[EMAIL PROTECTED] wrote:
>
> > > I suspect a generic chipset fault, or some design quirk that we are not
> > > working around. Note that the windoze drivers for these devices put them
> > > permanently in store-and-forward mode. if_de has the exact same problem on
> > > all of the systems above.
> ...
> > Store and forward mode introduces a horrible performance hit... Artesyn
> > wouldn't show us the source to their workaround :-(
>
> It should be noted that I was able to saturate a 100 Mbps Ethernet with
> FreeBSD 2.2 and a 21140 based card, using around 56% of the CPU of a
> PPro-200. This was done almost exactly three years ago, using the (then)
> standard if_de driver. I have no idea whether the card was operating in
> store-and-forward mode or not - but the performance was perfectly fine.
I can guarantee that it was not running in store and forward mode if you
were exceeding 30 Mbit throughput.
Was this a 21140, 21140A or other stepping? These variations varied
pretty wildly in their behavior (and I suspect in their actual
implementation as well).
Cheers,
J. Hicks
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-hackers" in the body of the message
