Date: Sat, 11 Feb 2006 00:54:20 +0000 (GMT) From: Robert Watson <rwatson@FreeBSD.org> To: Poul-Henning Kamp <phk@phk.freebsd.dk> Cc: Andrew Gallatin <gallatin@cs.duke.edu>, current@freebsd.org Subject: Re: [TEST/REVIEW] cpu time accounting patch, step 2 Message-ID: <20060211005336.F90460@fledge.watson.org> In-Reply-To: <5383.1139586916@critter.freebsd.dk> References: <5383.1139586916@critter.freebsd.dk>
next in thread | previous in thread | raw e-mail | index | archive | help
On Fri, 10 Feb 2006, Poul-Henning Kamp wrote: > In message <17388.44976.250463.383429@grasshopper.cs.duke.edu>, Andrew Gallatin > writes: > >> Here are some benchmark results from my "AMD Athlon(tm) 64 X2 Dual Core >> Processor 3800+" running an SMP kernel for both loopback and 10GbE TCP >> networks. The executive summary is that there's a 41% improvement in >> loopback pingpong, and a 12.5% improvement in a pingpong test with a remote >> linux machine over 10GbE. Since the linux machine isn't infinately fast, >> the improvement is probably a bit more than 12.5% ;) > > That's a LOT more than I had hoped for... This is due to our currently rather severe problems with preemption of the sending thread by the netisr on netisr wakeup for loopback transmit. Still cogitating on the right model change for that, but something will go into the tree before 7.0. Robert N M Watson
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20060211005336.F90460>