Re: tuning for high connection rates

At 05:19 PM 12/5/2007, Philipp Wuensche wrote:

After switching to net.isr.direct=0 and 346609775 good packets later, RX
overruns haven't increased by one! Thats nice. Still interrupt is using
up the CPU. I'm not quite sure if polling would help now!?

Polling is helpful to prevent livelock. Not sure if thats happening to you. What firewall (if any) are you using ? pf used to be a lot slower than ipfw.

The Yandex driver is at
http://people.yandex-team.ru/~wawa/
but its against RELENG_6 only I think.


Another thing to try is to turn back on Fast Interrupt handling. I think its currently disabled.

In if_em.h, try adding

#define EM_FAST_IRQ 1

and then recompile the kernel or just driver.


We will try disabling TSO to see if anything changes.

If you have TCP in your app, it seems thats the thing to do according to the Intel developer.

---Mike

_______________________________________________
freebsd-performance@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-performance
To unsubscribe, send any mail to "freebsd-performance-unsubscribe@xxxxxxxxxxx"

Relevant Pages

  • Re: [PATCH] kexec: force x86_64 arches to boot kdump kernels on boot cpu
    ... Thats what I'm doing at the moment. ... timer interrupt wasn't getting received (even though we could successfully run ... unicast interrupts delivered accross the ht bus to a single cpu, ... that that broadcast bit was set. ...
    (Linux-Kernel)
  • Re: [PATCH] kexec: force x86_64 arches to boot kdump kernels on boot cpu
    ... Thats what I'm doing at the moment. ... timer interrupt wasn't getting received (even though we could successfully run ... for nvidia chipsets, which scanned for hypertransport controllers, and ensured ... that that broadcast bit was set. ...
    (Linux-Kernel)
  • Re: [PATCH] kexec: force x86_64 arches to boot kdump kernels on boot cpu
    ... Thats what I'm doing at the moment. ... So I was working on a patch to do early apic enabling on ... timer interrupt wasn't getting received (even though we could successfully run ... that that broadcast bit was set. ...
    (Linux-Kernel)
  • Re: tuning for high connection rates
    ... Polling is helpful to prevent livelock. ... Not sure if thats happening to ... Disabling it at all gives no noticable performance boost ...
    (freebsd-performance)
  • Re: further optimizations
    ... But maybe thats cause I dont understand this thing that much. ... interrupt a user application without swapping SS:ESP and of course ... use the users STACK to push EFL,CS,EIP and pop them with IRET. ... What I probably mixed up was the diffrence between "stack" and "Stackframe". ...
    (alt.lang.asm)