2005-12-16 21:26:04

by Vladimir Lazarenko

[permalink] [raw]
Subject: tulip on alpha ds10l 100MbpsFD problem

Hello,

I'm trying to get this DS 10L going with Linux, and I hit the ceiling
with the onboard ethernet.
Whenever the device is in 10Mbps, and switch is in 10 Mbps,
everything works ok (half-duplex mode). Whenever I switch to 100Mbps,
I'm not able to receive packets. At all. Sending goes ok, i.e. I see
DHCP requests on another server, but nothing reaches the box when
data is sent to it.

Similar behaviour with both de4x5 and tulip drivers.

lspci -vvv:

0000:00:09.0 Ethernet controller: Digital Equipment Corporation
DECchip 21142/4) Subsystem: Digital Equipment Corporation
DE500B Fast Ethernet
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Ste- Status: Cap- 66MHz- UDF- FastB2B+ ParErr-
DEVSEL=medium >TAbort- <TAbor- Latency: 255 (5000ns min,
10000ns max), Cache Line Size: 0x10 (64 bytes) Interrupt: pin
A routed to IRQ 29
Region 0: I/O ports at 8000 [size=128]
Region 1: Memory at 0000000009080000 (32-bit, non-
prefetchable) [size=1] Expansion ROM at 0000000009000000
[disabled] [size=256K]

0000:00:0b.0 Ethernet controller: Digital Equipment Corporation
DECchip 21142/4) Subsystem: Digital Equipment Corporation
DE500B Fast Ethernet
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Ste- Status: Cap- 66MHz- UDF- FastB2B+ ParErr-
DEVSEL=medium >TAbort- <TAbor- Latency: 255 (5000ns min,
10000ns max), Cache Line Size: 0x10 (64 bytes) Interrupt: pin
A routed to IRQ 30
Region 0: I/O ports at 8080 [size=128]
Region 1: Memory at 0000000009081000 (32-bit, non-
prefetchable) [size=1] Expansion ROM at 0000000009040000
[disabled] [size=256K]


tulip-diag -aem:

da-ds10-l:~# tulip-diag -aem
tulip-diag.c:v2.17 5/6/2003 Donald Becker ([email protected])
http://www.scyld.com/diag/index.html
Index #1: Found a Digital DS21143 Tulip adapter at 0x8000.
Digital DS21143 Tulip chip registers at 0x8000:
0x00: f8a0e000 ffffffff ffffffff 9c2d6000 9c2d6800 f0000000 b2420200
fbfffbff
0x40: e0000000 ffffcbf8 ffffffff 00000000 000012ce ffff0001 fffbffff
8ff00000
Port selection is 10mpbs-serial, full-duplex.
Transmit stopped, Receive stopped.
The Rx process state is 'Stopped'.
The Tx process state is 'Stopped'.
The transmit threshold is 72.
The NWay status register is 000012ce.
EEPROM 64 words, 6 address bits.
PCI Subsystem IDs, vendor 1011, device 500b.
CardBus Information Structure at offset 00000000.
Ethernet MAC Station Address 00:10:64:30:44:55.
EEPROM transceiver/media description table.
Leaf node at offset 46, default media type 0800 (Autosense).
4 transceiver description blocks:
Media 10baseT, block type 2, length 6.
Serial transceiver for 10baseT (media type 0).
GP pin direction 08af GP pin data 00a0.
Media 10baseT-Full Duplex, block type 2, length 6.
Serial transceiver for 10baseT-Full Duplex (media type 4).
GP pin direction 08af GP pin data 00a0.
Media 100baseTx, block type 4, length 8.
SYM transceiver for 100baseTx (media type 3).
GP pin direction 08af GP pin data 00a0.
No media detection indication (command 80 61).
Media 100baseTx Full Duplex, block type 4, length 8.
SYM transceiver for 100baseTx Full Duplex (media type 5).
GP pin direction 08af GP pin data 00a0.
No media detection indication (command 80 61).
No MII transceivers found!
Internal autonegotiation state is 'Ability detect'.
Index #2: Found a Digital DS21143 Tulip adapter at 0x8080.
Digital DS21143 Tulip chip registers at 0x8080:
0x00: f8004800 ffffffff ffffffff 9faf0000 9faf0080 f0000000 b2420200
f3fe0000
0x40: e0000000 ffffcbf8 ffffffff 00000000 000021c6 ffff0001 fffbffff
8ff00000
Port selection is 10mpbs-serial, full-duplex.
Transmit stopped, Receive stopped.
The Rx process state is 'Stopped'.
The Tx process state is 'Stopped'.
The transmit threshold is 72.
The NWay status register is 000021c6.
EEPROM 64 words, 6 address bits.
PCI Subsystem IDs, vendor 1011, device 500b.
CardBus Information Structure at offset 00000000.
Ethernet MAC Station Address 00:10:64:30:44:56.
EEPROM transceiver/media description table.
Leaf node at offset 46, default media type 0800 (Autosense).
4 transceiver description blocks:
Media 10baseT, block type 2, length 6.
Serial transceiver for 10baseT (media type 0).
GP pin direction 08af GP pin data 00a0.
Media 10baseT-Full Duplex, block type 2, length 6.
Serial transceiver for 10baseT-Full Duplex (media type 4).
GP pin direction 08af GP pin data 00a0.
Media 100baseTx, block type 4, length 8.
SYM transceiver for 100baseTx (media type 3).
GP pin direction 08af GP pin data 00a0.
No media detection indication (command 80 61).
Media 100baseTx Full Duplex, block type 4, length 8.
SYM transceiver for 100baseTx Full Duplex (media type 5).
GP pin direction 08af GP pin data 00a0.
No media detection indication (command 80 61).
No MII transceivers found!
Internal autonegotiation state is 'Ability detect'.


Note, that it thinks that the device is in 10mbps full duplex mode,
however when i force 10mbps on the switch, it will report 10mbps half-
duplex, and that's the only configuration it works properly in.

SRM ewa0_mode is set to FastFD. It doesn't work with Auto-Negotiate
either.

Any ideas would be very welcome. I'd be very happy to get this baby
going in 100Mbps mode.
As a sidenote, alphaserver800 which is next to it, works in 100MbpsFD
perfectly.

Regards,
Vladimir


2005-12-17 00:18:42

by Tim Schmielau

[permalink] [raw]
Subject: Re: tulip on alpha ds10l 100MbpsFD problem

On Fri, 16 Dec 2005, Vladimir Lazarenko wrote:

> Hello,
>
> I'm trying to get this DS 10L going with Linux, and I hit the ceiling with the
> onboard ethernet.
> Whenever the device is in 10Mbps, and switch is in 10 Mbps, everything works
> ok (half-duplex mode). Whenever I switch to 100Mbps, I'm not able to receive
> packets. At all. Sending goes ok, i.e. I see DHCP requests on another server,
> but nothing reaches the box when data is sent to it.
>
> Similar behaviour with both de4x5 and tulip drivers.

Try to connect them to a different switch.

I had the same problem with some XP1000s some years ago. It went away when
I connected them to a different switch. Since then, the machines have been
on four different switches, and only had the problem with one of them.
Strangely it was an expensive managed switch that didn't work, while two
el cheapo 8 port switches had no problems. Somewhere I read this was a
hardware bug of the cards.

Good luck!
Tim