Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 161CB217A4
MIME-Version: 1.0
In-Reply-To: <20180124015734.GA11302@lerouge>
References: <20180123.112201.1263563609292212852.davem@davemloft.net>
 <1516726652.2554.58.camel@redhat.com> <CA+55aFwEMe=jV7pvV2dWFKrQ--mAw3buXhdJ-Qk2u-290rTQQQ@mail.gmail.com>
 <20180123.132424.1035340800864767853.davem@davemloft.net> <20180124015734.GA11302@lerouge>
From:   Frederic Weisbecker <frederic@kernel.org>
Date:   Wed, 24 Jan 2018 03:01:37 +0100
Message-ID: <CAFTL4hziZkz2TPr4-YXYEumcgypxQHG4DTLs=Qjcsuh_tySHYA@mail.gmail.com>
Subject: Re: [RFC PATCH 0/4] softirq: Per vector threading v3
To:     David Miller <davem@davemloft.net>
Cc:     torvalds@linux-foundation.org, pabeni@redhat.com,
        linux-kernel@vger.kernel.org, alexander.levin@verizon.com,
        peterz@infradead.org, mchehab@s-opensource.com,
        hannes@stressinduktion.org, paulmck@linux.vnet.ibm.com,
        wanpeng.li@hotmail.com, dima@arista.com, tglx@linutronix.de,
        akpm@linux-foundation.org, rrendec@arista.com, mingo@kernel.org,
        sgruszka@redhat.com, riel@redhat.com, edumazet@google.com
Content-Type: text/plain; charset="UTF-8"
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

2018-01-24 2:57 UTC+01:00, Frederic Weisbecker <frederic@kernel.org>:
> On Tue, Jan 23, 2018 at 01:24:24PM -0500, David Miller wrote:
>> From: Linus Torvalds <torvalds@linux-foundation.org>
>> Date: Tue, 23 Jan 2018 09:42:32 -0800
>>
>> > But I wonder if the test triggers the "lets run lots of workqueue
>> > threads", and then the single-threaded user space just gets blown out
>> > of the water by many kernel threads. Each thread gets its own "fair"
>> > amount of CPU, but..
>>
>> If a single cpu's softirq deferral can end up running on multiple
>> workqueue threads, indeed that's a serious problem.
>>
>> So if we're in a workqueue and it does a:
>>
>> 	schedule_work_on(this_cpu, currently_executing_work);
>>
>> it'll potentially make a new thread?
>>
>> That's exactly the code path that will get exercised during a UDP
>> flood the way that vector_work_func() is implemented.
>
> It shouldn't create a new thread unless all other workers in the CPU
> are voluntarily sleeping while executing a work. Also since softirqs
> can't sleep, we shouldn't even have two vectors running concurrently
> on workqueues.
>

...unless I misunderstood workqueue internals or missed something, in
which case we may try ordered workqueue.