Tulip NIC 100Mb/s - HOW?

Does anyone have direct experience with DEC Tulip NIC's?

The trouble is it'll only work at 10Mb/s. I'm sure there is a setting for
modules.conf, however I've not been able to make it happen. I'll go into
modules.conf, or in my case modprobe.conf - in a minute. First, here's what
DMESG tells me:

<snip>
Linux Tulip driver version 1.1.13 (May 11, 2002)
PCI: Found IRQ 10 for device 0000:00:0a.0
PCI: Sharing IRQ 10 with 0000:00:07.5
tulip0: EEPROM default media type Autosense.
tulip0: Index #0 - Media 10baseT (#0) described by a 21140 non-MII (0)
tulip0: Index #1 - Media 100baseTx (#3) described by a 21140 non-MII (0)
tulip0: Index #2 - Media 100baseTx-FDX (#5) described by a 21140 non-MII
tulip0: Index #3 - Media 10baseT-FDX (#4) described by a 21140 non-MII)
eth0: Digital DS21140 Tulip rev 17 at 0001e000, 00:00:F8:00:07:E8, IRQ 10.
NET: Registered protocol family 10
</snip>

Right, the module is loaded thus:

alias eth0 tulip

Then the next line should have:

options tulip {here is where I get stuck}

Do you know *exactly* what to stick in after "options"?

In order to save us all time and bother, if you do not know the answer to
this query then please don't bother replying - I've probably already tried
everything (except the one thing that works) and so I'd be grateful to hear
from anyone who KNOWS EXACTLY what to do. Cheers.

In news:(E-Mail Removed),
Peter Lowrie <(E-Mail Removed)> wrote:

> Right, the module is loaded thus:
>
> alias eth0 tulip
>
> Then the next line should have:
>
> options tulip {here is where I get stuck}
>
> Do you know *exactly* what to stick in after "options"?

options tulip 100baseTx-FDX

On Fri, 17 Nov 2006 17:08:32 +1300, Peter Lowrie <(E-Mail Removed)> wrote:

>Does anyone have direct experience with DEC Tulip NIC's?

Yup, chucked them in favour of Intel pro/100 with e100 driver :^)

Okay, still have a couple around to play with...

a)
Linux Tulip driver version 1.1.13 (May 11, 2002)
PCI: Found IRQ 9 for device 0000:01:07.0
PCI: Sharing IRQ 9 with 0000:00:01.0
tulip0: MII transceiver #1 config 1000 status 782d advertising 01e1.
eth2: Digital DS21140 Tulip rev 34 at cca3cc00, 00:00:E8:4F:4D:3D, IRQ 9.

b)
Linux Tulip driver version 1.1.13 (May 11, 2002)
ACPI: PCI Interrupt Link [LNKC] enabled at IRQ 3
PCI: setting IRQ 3 as level-triggered
ACPI: PCI Interrupt 0000:00:0f.0[A] -> Link [LNKC] -> GSI 3 (level, low) -> IRQ 3
tulip0: EEPROM default media type Autosense.
tulip0: Index #0 - Media MII (#11) described by a 21140 MII PHY (1) block.
tulip0: MII transceiver #1 config 1000 status 7809 advertising 01e1.
eth2: Digital DS21140 Tulip rev 34 at d8d6c000, 00:00:E8:4A:A0:C8, IRQ 3.
....
eth2: Setting full-duplex based on MII#1 link partner capability of 01e1.

Both boxen running 2.6.16.32.

Up at 100Mbps with crossed cable to another tulip, also links at 100Mbps
with Intel e100 or a 3com 3c59x. Casual test, nothing beyond 'modprobe
tulip', followed by ifconfig for a ping test.

>Do you know *exactly* what to stick in after "options"?

Nada, zip, zilch. No option line needed, use default auto-sense.

Check what device you're sharing interrupts with, you don't want it
sharing with another NIC. Try changing the NIC to another PCI slot.

Have you tried with / without kernel PCI shared memory option? I have
that turned on.

Which kernel you running?

Grant.
--
http://bugsplatter.mine.nu/

Peter Lowrie <(E-Mail Removed)> writes:

>Does anyone have direct experience with DEC Tulip NIC's?

>The trouble is it'll only work at 10Mb/s.

I have used these since the first Netgear tulips - FA310TX cards, I
think - that's a bunch of years ago. My original network hub was
10baseT, & that speed was listed in dmesg.

I replaced my hub with a 100baseT & cables with cat5, & without other
intervention I got 100baseT, but not full duplex - turns out the hub
wasn't capable of full duplex.

A few years later, I replaced my hub with a 100baseT switch that was
full duplex capable, & dmesg listed 100baseT full duplex without other
intervention.

All with the very same nic. Same experience with a Lite-On LNE100TX
(82c168) PNIC.

Google up a copy of tulip-diag.c, or try
(http://www.scyld.com/diag/index.html)
compile it & see if it can be of any help.

Good luck, Bob
--
<> Robert Geer & Donna Tomky | |||| We sure |||| <>
<> (E-Mail Removed) | == == find it == == <>
<> (E-Mail Removed) | == == enchanting == == <>
<> Albuquerque, NM USA | |||| here! |||| <>

Peter Lowrie wrote:

> <snip>
> Linux Tulip driver version 1.1.13 (May 11, 2002)
> PCI: Found IRQ 10 for device 0000:00:0a.0
> PCI: Sharing IRQ 10 with 0000:00:07.5
> tulip0: EEPROM default media type Autosense.
> tulip0: Index #0 - Media 10baseT (#0) described by a 21140 non-MII (0)
> tulip0: Index #1 - Media 100baseTx (#3) described by a 21140 non-MII (0)
> tulip0: Index #2 - Media 100baseTx-FDX (#5) described by a 21140 non-MII
> tulip0: Index #3 - Media 10baseT-FDX (#4) described by a 21140 non-MII)
> eth0: Digital DS21140 Tulip rev 17 at 0001e000, 00:00:F8:00:07:E8, IRQ 10.


i had the same problem and resolved it by forcing the options for the
driver

# /etc/modprobe.conf
alias tulip eth0
options tulip options=5

it comes down to the driver getting the auto-detection wrong; the '5'
in the options refers to the "100baseTx-FDX (#5)" as listed in the
dmesg. you will have to determine which category your NIC falls into
but option 5 would be the one to try -- the diff between the 100baseTx
options is that one is full duplex (ie FDX) whilst the other isnt.

try removing the tulip driver and then doing "modprobe tulip options=5"
and then check the LEDS on the NIC/router -- you should see the 100mbps
fullduplex LEDS lit if that is the correct parameters for the tulip
driver

ray

i posted to this same NG about this subject; the original thread can be
seen by googling "tulip initialization" for the thread that i started
and finished