2006-03-06 12:28:27

by PaNiC

[permalink] [raw]
Subject: Problem: NIC transmit timeouts

1. The problem is that the outbound interface in a Sun Enterprise 250
running maquerade gets transmit timeouts frequently.

2. I get the error "NETDEV WATCHDOG: eth0: transmit timed out" and a
couple of seconds later the interface jumps back up again. This
seems only to happen after masquerading and packet forwarding is
activated. When i have momentary traffic going in or out (web browsing
for instance) it happens mostly. It does not happen as often when
downloading large files, but will surely happen with somewhat heavy
traffic going in and out for instance while talking on Skype. I tried
reversing the interfaces so that the outbound would be the inbound and
vice versa, mostly to exclude a hardware problem. This moved the
timeouts again to the outbound interface. This was while using kernel
2.4.27. The same problem appears with 2.6.8 and 2.6.15.4 which i am
currently using. Though not as often as with 2.4.27. I have googled
immensely but have only found people with the same problem, no
solutions.

4. Linux version 2.6.15.4 (root@kronos) (gcc version 3.3.5 (Debian
1:3.3.5-13))

7. Sun Enterprise 250, UltrasparcII-400, Debian 3.1,

7.1
Linux kronos 2.6.15.4 #2 SMP Mon Feb 27 17:48:33 CET 2006 sparc64
GNU/Linux

Gnu C 3.3.5
Gnu make 3.80
binutils 2.15
util-linux 2.12p
mount 2.12p
module-init-tools 3.2-pre1
e2fsprogs 1.37
reiserfsprogs line
reiser4progs line
PPP 2.4.3
Linux C Library 2.3.2
Dynamic linker (ldd) 2.3.2
Procps 3.2.1
Net-tools 1.60
Kbd 83:
Sh-utils 5.2.1
Modules Loaded isofs

7.2
cpu : TI UltraSparc II (BlackBird)
fpu : UltraSparc II integrated FPU
promlib : Version 3 Revision 16
prom : 3.16.1
type : sun4u
ncpus probed : 1
ncpus active : 1
D$ parity tl1 : 0
I$ parity tl1 : 0
Cpu0Bogo : 798.72
Cpu0ClkTck : 0000000017d78400
MMU Type : Spitfire
State:
CPU0: online

7.3
isofs 34376 0 - Live 0x0000000010000000

7.4
ioports:
1fe02000000-1fe0200ffff : PSYCHO0 PBMA
1fe02010000-1fe0201ffff : PSYCHO0 PBMB
1fe02011000-1fe020110ff : 0001:00:03.0
1fe02011000-1fe020110ff : sym53c8xx
1fe02011400-1fe020114ff : 0001:00:03.1
1fe02011400-1fe020114ff : sym53c8xx

