Skip site navigation (1)Skip section navigation (2)
Date:      Tue, 27 Oct 1998 18:01:32 -0800
From:      "Gregory P. Smith" <greg@nas.nasa.gov>
To:        Bill Paul <wpaul@skynet.ctr.columbia.edu>
Cc:        hardware@FreeBSD.ORG
Subject:   Re: Call for testers for Winbond W89C840F driver 
Message-ID:  <199810280201.SAA29654@ryouko.nas.nasa.gov>
In-Reply-To: Your message of "Tue, 27 Oct 1998 20:44:02 EST." <199810280144.UAA11413@skynet.ctr.columbia.edu> 

next in thread | previous in thread | raw e-mail | index | archive | help

> driver. The Winbond 840F is a tulip clone with a couple of wrinkles:
> the registers are spaced 4 bytes apart instead of 8 and the receive
> filter is programmed by writing directly to registers instead of
> loading a setup frame via the transmit DMA engine. The receive filter
> also has only one perfect address and a 64-bit multicast hash filter
> instead of a 16 entry perfect filter ans a 512-bit hash table.

Sounds like a pretty sad chip design.  Let's see, get rid of the
64-bit structure alignment, dump its ability to do a reasonable amount
of multicast efficiently...  I hope these don't catch on considering
real tulip chip based cards are already dirt cheap.  ;)

Ah well, good to see it supported none the less.

> So far, the only card I've found that uses the Winbond chip is the
> Trendware TE100-PCIE (www.trendware.com). Trendware also makes DEC
> tulip and PNIC adapters. I believe Winbond may make their own cards with 
> this chip as well.

-G

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hardware" in the body of the message



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199810280201.SAA29654>