>> > I am willing to try the 'unsupported' com driver to see if it fixes this
>> > problem; all I need is that address...

>> Note that at some point (hopefully soon) this is going to find its way
>> into the post-release branch.

>Has anyone investigated the real cause of the com problems?  I'm still afraid
>that something is keeping the interrupts disabled for too long and this
>causes loss of data in the com driver _and_ in some Ethernet drivers on slower
>machines.

	I am suddenly reminded of a problem that I've been having but
forgot about. A couple of my machines at home have SMC cards, a WD8003E
and a WD8013. About a month ago I upgraded and suddenly was unable to
NFS-boot the machine with the 8013. The screen just scrolls with

	ed0: device timeout
	ed0: device timeout
	ed0: device timeout

and the machine hasn't successfully booted via NFS since.

At the same time, I started noticing my log files on my 386/33 PPP router
filling up with these messages whenever another machine at the apartment
starts FTP'ing over the link.

It happens with both the 8003 and 8013, but the Kingston clone NE2000 16-bit
card in the other machine (also "ed" driver) has had no problems at all.

Does this remind anyone of any changes that took place around that period?

On another note, this 386/33 with a zoom 14.4k modem (4 byte FIFO), during
full-bore FTP sessions, has ~2-3 silo overflows per 1 minute period. Not
perfect but definitely a lot better than it used to be.

Cheers,
Phil
--
Phillip F Knaack
Database Programmer, Information Development for Extension Audiences (IDEA)
Iowa State University Extension