iomem:
1ff00000000-1ff7fffffff : PSYCHO0 PBMA
1ff000a0000-1ff000bffff : Video RAM area
1ff000c0000-1ff000c7fff : Video ROM
1ff000f0000-1ff000fffff : System ROM
1ff80000000-1ffffffffff : PSYCHO0 PBMB
1ff800a0000-1ff800bffff : Video RAM area
1ff800c0000-1ff800c7fff : Video ROM
1ff800f0000-1ff800fffff : System ROM
1ff81000000-1ff81ffffff : 0001:01:00.0
1ff82000000-1ff827fffff : 0001:01:00.0
1ff83000000-1ff83ffffff : 0001:01:00.0
1ff84000000-1ff84007fff : 0001:01:00.1
1ff84000000-1ff84007fff : sunhme
1ff85000000-1ff85ffffff : 0001:01:01.0
1ff86000000-1ff867fffff : 0001:01:01.0
1ff87000000-1ff87ffffff : 0001:01:01.0
1ff88000000-1ff88007fff : 0001:01:01.1
1ff88000000-1ff88007fff : sunhme
1ff89000000-1ff89ffffff : 0001:01:02.0
1ff8a000000-1ff8a7fffff : 0001:01:02.0
1ff8b000000-1ff8bffffff : 0001:01:02.0
1ff8c000000-1ff8c007fff : 0001:01:02.1
1ff8c000000-1ff8c007fff : sunhme
1ff8d000000-1ff8dffffff : 0001:00:01.0
1ff8e000000-1ff8effffff : 0001:00:01.1
1ff8f000000-1ff8fffffff : 0001:01:00.1
1ff90000000-1ff90007fff : 0001:01:03.1
1ff90000000-1ff90007fff : sunhme
1ff90100000-1ff90107fff : 0001:00:01.1
1ff90100000-1ff90107fff : sunhme
1ff90108000-1ff901080ff : 0001:00:03.0
1ff90108000-1ff901080ff : sym53c8xx
1ff9010a000-1ff9010afff : 0001:00:03.0
1ff9010a000-1ff9010afff : sym53c8xx
1ff9010c000-1ff9010c0ff : 0001:00:03.1
1ff9010c000-1ff9010c0ff : sym53c8xx
1ff9010e000-1ff9010efff : 0001:00:03.1
1ff9010e000-1ff9010efff : sym53c8xx
1ff91000000-1ff91ffffff : 0001:01:01.1
1ff92000000-1ff92ffffff : 0001:01:02.1
1ff93000000-1ff93ffffff : 0001:01:03.1
1fff0000000-1fff0ffffff : 0001:00:01.0
1fff0000000-1fff00fffff : flashprom
1fff1000000-1fff17fffff : 0001:00:01.0
1fff1000000-1fff1001fff : eeprom
1fff1200000-1fff120007f : se
1fff1300398-1fff1300399 : ecpp
1fff13043bc-1fff13043cb : ecpp
1fff13062f8-1fff13062ff : su
1fff13083f8-1fff13083ff : su
1fff1400000-1fff140007f : se
1fff1500000-1fff1500007 : sc
1fff1504000-1fff1504002 : SUNW,pll
1fff1600000-1fff1600003 : SUNW,envctrltwo
1fff1700000-1fff170000f : ecpp
1fff1724000-1fff1724003 : power
1fff1726000-1fff1726003 : auxio
1fff1728000-1fff1728003 : auxio
1fff172a000-1fff172a003 : auxio
1fff172c000-1fff172c003 : auxio
1fff172f000-1fff172f003 : auxio

7.5
0000:80:00.0 Host bridge: Sun Microsystems Computer Corp. Psycho PCI
Bus Module
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr+ Stepping- SERR- FastB2B-
Status: Cap- 66MHz+ UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort+ >SERR- <PERR-
Latency: 64

0001:00:00.0 Host bridge: Sun Microsystems Computer Corp. Psycho PCI Bus
Module
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr+ Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort+ >SERR- <PERR-
Latency: 64

0001:00:01.0 Bridge: Sun Microsystems Computer Corp. EBUS (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr+ Stepping- SERR+ FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 6250ns max), Cache Line Size: 0x10 (64
bytes)
Region 0: Memory at 000001fff0000000 (32-bit, non-prefetchable)
[size=16M]
Region 1: Memory at 000001fff1000000 (32-bit, non-prefetchable)
[size=8M]
Expansion ROM at 000000008d000000 [disabled] [size=16M]

0001:00:01.1 Ethernet controller: Sun Microsystems Computer Corp. Happy
Meal (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 1250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin ? routed to IRQ 8038560
Region 0: Memory at 000001ff90100000 (32-bit, non-prefetchable)
[size=32K]
Expansion ROM at 000000008e000000 [disabled] [size=16M]

