[PATCH] if_sk(4) rx/tx "hangs"

[Hans Lambermont]

Bjoern A. Zeeb wrote:

> On Fri, 12 Nov 2004, Peter Wemm wrote:
>> Anyway, the point was that it worked!  Could the problem with the K8V
>> SE really be as simple as we've been hardcoding 128K of ram for a
>> device that only has 64K?
> 
> It's not directly a matter of the K8V SE.
> 
> It's the 88E8001 (but at least my K8V SE deluxe has this one acording
> to the PN from VPD data).  The 88E8010 already has 128k according to
> the pdf mentioned in my diff.
> 
> Anyone actually tested it ? Mine still seems to be fine after the
> night.

I have a Marvell Semiconductor, Inc. Yukon 88E8000 on my Asus P4P800-E
(according to dmesg that is, the booklet claims it's a 88E8001).

Using the patch on a stock 5.3R system did not help (could not even get
a dhcp lease). Adding debug.mpsafenet="0" in /boot/loader.conf also did
not help.

Adding SK_LOCK/SK_UNLOCK patch of
http://www.freebsd.org/cgi/query-pr.cgi?pr=kern%2F73038
also did not help.

Someone suggested upgrading the BIOS to >1003, the system now runs
1005.002, which ;-) did not help.

Hans Lambermont
-- 
http://lambermont.webhop.org/ () ASCII-ribbon campaign against vCards,
                              /\ HTML-mail and proprietary formats.