Re: tulip driver problem

Re: tulip driver problem

[Colonel]

In clouddancer.list.kernel, Tom Rini wrote:
>
>On Mon, Aug 06, 2001 at 12:19:10PM -0400, Albert D. Cahalan wrote:
>
>> This is the Force PowerCore 6750 single-board computer with
>> a PowerPC processor and the DEC 21143 Ethernet chip.
>
>Just wondering, but when booting 2.4.x, do you see something like:
>"Unknown bridge resource %d: assuming transparent"
>for the tulip?
>

Linux ns1.clouddancer.com 2.4.8-pre4 #3 SMP Sat Aug 4 12:32:55 PDT 2001 i686 unknown
...
Unknown bridge resource 0: assuming transparent
Unknown bridge resource 1: assuming transparent
Unknown bridge resource 2: assuming transparent
...
Linux Tulip driver version 0.9.15-pre6 (July 2, 2001)
tulip0:  EEPROM default media type Autosense.
tulip0:  Index #0 - Media MII (#11) described by a 21140 MII PHY (1) block.
tulip0:  MII transceiver #0 config 1000 status 782d advertising 01e1.
eth0: Digital DS21140 Tulip rev 34 at 0xb400, 00:80:C8:45:97:AF, IRQ 12.


It's been working fine here, 3 cards over 2 computers.



-- 
Windows 2001: "I'm sorry Dave ...  I'm afraid I can't do that."

Re: tulip driver problem

[Tom Rini]

On Mon, Aug 06, 2001 at 07:32:00PM -0700, Colonel wrote:
> In clouddancer.list.kernel, Tom Rini wrote:
> >
> >On Mon, Aug 06, 2001 at 12:19:10PM -0400, Albert D. Cahalan wrote:
> >
> >> This is the Force PowerCore 6750 single-board computer with
> >> a PowerPC processor and the DEC 21143 Ethernet chip.
> >
> >Just wondering, but when booting 2.4.x, do you see something like:
> >"Unknown bridge resource %d: assuming transparent"
> >for the tulip?
> 
> Linux ns1.clouddancer.com 2.4.8-pre4 #3 SMP Sat Aug 4 12:32:55 PDT 2001 i686 unknown
> ...
> Unknown bridge resource 0: assuming transparent
> Unknown bridge resource 1: assuming transparent
> Unknown bridge resource 2: assuming transparent
> ...
> Linux Tulip driver version 0.9.15-pre6 (July 2, 2001)
> tulip0:  EEPROM default media type Autosense.
> tulip0:  Index #0 - Media MII (#11) described by a 21140 MII PHY (1) block.
> tulip0:  MII transceiver #0 config 1000 status 782d advertising 01e1.
> eth0: Digital DS21140 Tulip rev 34 at 0xb400, 00:80:C8:45:97:AF, IRQ 12.
> 
> It's been working fine here, 3 cards over 2 computers.

Yes, but what arch and do either of the above corrispond to the tulip?
I asked because sometimes when the code prints msgs like the above, it's
doing the wrong thing and ends up wreaking havoc.

-- 
Tom Rini (TR1265)
http://gate.crashing.org/~trini/

RE: tulip driver problem

[chen, xiangping]

I have similar problem. The tulip driver in 2.4.7 doesn't work.
The driver version is 0.9.15-pre6.
Error msg:
	tulip0: ***Warning*** : No MII transeiver found.

The tulip driver version 0.9.14 in 2.4.2 works, though.


-----Original Message-----
From: Albert D. Cahalan [mailto:acahalan@cs.uml.edu]
Sent: Monday, August 06, 2001 12:19 PM
To: linux-kernel@vger.kernel.org
Subject: tulip driver problem



The tulip driver in 2.4.8-pre3 does not work.
The tulip driver version 0.9.14 from sourceforge does not work.
The tulip driver in 2.2.14 is OK.
The de4x5 driver does not work.
I think I had the driver in 2.4.3 working at one point.