0001:00:02.0 PCI bridge: Digital Equipment Corporation DECchip 21153
(rev 04) (prog-if 00 [Normal decode])
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 64, Cache Line Size: 0x10 (64 bytes)
Bus: primary=00, secondary=01, subordinate=01, sec-latency=64
I/O behind bridge: 00001000-00000fff
Memory behind bridge: 00100000-100fffff
Prefetchable memory behind bridge:
fffffffffff00000-0000000000000000
BridgeCtl: Parity- SERR- NoISA- VGA- MAbort- >Reset- FastB2B-
Capabilities: [dc] Power Management version 1
Flags: PMEClk- DSI- D1- D2- AuxCurrent=220mA
PME(D0-,D1-,D2-,D3hot-,D3cold-)
Status: D0 PME-Enable- DSel=0 DScale=0 PME-
Bridge: PM- B3+

0001:00:03.0 SCSI storage controller: LSI Logic / Symbios Logic 53c875
(rev 14)
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop-
ParErr+ Stepping- SERR+ FastB2B-
Status: Cap- 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 17 (4250ns min, 16000ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin A routed to IRQ 8038528
Region 0: I/O ports at 2011000 [size=256]
Region 1: Memory at 000001ff90108000 (32-bit, non-prefetchable)
[size=256]
Region 2: Memory at 000001ff9010a000 (32-bit, non-prefetchable)
[size=4K]

0001:00:03.1 SCSI storage controller: LSI Logic / Symbios Logic 53c875
(rev 14)
Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV+ VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B- ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 17 (4250ns min, 16000ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin B routed to IRQ 8038720
Region 0: I/O ports at 2011400 [size=256]
Region 1: Memory at 000001ff9010c000 (32-bit, non-prefetchable)
[size=256]
Region 2: Memory at 000001ff9010e000 (32-bit, non-prefetchable)
[size=4K]

0001:01:00.0 Bridge: Sun Microsystems Computer Corp. EBUS (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 6250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin A routed to IRQ 0
Region 0: Memory at 000001ff81000000 (32-bit, non-prefetchable)
[size=16M]
Region 1: Memory at 000001ff82000000 (32-bit, non-prefetchable)
[size=8M]
Expansion ROM at 0000000083000000 [disabled] [size=16M]

0001:01:00.1 Ethernet controller: Sun Microsystems Computer Corp. Happy
Meal (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 1250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin B routed to IRQ 8038048
Region 0: Memory at 000001ff84000000 (32-bit, non-prefetchable)
[size=32K]
Expansion ROM at 000000008f000000 [disabled] [size=16M]

0001:01:01.0 Bridge: Sun Microsystems Computer Corp. EBUS (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 6250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin A routed to IRQ 0
Region 0: Memory at 000001ff85000000 (32-bit, non-prefetchable)
[size=16M]
Region 1: Memory at 000001ff86000000 (32-bit, non-prefetchable)
[size=8M]
Expansion ROM at 0000000087000000 [disabled] [size=16M]

0001:01:01.1 Ethernet controller: Sun Microsystems Computer Corp. Happy
Meal (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 1250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin B routed to IRQ 8038080
Region 0: Memory at 000001ff88000000 (32-bit, non-prefetchable)
[size=32K]
Expansion ROM at 0000000091000000 [disabled] [size=16M]

0001:01:02.0 Bridge: Sun Microsystems Computer Corp. EBUS (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 6250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin A routed to IRQ 0
Region 0: Memory at 000001ff89000000 (32-bit, non-prefetchable)
[size=16M]
Region 1: Memory at 000001ff8a000000 (32-bit, non-prefetchable)
[size=8M]
Expansion ROM at 000000008b000000 [disabled] [size=16M]

0001:01:02.1 Ethernet controller: Sun Microsystems Computer Corp. Happy
Meal (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 1250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin B routed to IRQ 8038112
Region 0: Memory at 000001ff8c000000 (32-bit, non-prefetchable)
[size=32K]
Expansion ROM at 0000000092000000 [disabled] [size=16M]

