From mboxrd@z Thu Jan  1 00:00:00 1970
From: Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [RFD] Managed interrupt affinities [ Was: mlx5 broken affinity
 ]
Date: Fri, 10 Nov 2017 14:03:08 +0100 (CET)
Message-ID: <alpine.DEB.2.20.1711101359290.1923@nanos>
References: <2187e555-2c4e-ef55-1c3a-17f5af54d762@fb.com>  <d65d6156-07da-14dc-88ff-fcb5c0ef5f83@grimberg.me>  <f8624a01-f77a-ac0b-ba19-19c7150ca133@fb.com>  <CALzJLG-Z2fJibyZ74XU7XCSx-g3=YapV9VyekAYZkS5JwvkwWw@mail.gmail.com>  <573060a9-19a7-9133-ef52-fa947088dabb@fb.com>
  <3af0c164-8dde-b6f0-45e1-edbbb28e7f73@mellanox.com>  <83d3944f-8a31-eb31-93db-294906630b0e@grimberg.me>  <556f3ff5-c1d4-25c6-7bfc-9866c0d9b323@fb.com>  <9384acdc-a5d8-872c-0034-9a3869f4fc8b@grimberg.me>  <alpine.DEB.2.20.1711021907420.2824@nanos>
 <1d2e9304-089a-a769-9f38-a742dc066baf@grimberg.me>  <alpine.DEB.2.20.1711071551330.1716@nanos>  <67a1157e-06e7-479c-993e-bdf42fd613c6@fb.com>  <alpine.DEB.2.20.1711081822450.1937@nanos>  <69a46009-184f-d925-289c-6036f0bf2554@grimberg.me>
 <alpine.DEB.2.20.1711091429000.1839@nanos>  <7f7aab77-b0b6-f8fa-0b57-3e3c1755eeaa@fb.com>  <alpine.DEB.2.20.1711091759350.1839@nanos>  <4888ac89-22d5-0700-daa1-d604e4c54970@fb.com>  <alpine.DEB.2.20.1711092226080.2690@nanos>
 <1510293390.5405.9.camel@mellanox.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Cc: Jens Axboe <axboe@fb.com>, Sagi Grimberg <sagi@grimberg.me>,
        Jes Sorensen <jsorensen@fb.com>,
        Tariq Toukan <tariqt@mellanox.com>,
        Saeed Mahameed <saeedm@dev.mellanox.co.il>,
        Networking <netdev@vger.kernel.org>,
        Leon Romanovsky <leonro@mellanox.com>,
        Kernel Team <kernel-team@fb.com>,
        Christoph Hellwig <hch@lst.de>
To: Saeed Mahameed <saeedm@mellanox.com>
Return-path: <netdev-owner@vger.kernel.org>
Received: from Galois.linutronix.de ([146.0.238.70]:55208 "EHLO
        Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S1751142AbdKJNDM (ORCPT
        <rfc822;netdev@vger.kernel.org>); Fri, 10 Nov 2017 08:03:12 -0500
In-Reply-To: <1510293390.5405.9.camel@mellanox.com>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Fri, 10 Nov 2017, Saeed Mahameed wrote:

> Well, I can speak for mlx5 case or most of the network drivers, where
> all of the queues associated with an interrupt, move with it, so i
> don't think our current driver have this issue. I don't believe there
> are network driver with fixed Per cpu resources, but it worth double
> checking.

At least the intel i40e[vf] drivers use the affinity notifier. Same for
infiniband drivers and scsi. So there seems to be requirements.

> Regarding the below solutions, any one that will gurantee the initial
> managed spreading and still allow the user to modify affinity via
> /proc/irq/xyz/smp_afinity will be acceptable, since many tools and user
> rely on this sysfs entry e.g. (irqbalance)

Well, any one is not very specific.

This needs to be discussed and agreed on in detail, otherwise we end up
with the same situation 3 month after the changes hit mainline and drivers
got converted over.

Thanks,

	tglx