The error I'm getting from the 0.9.14 driver:
NETDEV WATCHDOG: eth0: transmit timed out

This is the Force PowerCore 6750 single-board computer with
a PowerPC processor and the DEC 21143 Ethernet chip.

Booting 2.2.14 shows:

POSIX conformance testing by UNIFIX
PCI: Probing PCI hardware
remapping IO (0x10e3:0x0 BAR 0): 0xbff001 -> 0xbff000
remapping MEM(0x10e3:0x0 BAR 1): fcfff000 -> 0xfcfff000
remapping IO (0x1011:0x19 BAR 0): 0xbfef81 -> 0xbfef80
remapping MEM(0x1011:0x19 BAR 1): fcffec00 -> 0xfcffec00
Linux NET4.0 for Linux 2.2
...
RAM disk driver initialized:  16 RAM disks of 4096K size
tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov
eth0: Digital DS21143 Tulip rev 65 at 0xbfef80, 00:80:42:0E:CF:E8, IRQ 9.
eth0:  EEPROM default media type Autosense.
eth0:  Index #0 - Media MII (#11) described by a 21142 MII PHY (3) block.
eth0:  Index #1 - Media AUI (#2) described by a 21142 Serial PHY (2) block.
eth0:  MII transceiver #1 config 1000 status 782d advertising 00a1.
Looking up port of RPC 100003/2 on 172.16.101.112
Looking up port of RPC 100005/1 on 172.16.101.112
VFS: Mounted root (NFS filesystem).
Freeing unused kernel memory: 68k init 32k prep 8k pmac 12k open firmware

The firmware (which likes the Ethernet just fine) reports the
host bridge and Ethernet as:

Probing PCIbus at 0x80000000
Device ID = 0x0002; Vendor ID = 0x1057; 
Status    = 0x0080; Command   = 0x0146; 
Base Class= 0x06;   Sub Class = 0x00;   Prg. Inter= 0x00;   Rev. ID   =
0x40; 
BIST      = 0x00;   Header Typ= 0x00;   Latency Ti= 0x00;   Cache Line=
0x08; 
base addr0= 0x00000000, base addr1= 0x00000000; 
Max Lat   = 0x00;   Min Gnt   = 0x00;   IRQ Pin   = 0x00;   IRQ Line  =
0x00; 
Found PCI device: Motorola MPC106 PowerPC PCI bridge

Probing PCIbus at 0x8000D800
Device ID = 0x0019; Vendor ID = 0x1011; 
Status    = 0x0280; Command   = 0x0005; 
Base Class= 0x02;   Sub Class = 0x00;   Prg. Inter= 0x00;   Rev. ID   =
0x41; 
BIST      = 0x00;   Header Typ= 0x00;   Latency Ti= 0x20;   Cache Line=
0x00; 
base addr0= 0x00850001, base addr1= 0x00000000; 
Max Lat   = 0x28;   Min Gnt   = 0x14;   IRQ Pin   = 0x01;   IRQ Line  =
0xFF; 
Found PCI device: DEC 21143 PCI/Cardbus Ethernet LAN

When using lspci I get:

00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
        Flags: bus master, fast devsel, latency 0

00:1b.0 Ethernet controller: Digital Equipment Corporation DECchip 21142/43
(rev 41)
        Flags: bus master, medium devsel, latency 32, IRQ 9
        I/O ports at bfef80
        Memory at fcffec00 (32-bit, non-prefetchable)
        Expansion ROM at b7fc0000 [disabled]
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

tulip driver problem

[Albert D. Cahalan]

The tulip driver in 2.4.8-pre3 does not work.
The tulip driver version 0.9.14 from sourceforge does not work.
The tulip driver in 2.2.14 is OK.
The de4x5 driver does not work.
I think I had the driver in 2.4.3 working at one point.

The error I'm getting from the 0.9.14 driver:
NETDEV WATCHDOG: eth0: transmit timed out

This is the Force PowerCore 6750 single-board computer with
a PowerPC processor and the DEC 21143 Ethernet chip.

