linux-kernel - Re:Re: [PATCH,RESEND] softirq: Introduce SOFTIRQ_FORCED

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] [day] [month] [year] [list]

Message-ID: <87wno5hdcp.ffs@tglx>
Date:   Sat, 28 Aug 2021 16:07:34 +0200
From:   Thomas Gleixner <tglx@...utronix.de>
To:     王擎 <wangqing@...o.com>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        Ingo Molnar <mingo@...nel.org>,
        Frederic Weisbecker <frederic@...nel.org>,
        Michal Hocko <mhocko@...nel.org>,
        Davidlohr Bueso <dave@...olabs.net>,
        Will Deacon <will@...nel.org>,
        "Paul E. McKenney" <paulmck@...nel.org>,
        Dirk Behme <dirk.behme@...bosch.com>,
        linux-kernel@...r.kernel.org
Subject: Re:Re: [PATCH,RESEND] softirq: Introduce SOFTIRQ_FORCED_THREADING

Qing,

On Sat, Aug 28 2021 at 10:18, 王擎 wrote:
>> On Mon, Aug 23 2021 at 11:33, Wang Qing wrote:
>> What you are proposing here is completly different as you enforce
>> softirq execution in context of ksoftirqd only.
>
> Thank you for reply and explanation, I just provide a choice to balance
> the execution of softirq according to their own business scenarios.

That's not a choice. Forced interrupt threading is a boot-time option
and not a compile time boolean. So with your change you even changed the
behaviour of the kernel when your magic config switch is not selected by
the user.

>> What are you referring to? PREEMPT_RT does not modify the priority of
>> ksoftirqd. If system designers want to do that, then they can do so from
>> user space. 
>
> I refer to the kernel-3.14 RT Patches. I used it at that time and achieved 
> very good results.

There is a reason why RT does not use this anymore and switched to a
different model. As I said before. Just because it works for you, it's
not necessarily a solution which should be exposed for general
consumption.

> I remember where I saw that softirqd was split into the original process 
> and the RT process. This can partially solve my problem.

Your patch has absolutely nothing to do with that. You just picked some
random part out of those 7+ years old patches and then claim that it's
something RT does, which is just not true.

Thanks,

        tglx