2.4.18-rc1: Tulip driver fails to initialize 21140 card

2.4.18-rc1: Tulip driver fails to initialize 21140 card

Post by Jean-Francois Cardos » Fri, 15 Feb 2002 23:50:09



The tulip driver in 2.4.18-rc1 does not behave with my 21140 card:

1) At boot time, the tulip.o module is loaded OK but ifconfig only
   shows errors for both RX and TX --> no network.

2) The card can be made to work by briefly unplugging/replugging the
   ethernet wire into the hub.  After that, networking works, even
   though the log says eth0 uses half-duplex which does not sound good
   to me (who is an ignorant).

3) The old_tulip driver in vmlinuz-2.2.19-6.3mdk work well with this
   card.

I provide various outputs below.  Let me know if any other bits are needed.

Cheers, JF.

(PS: Jeff has been CC'ed in another message)

========================================================================

vmlinuz-2.2.19-6.3mdk messages
-------------------------------------------------------------------

        eth0: Digital DS21140 Tulip at 0x1000, 00 c0 fd 02 11 6f, IRQ 10.
        eth0:  EEPROM default media type Autosense.
        eth0:  Index #0 - Media MII (#11) described by a 21140 MII PHY (1) block.
        eth0:  MII transceiver found at MDIO address 1, config 3000 status 782d.

2.4.18-rc1 messages
-----------------------------------------------------
        Linux Tulip driver version 0.9.15-pre9 (Nov 6, 2001)
        PCI: Found IRQ 10 for device 00:0d.0
        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 3000 status 782d advertising 05e1.
        eth0: Digital DS21140 Tulip rev 34 at 0x1000, 00:C0:FD:02:11:6F, IRQ 10.

    On bringing up the interface:
        eth0: Setting half-duplex based on MII#1 link partner capability of 0000.

lspci output
------------

   Verbose for the NIC:

        00:0d.0 Ethernet controller: Digital Equipment Corporation DECchip 21140 [FasterNet] (rev 22)
        Flags: bus master, medium devsel, latency 165, IRQ 10
        I/O ports at 1000 [size=128]
        Memory at f4000000 (32-bit, non-prefetchable) [size=128]
        Expansion ROM at <unassigned> [disabled] [size=256K]

   Terse for the rest:

        00:00.0 Host bridge: Intel Corporation 440BX/ZX - 82443BX/ZX Host bridge (rev 02)
        00:01.0 PCI bridge: Intel Corporation 440BX/ZX - 82443BX/ZX AGP bridge (rev 02)
        00:07.0 ISA bridge: Intel Corporation 82371AB PIIX4 ISA (rev 02)
        00:07.1 IDE interface: Intel Corporation 82371AB PIIX4 IDE (rev 01)
        00:07.2 USB Controller: Intel Corporation 82371AB PIIX4 USB (rev 01)
        00:07.3 Bridge: Intel Corporation 82371AB PIIX4 ACPI (rev 02)
        00:0c.0 Multimedia audio controller: Ensoniq ES1371 [AudioPCI-97] (rev 02)
        00:0d.0 Ethernet controller: Digital Equipment Corporation DECchip 21140 [FasterNet] (rev 22)
        01:00.0 VGA compatible controller: NVidia / SGS Thomson (Joint Venture) Riva128 (rev 21)

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

 
 
 

1. Tulip driver broken in 2.4.18

Hello!

Tulip driver included in 2.4.18 doesn't work for me. I have tulip-based
card which is connected to 10Mbps HUB. The driver fails to set 10Mbps
halfduplex, and because of that the link between computer and HUB doesn't
work. HUB keeps flashing it's lights wildly (because my tulip tries to
enable 100Mbps..), and the link-led doesn't turn on..

Tulip driver 0.9.14 (available from sourceforge) works OK and correctly
sets 10Mbps halfduplex.

This is what mii-diag says when using 2.4.18:

Basic registers of MII PHY #32:  2000 7848 0000 0000 01e1 0000 0000 0000.
 Basic mode control register 0x2000: Auto-negotiation disabled, with
 Speed fixed at 100 mbps, half-duplex.
 Basic mode status register 0x7848 ... 7848.
   Link status: not established.
 Link partner information information is not exchanged when in fixed speed
mode.   End of basic transceiver informaion.

And after a couple of seconds it says this:

Basic registers of MII PHY #32:  1000 7848 0000 0000 01e1 0000 0000 0000.
 Basic mode control register 0x1000: Auto-negotiation enabled.
 Basic mode status register 0x7848 ... 7848.
   Link status: not established.
   End of basic transceiver informaion.

And then after a couple of seconds comes the 100Mbps again (the first
mii-diag paste) and loop continues..

Linux Tulip driver version 0.9.15-pre9 (Nov 6, 2001)
PCI: Found IRQ 5 for device 00:09.0
PCI: Sharing IRQ 5 with 00:04.2
tulip0:  EEPROM default media type Autosense.
tulip0:  Index #0 - Media 10baseT (#0) described by a 21142 Serial PHY (2) block.
tulip0:  Index #1 - Media 10baseT-FDX (#4) described by a 21142 Serial PHY  (2) block.
tulip0:  Index #2 - Media 100baseTx (#3) described by a 21143 SYM PHY (4) block.
tulip0:  Index #3 - Media 100baseTx-FDX (#5) described by a 21143  SYM PHY (4) block.
eth1: Digital DS21143 Tulip rev 65 at 0xb000, 00:C0:CA:20:3C:A5, IRQ 5.

Any ideas?

- Pasi K?rkk?inen

                                   ^
                                .     .
                                 Linux
                              /    -    \
                             Choice.of.the
                           .Next.Generation.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in

More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

2. XFree86, Radeon 8500LE 128DDR

3. Linux 2.4.18rc1 hangs when using hwclock, 2.4.18pre9 is fine

4. case insensitive file complete with tcsh

5. Linux Kernel Crash - Vanilla 2.4.18/Redhat 2.4.18-5

6. RH6.0

7. Linux Kernel Crash - Vanilla 2.4.18/Redhat 2.4.18-5 (2nd try =) )

8. WAN/MAN using BGP?

9. newnat13-and-helpers-2.4.18.gz Failed to patch with linux 2.4.18 kernel?

10. what is the difference between 2.4.18-14 and 2.4.18-17.8.0

11. Very High Load on Disk Activity in 2.4.18 (and 2.4.18-pre8)

12. 2.4.18-rc1

13. 2.4.18-rc1, trylock for read/write semaphores