netdev - Re: [PATCH v3 net-next 2/4] net: Allow to use SMP threads for backlog NAPI.

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: <20240306081802.4MgVOkad@linutronix.de>
Date: Wed, 6 Mar 2024 09:18:02 +0100
From: Sebastian Andrzej Siewior <bigeasy@...utronix.de>
To: Jakub Kicinski <kuba@...nel.org>
Cc: Paolo Abeni <pabeni@...hat.com>, netdev@...r.kernel.org,
	"David S. Miller" <davem@...emloft.net>,
	Eric Dumazet <edumazet@...gle.com>,
	Jesper Dangaard Brouer <hawk@...nel.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Wander Lairson Costa <wander@...hat.com>,
	Yan Zhai <yan@...udflare.com>
Subject: Re: [PATCH v3 net-next 2/4] net: Allow to use SMP threads for
 backlog NAPI.

On 2024-03-05 07:23:34 [-0800], Jakub Kicinski wrote:
> On Tue, 5 Mar 2024 11:35:30 +0100 Sebastian Andrzej Siewior wrote:
> > I had RH benchmarking this and based on their 25Gbe and 50Gbe NICs and
> > the results look good. If anything it looked a bit better with this on
> > the 50Gbe NICs but since those NICs have RSS…
> 
> TBH if y'all tested this with iperf that's pretty meaningless.
> The concern is not as much throughput on an idle system as it 
> is the fact that we involve scheduler with it's heuristics
> for every NAPI run.

Yeah, I thought so. According to RH, they don't have a RPS workload left
since all their NICs use RSS now. That iperf test was the next best
thing. 

> But I recognize that your access to production workloads may 
> be limited and you did more than most, so 🤷️

Sebastian