netdev - Re: Re: Synopsys Ethernet QoS

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]

Date:   Wed, 14 Dec 2016 12:54:24 +0100
From:   Pavel Machek <pavel@....cz>
To:     Niklas Cassel <niklas.cassel@...s.com>
Cc:     Joao Pinto <Joao.Pinto@...opsys.com>,
        Florian Fainelli <f.fainelli@...il.com>,
        Andy Shevchenko <andy.shevchenko@...il.com>,
        David Miller <davem@...emloft.net>,
        Giuseppe CAVALLARO <peppe.cavallaro@...com>, larper@...s.com,
        rabinv@...s.com, netdev <netdev@...r.kernel.org>,
        CARLOS.PALMINHA@...opsys.com, Jie.Deng1@...opsys.com,
        Stephen Warren <swarren@...dia.com>
Subject: Re: Re: Synopsys Ethernet QoS

Hi!

> There are some performance problems with the stmmac driver though:
> 
> When running iperf3 with 3 streams:
> iperf3 -c 192.168.0.90 -P 3 -t 30
> iperf3 -c 192.168.0.90 -P 3 -t 30 -R
> 
> I get really bad fairness between the streams.
> 
> This appears to be an issue with how TX IRQ coalescing is implemented in stmmac.
> Disabling TX IRQ coalescing in the stmmac driver makes the problem go away.

Yes, I'm hitting the same problem

https://lkml.org/lkml/2016/12/11/90

> Also netperf TCP_RR and UDP_RR gives really bad results compared to the
> dwceqos driver (without IRQ coalescing).
> 2000 transactions/sec vs 9000 transactions/sec.
> Turning TX IRQ coalescing off and RX interrupt watchdog off in stmmac
> gives the same performance. I guess it's a trade off, low CPU usage
> vs low latency, so I don't know how important TCP_RR/UDP_RR really is.

Well... 75% performance hit is a bug, plain and simple, not CPU tradeoff.

> The best thing would be to get a good working TX IRQ coalesce
> implementation with HR timers in stmmac.

Actually it seems that using HR timers is not the only problem, AFAICT
the logic is wrong way around. (But yes, it needs to be HR timer, too,
and probably packet count needs to be reduced, too.)

Best regards,

									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html

Download attachment "signature.asc" of type "application/pgp-signature" (182 bytes)