netdev - Re: mlx5 broken affinity

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <1d2e9304-089a-a769-9f38-a742dc066baf@grimberg.me>
Date:   Sun, 5 Nov 2017 10:36:00 +0200
From:   Sagi Grimberg <sagi@...mberg.me>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Jes Sorensen <jsorensen@...com>,
        Tariq Toukan <tariqt@...lanox.com>,
        Saeed Mahameed <saeedm@....mellanox.co.il>,
        Networking <netdev@...r.kernel.org>,
        Leon Romanovsky <leonro@...lanox.com>,
        Saeed Mahameed <saeedm@...lanox.com>,
        Kernel Team <kernel-team@...com>,
        Christoph Hellwig <hch@....de>
Subject: Re: mlx5 broken affinity


>>> 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.

I do agree that the user would lose better cpu online/offline behavior,
but it seems that users want to still have some control over the IRQ
affinity assignments even if they lose this functionality.

Would it be possible to keep the managed facility until a user overrides
an affinity assignment? This way if the user didn't touch it, we keep
all the perks, and in case the user overrides it, we log the implication
so the user is aware?