From mboxrd@z Thu Jan  1 00:00:00 1970
From: Tom Herbert <tom@herbertland.com>
Subject: Re: [net-next PATCH v2 2/4] net: Enable Tx queue selection based on
 Rx queues
Date: Mon, 21 May 2018 07:51:27 -0700
Message-ID: <CALx6S36h=gGb1LkLuJ80DUrE=m+FhbcQ0AD94AdtEUvxJfHf=g@mail.gmail.com>
References: <152643356116.4991.7215767041139726872.stgit@anamdev.jf.intel.com>
 <152643400925.4991.5029989601625953592.stgit@anamdev.jf.intel.com>
 <CALx6S3652mgSjpQtF+9N90CYYhvyaMjS94zbgyhcFAa_G0h4tQ@mail.gmail.com>
 <CAF=yD-JXpiJwxM_mHvAgJ6qhsgq4uOZYbsMBVvcOmZawbueayQ@mail.gmail.com> <CAF=yD-JghZY5NN6cHGdHeOTs8xb9KF=mQ=J2P49ojrvp+MsD8w@mail.gmail.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="UTF-8"
Cc: Amritha Nambiar <amritha.nambiar@intel.com>,
        Linux Kernel Network Developers <netdev@vger.kernel.org>,
        "David S. Miller" <davem@davemloft.net>,
        Alexander Duyck <alexander.h.duyck@intel.com>,
        Sridhar Samudrala <sridhar.samudrala@intel.com>,
        Eric Dumazet <edumazet@google.com>,
        Hannes Frederic Sowa <hannes@stressinduktion.org>
To: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from mail-qk0-f194.google.com ([209.85.220.194]:42110 "EHLO
        mail-qk0-f194.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751504AbeEUOv3 (ORCPT
        <rfc822;netdev@vger.kernel.org>); Mon, 21 May 2018 10:51:29 -0400
Received: by mail-qk0-f194.google.com with SMTP id b22-v6so11972961qkj.9
        for <netdev@vger.kernel.org>; Mon, 21 May 2018 07:51:29 -0700 (PDT)
In-Reply-To: <CAF=yD-JghZY5NN6cHGdHeOTs8xb9KF=mQ=J2P49ojrvp+MsD8w@mail.gmail.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Sat, May 19, 2018 at 1:27 PM, Willem de Bruijn
<willemdebruijn.kernel@gmail.com> wrote:
> On Sat, May 19, 2018 at 4:13 PM, Willem de Bruijn
> <willemdebruijn.kernel@gmail.com> wrote:
>> On Fri, May 18, 2018 at 12:03 AM, Tom Herbert <tom@herbertland.com> wrote:
>>> On Tue, May 15, 2018 at 6:26 PM, Amritha Nambiar
>>> <amritha.nambiar@intel.com> wrote:
>>>> This patch adds support to pick Tx queue based on the Rx queue map
>>>> configuration set by the admin through the sysfs attribute
>>>> for each Tx queue. If the user configuration for receive
>>>> queue map does not apply, then the Tx queue selection falls back
>>>> to CPU map based selection and finally to hashing.
>>>>
>>>> Signed-off-by: Amritha Nambiar <amritha.nambiar@intel.com>
>>>> Signed-off-by: Sridhar Samudrala <sridhar.samudrala@intel.com>
>>>> ---
>
>>>> +static int get_xps_queue(struct net_device *dev, struct sk_buff *skb)
>>>> +{
>>>> +#ifdef CONFIG_XPS
>>>> +       enum xps_map_type i = XPS_MAP_RXQS;
>>>> +       struct xps_dev_maps *dev_maps;
>>>> +       struct sock *sk = skb->sk;
>>>> +       int queue_index = -1;
>>>> +       unsigned int tci = 0;
>>>> +
>>>> +       if (sk && sk->sk_rx_queue_mapping <= dev->real_num_rx_queues &&
>>>> +           dev->ifindex == sk->sk_rx_ifindex)
>>>> +               tci = sk->sk_rx_queue_mapping;
>>>> +
>>>> +       rcu_read_lock();
>>>> +       while (queue_index < 0 && i < __XPS_MAP_MAX) {
>>>> +               if (i == XPS_MAP_CPUS)
>>>
>>> This while loop typifies exactly why I don't think the XPS maps should
>>> be an array.
>>
>> +1
>
> as a matter of fact, as enabling both cpu and rxqueue map at the same
> time makes no sense, only one map is needed at any one time. The
> only difference is in how it is indexed. It should probably not be possible
> to configure both at the same time. Keeping a single map probably also
> significantly simplifies patch 1/4.

Willem,

I think it might makes sense to have them both. Maybe one application
is spin polling that needs this, where others might be happy with
normal CPU mappings as default.

Tom