Holger Weiss wrote:
> * Matthias Scheler <tron@zhadum.org.uk> [2007-10-10 13:22]:
>> On Wed, Oct 10, 2007 at 05:06:07AM +0200, Holger Weiss wrote:
>>> * Chris Ross <cross+netbsd@distal.com> [2007-09-23 22:44]:
>>>>   I wanted to report that this interface works just fine on my sparc64.  At
>>>> least at 100Mbit, full-duplex, which is the switch I have it hooked to.
>>> FTR, sk(4) negotiates 1000Base-T (full-duplex) just fine on my Ultra 80,
>>> though I'm not able to actually pipe more than ~90 Mb/s through the NIC
>>> on that system.
>> That's because the CPUs are too slow. Even Solaris can't do that on an
>> UltraSPARC II system despite better platform support (including a better
>> compiler) and all the optimizations to the network stack in Solaris 10.
> 
> Yah, I didn't expect much more.
> 
>> The rule of thumb is that you need 1MHz per 1MBit for real workloads.
> 
> That would give me 450 MBit ;-)

Maybe I'm being dense here -- but...

An Ultra-80 has 450Mhz cpus, generally.  Did the sk(4) interface, when
negotiated at 1000Base-T full-duplex, really only achieve
90 megabits/second?

I have a Netra 1125 with 2x440Mhz cpus in it, and it easily gets to
550 megabits/second using an Intel gigabit ethernet card. (Solaris
e1000g driver, running Solaris 10 Update 3.)

I would think that 90 mbits/sec is way too low a number...

-Kurt