From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jarek Poplawski Subject: Re: [RFC] multiqueue changes Date: Fri, 9 Oct 2009 08:51:07 +0000 Message-ID: <20091009085107.GA7711@ff.dom.local> References: <4ACD9255.4020008@gmail.com> <20091008090344.GA7409@ff.dom.local> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "David S. Miller" , Patrick McHardy , Linux Netdev List To: Eric Dumazet Return-path: Received: from mail-fx0-f227.google.com ([209.85.220.227]:59743 "EHLO mail-fx0-f227.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755161AbZJIIvu (ORCPT ); Fri, 9 Oct 2009 04:51:50 -0400 Received: by fxm27 with SMTP id 27so6208711fxm.17 for ; Fri, 09 Oct 2009 01:51:13 -0700 (PDT) Content-Disposition: inline In-Reply-To: <20091008090344.GA7409@ff.dom.local> Sender: netdev-owner@vger.kernel.org List-ID: On Thu, Oct 08, 2009 at 09:03:44AM +0000, Jarek Poplawski wrote: > On Thu, Oct 08, 2009 at 09:18:45AM +0200, Eric Dumazet wrote: ... > > Just wondering if it could hurt some people. > > > > Anyway, should we change bnx2/tg3 drivers so that single queue devices > > have same default qdisc/class than before ? > > > > eg : > > > > diff --git a/drivers/net/bnx2.c b/drivers/net/bnx2.c > > index 08cddb6..7cac205 100644 > > --- a/drivers/net/bnx2.c > > +++ b/drivers/net/bnx2.c > > @@ -6152,6 +6152,7 @@ bnx2_setup_int_mode(struct bnx2 *bp, int dis_msi) > > > > bp->num_tx_rings = rounddown_pow_of_two(bp->irq_nvecs); > > bp->dev->real_num_tx_queues = bp->num_tx_rings; > > + bp->dev->num_tx_queues = bp->dev->real_num_tx_queues; > > > > bp->num_rx_rings = bp->irq_nvecs; > > } > > It doesn't look consistent to me wrt. the comment in netdevice.h on > num_tx_queues. But it seems we should rather use more often > real_num_tx_queue in schedulers code like dumps and maybe more > (like e.g. sch_multiq does). So, according to my current understanding, we should probably let drivers to reset the tx_queues allocation with a new num_tx_queues in a place like this (ndo_open). Jarek P.