Bug 62094

Summary:	3com 940 gigabit ethernet dhcp failure
Product:	Base System	Reporter:	Kenneth Culver <culverk>
Component:	kern	Assignee:	freebsd-bugs (Nobody) <bugs>
Status:	Closed FIXED
Severity:	Affects Only Me
Priority:	Normal
Version:	5.2-CURRENT
Hardware:	Any
OS:	Any

Description Kenneth Culver 2004-01-30 04:00:25 UTC

	On an Asus K8V Deluxe motherboard (athlon 64) in both the i386 and
	amd64 versions of FreeBSD, when the machine first boots up, and
	dhclient starts, it hangs, and eventually times out. During this time
	the lights on the switch showing whether the ethernet card is connected
	keep flashing on and off (not the activity lights but the ones that
	show whether it's actually got a connection).

	A few minutes after boot (and breaking out of the dhclient that happens
	on boot using ctrl-c) I can type "dhclient sk0" on the commandline,
	and there is no problem. This leads me to believe that the auto-
	negotiaton code for my PHY is messed up. Here is the relevent output
	from dmesg:

skc0: <3Com 3C940 Gigabit Ethernet> port 0xdc00-0xdcff mem 0xfdc00000-0xfdc03fff irq 10 at device 10.0 on pci0
skc0: 3Com Gigabit LOM (3C940)
sk0: <Marvell Semiconductor, Inc. Yukon> on skc0
sk0: Ethernet address: 00:0c:6e:ba:a7:49
miibus0: <MII bus> on sk0
e1000phy0: <Marvell 88E1000 Gigabit PHY> on miibus0
e1000phy0:  10baseT, 10baseT-FDX, 100baseTX, 100baseTX-FDX, 1000baseTX-FDX, auto

Fix: 

Currently the only workaround is to set a static ip, and hope
	nobody else is using it.
How-To-Repeat: 	All that's needed is to have this on-board ethernet controller, and try
	to use dhclient to get network configuration.

Comment 1 wouter 2004-05-09 10:16:28 UTC

Same problem here, on-board in an Asus L5800C notebook (Intel P4)

Any news on this issue yet?

Greetings,

-- 
Wouter de Jong
wouter@de-jong.biz

Comment 2 Bruce M Simpson freebsd_committer

2004-07-09 07:15:11 UTC

Responsible Changed
From-To: freebsd-bugs->wpaul

Assign to driver maintainer

Comment 3 Bruce M Simpson freebsd_committer

2004-07-09 07:20:41 UTC

State Changed
From-To: open->feedback

Can you please supply full dmesg output from a 'boot -v' boot (that is, 
ensure debug.bootverbose is set to 1) for the failure case here? Thanks.

Comment 4 Bill Paul freebsd_committer

2004-07-09 07:31:02 UTC

Responsible Changed
From-To: wpaul->freebsd-bugs

I wrote the original sk(4) driver. I did not write the patch to add support 
for the Marvell parts, and I have no sample hardware and hence no way to 
debug this. I am also not inclined to do any favors for SysKonnect and/or 
Marvell.

Comment 5 Kris Kennaway freebsd_committer

2005-07-30 02:26:23 UTC

State Changed
From-To: feedback->closed

Feedback timeout