On 2013-05-23 02:37, Ermal Luçi wrote:
> On Thu, May 23, 2013 at 4:28 AM, Adam Thompson <[email protected] [1]> wrote: > >> I wonder, do you know of any heuristics to help judge (without doing extensive benchmarking, since I don't happen to have any Ixia gear handy...) when we're just as well off disabling all the offload functions? I wonder particularly about the code path to fix up tunnel packets so the igb/em/bce/bge can do their apparently-very-special cksum offload. > > I think it all comes down to latency. > If you have cpu contention and high load it would be beneficial to have the hardware(nic) do the checksum work for you anyway since the latency will be lower. So in the usual case, where my pfSense routers have excessively-powerful CPUs, and bge(4) NICs... are you saying I should expect lower latency with h/w offload turned ON or OFF? And, why does pfSense turn off h/w offloading by default? Is that only done automatically inside VMs? I know this has been discussed here previously, but I can't find the thread now. -Adam Links: ------ [1] mailto:[email protected]
_______________________________________________ List mailing list [email protected] http://lists.pfsense.org/mailman/listinfo/list
