From mboxrd@z Thu Jan  1 00:00:00 1970
From: Changli Gao <xiaosuo@gmail.com>
Subject: Re: [RFC] rps: shortcut net_rps_action()
Date: Mon, 19 Apr 2010 20:28:27 +0800
Message-ID: <p2u412e6f7f1004190528l65cc6253w8b7f43f2657b551d@mail.gmail.com>
References: <t2p65634d661004141031xf80f62e7sb64362ea1ce10a1f@mail.gmail.com>
	<1271395106.16881.3645.camel@edumazet-laptop> <1271424065.4606.31.camel@bigi>
	<1271489739.16881.4586.camel@edumazet-laptop> <1271525519.3929.3.camel@bigi>
	<1271583573.16881.4798.camel@edumazet-laptop> <1271590476.16881.4925.camel@edumazet-laptop>
	<1271669822.16881.7520.camel@edumazet-laptop> <k2u412e6f7f1004190248s8ac633beof2475645e799f2ea@mail.gmail.com>
	<1271679244.3845.43.camel@edumazet-laptop>
Mime-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: Tom Herbert <therbert@google.com>,
	David Miller <davem@davemloft.net>,
	netdev <netdev@vger.kernel.org>
To: Eric Dumazet <eric.dumazet@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-iw0-f199.google.com ([209.85.223.199]:62177 "EHLO
	mail-iw0-f199.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
	with ESMTP id S1753208Ab0DSM2s convert rfc822-to-8bit (ORCPT
	<rfc822;netdev@vger.kernel.org>); Mon, 19 Apr 2010 08:28:48 -0400
Received: by iwn37 with SMTP id 37so398139iwn.15
        for <netdev@vger.kernel.org>; Mon, 19 Apr 2010 05:28:47 -0700 (PDT)
In-Reply-To: <1271679244.3845.43.camel@edumazet-laptop>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Mon, Apr 19, 2010 at 8:14 PM, Eric Dumazet <eric.dumazet@gmail.com> =
wrote:
>
> As several packets can be enqueued for a given cpu, we would need to
> keep bitmasks.
> We would have to add one test in enqueue_to_backlog()
>
> if (cpu_test_and_set(cpu, mask)) {
> =C2=A0 =C2=A0 =C2=A0 =C2=A0__raise_softirq_irqoff(NET_RX_SOFTIRQ);
> =C2=A0 =C2=A0 =C2=A0 =C2=A0array[nb++] =3D cpu;
> }

        rps_lock(queue);
        if (queue->input_pkt_queue.qlen <=3D netdev_max_backlog) {
                if (queue->input_pkt_queue.qlen) {
=2E..
                if (napi_schedule_prep(&queue->backlog)) {
#ifdef CONFIG_RPS
                        if (cpu !=3D smp_processor_id()) {
                                struct rps_remote_softirq_cpus *rcpus =3D
                                    &__get_cpu_var(rps_remote_softirq_c=
pus);

                                cpu_set(cpu, rcpus->mask[rcpus->select]=
);
                                __raise_softirq_irqoff(NET_RX_SOFTIRQ);
                                goto enqueue;
                        }
#endif
                        __napi_schedule(&queue->backlog);
                }

Only the first packet of a softnet.input_pkt_queue may trigger IPI, so
we don't need to keep bitmasks.

--=20
Regards=EF=BC=8C
Changli Gao(xiaosuo@gmail.com)