Booting 2.2.14 shows:

POSIX conformance testing by UNIFIX
PCI: Probing PCI hardware
remapping IO (0x10e3:0x0 BAR 0): 0xbff001 -> 0xbff000
remapping MEM(0x10e3:0x0 BAR 1): fcfff000 -> 0xfcfff000
remapping IO (0x1011:0x19 BAR 0): 0xbfef81 -> 0xbfef80
remapping MEM(0x1011:0x19 BAR 1): fcffec00 -> 0xfcffec00
Linux NET4.0 for Linux 2.2
...
RAM disk driver initialized:  16 RAM disks of 4096K size
tulip.c:v0.91g-ppc 7/16/99 becker@cesdis.gsfc.nasa.gov
eth0: Digital DS21143 Tulip rev 65 at 0xbfef80, 00:80:42:0E:CF:E8, IRQ 9.
eth0:  EEPROM default media type Autosense.
eth0:  Index #0 - Media MII (#11) described by a 21142 MII PHY (3) block.
eth0:  Index #1 - Media AUI (#2) described by a 21142 Serial PHY (2) block.
eth0:  MII transceiver #1 config 1000 status 782d advertising 00a1.
Looking up port of RPC 100003/2 on 172.16.101.112
Looking up port of RPC 100005/1 on 172.16.101.112
VFS: Mounted root (NFS filesystem).
Freeing unused kernel memory: 68k init 32k prep 8k pmac 12k open firmware

The firmware (which likes the Ethernet just fine) reports the
host bridge and Ethernet as:

Probing PCIbus at 0x80000000
Device ID = 0x0002; Vendor ID = 0x1057; 
Status    = 0x0080; Command   = 0x0146; 
Base Class= 0x06;   Sub Class = 0x00;   Prg. Inter= 0x00;   Rev. ID   = 0x40; 
BIST      = 0x00;   Header Typ= 0x00;   Latency Ti= 0x00;   Cache Line= 0x08; 
base addr0= 0x00000000, base addr1= 0x00000000; 
Max Lat   = 0x00;   Min Gnt   = 0x00;   IRQ Pin   = 0x00;   IRQ Line  = 0x00; 
Found PCI device: Motorola MPC106 PowerPC PCI bridge

Probing PCIbus at 0x8000D800
Device ID = 0x0019; Vendor ID = 0x1011; 
Status    = 0x0280; Command   = 0x0005; 
Base Class= 0x02;   Sub Class = 0x00;   Prg. Inter= 0x00;   Rev. ID   = 0x41; 
BIST      = 0x00;   Header Typ= 0x00;   Latency Ti= 0x20;   Cache Line= 0x00; 
base addr0= 0x00850001, base addr1= 0x00000000; 
Max Lat   = 0x28;   Min Gnt   = 0x14;   IRQ Pin   = 0x01;   IRQ Line  = 0xFF; 
Found PCI device: DEC 21143 PCI/Cardbus Ethernet LAN

When using lspci I get:

00:00.0 Host bridge: Motorola MPC106 [Grackle] (rev 40)
        Flags: bus master, fast devsel, latency 0

00:1b.0 Ethernet controller: Digital Equipment Corporation DECchip 21142/43 (rev 41)
        Flags: bus master, medium devsel, latency 32, IRQ 9
        I/O ports at bfef80
        Memory at fcffec00 (32-bit, non-prefetchable)
        Expansion ROM at b7fc0000 [disabled]

Re: tulip driver problem

[Tom Rini]

On Mon, Aug 06, 2001 at 12:19:10PM -0400, Albert D. Cahalan wrote:

> This is the Force PowerCore 6750 single-board computer with
> a PowerPC processor and the DEC 21143 Ethernet chip.

Just wondering, but when booting 2.4.x, do you see something like:
"Unknown bridge resource %d: assuming transparent"
for the tulip?

