Re: Multiport NICs - VLAN and Polling Support?

From: Danial Thom (danial_thom_at_yahoo.com)
Date: 10/14/05

  • Next message: Francisco Reyes: "Distributed authentication. Which one?" 
    Date: Fri, 14 Oct 2005 14:28:08 -0700 (PDT)
To: Kurt Jaeger <lists@complx.LF.net>

    --- Kurt Jaeger <lists@complx.LF.net> wrote:

    > Hi!
    >
    > > polling is almost never a performance
    > advantage
    > > for ethernet, as virtually all modern
    > controllers
    > > have some sort of interrupt moderation
    > built-in.
    >
    > We had the case on fbsd 5.4p7 with SMP where
    > bge drivers produced
    > much slower throughput if we did not use
    > polling. We measured using
    > ttcp and the difference was from 2-3Mbyte/sec
    > to 11 Mbyte/sec.
    >
    > Yes, it used more CPU, but for that increase in
    > throughput, it
    > was worth the expense.
    >
    > So it looks a bit strange from my part of the
    > world to.
    >
    > --
    > MfG/Best regards, Kurt Jaeger

    You have some other forces are work there
    obviously. There is no possible way that polling
    would yield a 5 fold increase in performance.
    You're still using the same cpu to do the tasks.
    11MG/s is such a small number of packets that it
    wouldn't generate enough interrrupts to make much
    difference to a modern processor. Thats what
    happens when you use benchmarks. Benchmarks are
    BS.

    Probably one of the many "features" that FreeBSD
    5.x offers. It likely it has more to do with
    FreeBSD 5s handling of interrupts in an SMP
    environment than anything having to do with
    polling.

    DT

    Danial

            
                    
    __________________________________
    Yahoo! Mail - PC Magazine Editors' Choice 2005
    http://mail.yahoo.com
    _______________________________________________
    freebsd-isp@freebsd.org mailing list
    http://lists.freebsd.org/mailman/listinfo/freebsd-isp
    To unsubscribe, send any mail to "freebsd-isp-unsubscribe@freebsd.org"

  • Next message: Francisco Reyes: "Distributed authentication. Which one?"

    Relevant Pages

    • Re: new version of polling for FreeBSD 6.x
      ... I am pretty sure you have removed all the debug OPTIONs from the kernel, ... The result on this test show that polling is a great benefit to network application vs interrupt. ... Yes it cost more in term of packet latency (where FreeBSD 4.11 was better than 6.x / 7.x in all mode) but under high pps with interrupt the DUT is unresponsive. ... For us polling have always worked better than interrupt under FreeBSD 4.x, under FreeBSD 6.x it is not the case and under one of my application benchmark you can see that it really have a problem to sustain the load. ...
      (freebsd-net)
    • Re: Polling, Interrupts, DMA, Synchronous, Asynchronous I/O Definitions
      ... > then it is not polling. ... > you would also have 100% CPU usage in this polling loop. ... polling can be *more* efficient than an interrupt system ...
      (alt.lang.asm)
    • RE: Polling For 100 mbps Connections? (Was Re: Freebsd Theme Song)
      ... > (Was Re: Freebsd Theme ... >>> Well, if polling does no good for fxp, due ... >>drop packets instead of going into livelock. ... >>> not interrupt more than this throttled rate ...
      (freebsd-questions)
    • Re: Polling, Interrupts, DMA, Synchronous, Asynchronous I/O Definitions
      ... > is no problem to do polling with this processor. ... > to notice when your friend arrives, ... > instead of always looking to the door and he should interrupt ...
      (alt.lang.asm)
    • RE: Polling For 100 mbps Connections? (Was Re: Freebsd Theme Song)
      ... >> polling is less ... It depends on the hardware. ... >has built-in hardware interrupt throttling (such ...
      (freebsd-questions)