linux-kernel - Re: [RFC PATCH 0/4] softirq: Per vector threading v3

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: <20180123.112201.1263563609292212852.davem@davemloft.net>
Date:   Tue, 23 Jan 2018 11:22:01 -0500 (EST)
From:   David Miller <davem@...emloft.net>
To:     pabeni@...hat.com
Cc:     frederic@...nel.org, linux-kernel@...r.kernel.org,
        alexander.levin@...izon.com, peterz@...radead.org,
        mchehab@...pensource.com, torvalds@...ux-foundation.org,
        hannes@...essinduktion.org, paulmck@...ux.vnet.ibm.com,
        wanpeng.li@...mail.com, dima@...sta.com, tglx@...utronix.de,
        akpm@...ux-foundation.org, rrendec@...sta.com, mingo@...nel.org,
        sgruszka@...hat.com, riel@...hat.com, edumazet@...gle.com
Subject: Re: [RFC PATCH 0/4] softirq: Per vector threading v3

From: Paolo Abeni <pabeni@...hat.com>
Date: Tue, 23 Jan 2018 11:13:52 +0100

> Hi,
> 
> On Fri, 2018-01-19 at 16:46 +0100, Frederic Weisbecker wrote:
>> As per Linus suggestion, this take doesn't limit the number of occurences
>> per jiffy anymore but instead defers a vector to workqueues as soon as
>> it gets re-enqueued on IRQ tail.
>> 
>> No tunable here, so testing should be easier.
>> 
>> git://git.kernel.org/pub/scm/linux/kernel/git/frederic/linux-dynticks.git
>> 	softirq/thread-v3
>> 
>> HEAD: 6835e92cbd70ef4a056987d2e1ed383b294429d4
> 
> I tested this series in the UDP flood scenario, binding the user space
> process receiving the packets on the same CPU processing the related
> IRQ, and the tput sinks nearly to 0, like before Eric's patch.
> 
> The perf tool says that almost all the softirq processing is done
> inside the workqueue, but the user space process is scheduled very
> rarely, while before this series, in this scenario, ksoftirqd and the
> user space process got a fair share of the CPU time.

Do workqueue threads get a higher scheduling priority than user
processes?  If so, that's going to work against the entire point of
deferring softirq work into a thread.

Or is it that the workqueue execution is simply not yielding for some
reason?