-- 
Tom Rini (TR1265)
http://gate.crashing.org/~trini/

Re: tulip driver problem

[Albert D. Cahalan]

Tom Rini writes:
> On Mon, Aug 06, 2001 at 12:19:10PM -0400, Albert D. Cahalan wrote:

>> This is the Force PowerCore 6750 single-board computer with
>> a PowerPC processor and the DEC 21143 Ethernet chip.
>
> Just wondering, but when booting 2.4.x, do you see something like:
> "Unknown bridge resource %d: assuming transparent"
> for the tulip?

Nope. The closest is "Memory resource not set for host bridge 0".

This is the VME 6750 BTW, not the CompactPCI 6750.
I'll put my IRQ table adjustments below, in case they are wrong.

Right now, a 2.4.8-pre4 boot gives this:

ip_conntrack (2048 buckets, 16384 max)
ip_tables: (c)2000 Netfilter core team
NET4: Unix domain sockets 1.0/SMP for Linux NET4.0.
Looking up port of RPC 100003/2 on 172.16.101.112
eth0: Setting half-duplex based on MII#1 link partner capability of 0021.
NETDEV WATCHDOG: eth0: transmit timed out
NETDEV WATCHDOG: eth0: transmit timed out
NETDEV WATCHDOG: eth0: transmit timed out
NETDEV WATCHDOG: eth0: transmit timed out

That "eth0: Setting half-duplex" line doesn't show if I use the
other driver or forget to adjust the IRQ mappings; the rest is
the same no matter what.


        static char pci_irq_table[][4] =
        /*
         *      PCI IDSEL/INTPIN->INTLINE
         *      A       B       C       D
         */
        {
                {11,    14,     5,      10},    /* IDSEL 24 - Universe II */
                {0,     0,      0,      0},     /* IDSEL 25 - unused */
                {10,    11,     14,     5},     /* IDSEL 26 - Winbond */
                {10,    11,     14,     5},     /* IDSEL 27 - DEC 21143 */
                {14,    5,      10,     11},    /* IDSEL 28 - PMC I */
                {11,    14,     5,      10},    /* IDSEL 29 - PMC II */
                {0,     0,      0,      0},     /* IDSEL 30 - unused */
#if 0
                {9,     10,     11,     12},    /* IDSEL 24 - DEC 21554 */
                {10,    0,      0,      0},     /* IDSEL 25 - DEC 21143 */
                {11,    12,     9,      10},    /* IDSEL 26 - PMC I */
                {12,    9,      10,     11},    /* IDSEL 27 - PMC II */
                {0,     0,      0,      0},     /* IDSEL 28 - unused */
                {0,     0,      9,      0},     /* IDSEL 29 - unused */
                {0,     0,      0,      0},     /* IDSEL 30 - Winbond */
#endif
                };

Re: tulip driver problem

[Lech Szychowski]

> eth0: Setting half-duplex based on MII#1 link partner capability of 0021.
> NETDEV WATCHDOG: eth0: transmit timed out
> NETDEV WATCHDOG: eth0: transmit timed out
> NETDEV WATCHDOG: eth0: transmit timed out
> NETDEV WATCHDOG: eth0: transmit timed out

Does eth0 share IRQ with any other card? I got the very same messages
when one of my 4-way Adaptec (tulip based) ports got assigned the same
IRQ as my SCSI card (53c810a). BTW, that forced me to swich to using
de4x5 driver...

-- 
	Leszek.

-- lech7@pse.pl 2:480/33.7          -- REAL programmers use INTEGERS --
-- speaking just for myself...

Re: tulip driver problem

[Colonel]

In clouddancer.list.kernel, you wrote:
>
>Does eth0 share IRQ with any other card? I got the very same messages
>when one of my 4-way Adaptec (tulip based) ports got assigned the same
>IRQ as my SCSI card (53c810a). BTW, that forced me to swich to using
>de4x5 driver...
>-- lech7@pse.pl 2:480/33.7          -- REAL programmers use INTEGERS --



Goto Becker's website and fetch the IRQ sharing info, then patch the
scsi driver.


-- 
Windows 2001: "I'm sorry Dave ...  I'm afraid I can't do that."