From mboxrd@z Thu Jan  1 00:00:00 1970
From: Thomas Gleixner <tglx@linutronix.de>
Subject: Re: mlx5 broken affinity
Date: Thu, 2 Nov 2017 19:10:00 +0100 (CET)
Message-ID: <alpine.DEB.2.20.1711021907420.2824@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>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Cc: 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>,
        Saeed Mahameed <saeedm@mellanox.com>,
        Kernel Team <kernel-team@fb.com>,
        Christoph Hellwig <hch@lst.de>
To: Sagi Grimberg <sagi@grimberg.me>
Return-path: <netdev-owner@vger.kernel.org>
Received: from Galois.linutronix.de ([146.0.238.70]:55762 "EHLO
        Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org
        with ESMTP id S933025AbdKBSKF (ORCPT
        <rfc822;netdev@vger.kernel.org>); Thu, 2 Nov 2017 14:10:05 -0400
In-Reply-To: <9384acdc-a5d8-872c-0034-9a3869f4fc8b@grimberg.me>
Sender: netdev-owner@vger.kernel.org
List-ID: <netdev.vger.kernel.org>

On Thu, 2 Nov 2017, Sagi Grimberg wrote:

> 
> > This wasn't to start a debate about which allocation method is the
> > perfect solution. I am perfectly happy with the new default, the part
> > that is broken is to take away the user's option to reassign the
> > affinity. That is a bug and it needs to be fixed!
> 
> Well,
> 
> I would really want to wait for Thomas/Christoph to reply, but this
> simple change fixed it for me:
> --
> diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
> index 573dc52b0806..eccd06be5e44 100644
> --- a/kernel/irq/manage.c
> +++ b/kernel/irq/manage.c
> @@ -146,8 +146,7 @@ bool irq_can_set_affinity_usr(unsigned int irq)
>  {
>         struct irq_desc *desc = irq_to_desc(irq);
> 
> -       return __irq_can_set_affinity(desc) &&
> -               !irqd_affinity_is_managed(&desc->irq_data);
> +       return __irq_can_set_affinity(desc);

Which defeats the whole purpose of the managed facility, which is _not_ to
break the affinities on cpu offline and bring the interrupt back on the CPU
when it comes online again.

What I can do is to have a separate flag, which only uses the initial
distribution mechanism, but I really want to have Christophs opinion on
that.

Thanks,

	tglx