: from mail-ea0-f181.google.com ([209.85.215.181]:34030 "EHLO mail-ea0-f181.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751900Ab3CKGPD (ORCPT ); Mon, 11 Mar 2013 02:15:03 -0400 Received: by mail-ea0-f181.google.com with SMTP id z10so994943ead.40 for ; Sun, 10 Mar 2013 23:15:01 -0700 (PDT) In-Reply-To: <20130310083800.GC10448@gondor.apana.org.au> Sender: netdev-owner@vger.kernel.org List-ID: On Sun, 2013-03-10 at 16:38 +0800, Herbert Xu wrote: > On Wed, Mar 06, 2013 at 01:52:57PM -0800, Eric Dumazet wrote: > > > > - BQL (incurring more TX completion rounds and possibility to > > block/unblock a qdisc) > > - ticket spinlocks, and even with the guard of qdisc busylock > > > > -> we can have a starvation problem. > > This only happens in cases where we aren't using multiqueue or > we're using it incorrectly, resulting in TX work from being split > over CPUs. > Or using qdisc like HTB ;) > In that case it's not clear that it is starvation if we keep the > TX processing on one CPU. Thats not always the case. TX path has interesting features like XPS ... We probably could add instrumentation and keep track of the maximum time we are spending in this dark area.