Re: Much improved sendfile(2) kernel implementation


On Sat, 23 Sep 2006, Andre Oppermann wrote:

Without patch:
87380 393216 393216 10.00 2163.08 100.00 19.35 3.787 1.466 Without patch + TSO:
87380 393216 393216 10.00 4367.18 71.54 42.07 1.342 1.578 With patch:
87380 393216 393216 10.01 1882.73 86.15 18.43 3.749 1.604 With patch + TSO:
87380 393216 393216 10.00 6961.08 47.69 60.11 0.561 1.415

The impact of TSO is clearly dramatic, especially when combined with the patch, but I'm a bit concerned by the drop in performance in the patched non-TSO case. For network cards which will always have TSO enabled, this isn't an issue, but do we see a similar affect for drivers without TSO? What can we put this drop down to?

Robert N M Watson
Computer Laboratory
University of Cambridge
_______________________________________________
freebsd-current@xxxxxxxxxxx mailing list
http://lists.freebsd.org/mailman/listinfo/freebsd-current
To unsubscribe, send any mail to "freebsd-current-unsubscribe@xxxxxxxxxxx"

Relevant Pages

  • Re: RFC: TSO patch for current
    ... perhaps they lifted it from your patch then. ... >> similiar that's why I thought it came from NetBSD. ... > The driver related changes can follow.. ... > Later we also need to fix ifconfig so that user can enable/disable TSO on the interface. ...
    (freebsd-current)
  • Re: RFC: TSO patch for current
    ... perhaps they lifted it from your patch then. ... >> similiar that's why I thought it came from NetBSD. ... > The driver related changes can follow.. ... > Later we also need to fix ifconfig so that user can enable/disable TSO on the interface. ...
    (freebsd-net)
  • Re: re(4) problem
    ... The attached patch didn't help, the box is again in the same status. ... get packet loss even when I ping it from another server in the same lan. ... It seems that you've already disabled TSO on reinterface. ... Disabling system-wise TSO wouldn't affect re's TSO operation ...
    (freebsd-current)
  • Re: [PATCH 0/4] atl1: Revised Attansic L1 ethernet driver
    ... I did the the atl1-2.0.2 patch against the 2.6.19.rc6 kernel. ... I also tried turninig off TSO as adviced, but i got the message "operation ... I'm attaching what I'm using here. ... perche` vuol dire che ti ...
    (Linux-Kernel)
  • Re: Much improved sendfile(2) kernel implementation
    ... On Sat, 23 Sep 2006, Andre Oppermann wrote: ... Without patch + TSO: ... University of Cambridge ...
    (freebsd-net)