linux-kernel - Re:[PATCH v3 0/2] nvmet: use unbound

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: <20240726023407.70799-1-jacky_gam_2001@163.com>
Date: Fri, 26 Jul 2024 10:34:06 +0800
From: Ping Gan <jacky_gam_2001@....com>
To: sagi@...mberg.me,
	hare@...e.de,
	hch@....de
Cc: ping.gan@...l.com,
	kch@...dia.com,
	linux-nvme@...ts.infradead.org,
	linux-kernel@...r.kernel.org
Subject: Re:[PATCH v3 0/2] nvmet: use unbound_wq for RDMA and TCP by default

> On 19/07/2024 12:19, Ping Gan wrote:
>> When running nvmf on SMP platform, current nvme target's RDMA and
>> TCP use bounded workqueue to handle IO, but when there is other high
>> workload on the system(eg: kubernetes), the competition between the
>> bounded kworker and other workload is very radical. To decrease the
>> resource race of OS among them, this patchset will switch to
>> unbounded
>> workqueue for nvmet-rdma and nvmet-tcp; besides that, it can also
>> get some performance improvement. And this patchset bases on previous
>> discussion from below session.
>>
>> https://lore.kernel.org/lkml/20240719084953.8050-1-jacky_gam_2001@163.com/
>
> Hold your horses.
>
> This cannot be just switched without a thorough testing and actual 
> justification/proof of
> a benefit beyond just a narrow use-case brought initially by Ping Gan.
>
> If the ask is to universally use an unbound workqueue, please provide 
> detailed
> benchmarking convincing us that this makes sense.

So you think we should not do a radical change for the narrow usecase
but
keep the parameter to enable it in previous version patch, right?

Thanks,
Ping