0001:01:03.1 Ethernet controller: Sun Microsystems Computer Corp. Happy
Meal (rev 01)
Control: I/O- Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop-
ParErr- Stepping- SERR- FastB2B-
Status: Cap- 66MHz- UDF- FastB2B+ ParErr- DEVSEL=medium >TAbort-
<TAbort- <MAbort- >SERR- <PERR-
Latency: 10 (2500ns min, 1250ns max), Cache Line Size: 0x10 (64
bytes)
Interrupt: pin B routed to IRQ 8038016
Region 0: Memory at 000001ff90000000 (32-bit, non-prefetchable)
[size=32K]
Expansion ROM at 0000000093000000 [disabled] [size=16M]

7.6
Attached devices:
Host: scsi0 Channel: 00 Id: 00 Lun: 00
Vendor: SEAGATE Model: ST318203LSUN18G Rev: 034A
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 06 Lun: 00
Vendor: PIONEER Model: CD-ROM DR-U16S Rev: 1.01
Type: CD-ROM ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 09 Lun: 00
Vendor: IBM-PCCO Model: DDRS-39130Y !# Rev: S97B
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 10 Lun: 00
Vendor: IBM-PCCO Model: DDRS-39130Y !# Rev: S97B
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 11 Lun: 00
Vendor: IBM-PCCO Model: DDRS-39130Y !# Rev: S97B
Type: Direct-Access ANSI SCSI revision: 02
Host: scsi0 Channel: 00 Id: 12 Lun: 00
Vendor: IBM-PCCO Model: DDRS-39130Y !# Rev: S97B
Type: Direct-Access ANSI SCSI revision: 02

7.7 Interfaces are a Sun Happy Meal and a Sun quad interface.

I will be thankful for any light on this. If i can do anything to help,
i will.
Thanks for your time.

Best regards
Jonas Bengtsson



2006-03-06 17:34:06

by John Z. Bohach

[permalink] [raw]
Subject: Re: Problem: NIC transmit timeouts

On Monday 06 March 2006 04:28, PaNiC wrote:
> 1. The problem is that the outbound interface in a Sun Enterprise 250
> running maquerade gets transmit timeouts frequently.
>
> 2. I get the error "NETDEV WATCHDOG: eth0: transmit timed out" and a
> couple of seconds later the interface jumps back up again. This

I can't say what the cause of your particular NETDEV WATCHDOG timeout may
be, but I had the same problem, and I root-caused it to the host bus <--> PCI bridge
configuration. In particular, the multi-transaction timeout register in the bridge
wasn't programmed, and heavy PCI traffic would cause aborts. Also, the
ICH configuration register had to be programmed according to the manufacturer's
recommendations.

This was on Intel h/w, and the registers to which I refer are
proprietary, so its a bit difficult to know what values to program where,
but it might give you a place to start. On the other hand, some people have
reported issues with their device driver causing some timeouts, but your symptoms
seem to more closely resemble what I was seeing than those folks who had
s/w issues.

Regards,
John

2006-03-07 12:08:50

by PaNiC

[permalink] [raw]
Subject: Re: Problem: NIC transmit timeouts

Thanks for your reply.
I have an idea what you're talking about, but no more i'm afraid.
I'm no programmer and i don't know how to try what you suggested.
What i did try was applying some patches manually that i found in the
mailing list archives.

This is one of them:

--- drivers/net/sunhme.c.~1~ Sun Aug 11 18:37:34 2002
+++ drivers/net/sunhme.c Sun Aug 11 18:38:17 2002
@@ -1640,6 +1640,7 @@
HMD((", enable global interrupts, "));
hme_write32(hp, gregs + GREG_IMASK,
(GREG_IMASK_GOTFRAME | GREG_IMASK_RCNTEXP |
+ GREG_IMASK_TXALL |
GREG_IMASK_SENTFRAME | GREG_IMASK_TXPERR));

/* Set the transmit ring buffer size. */
@@ -2125,8 +2126,8 @@
happy_meal_mif_interrupt(hp);
}

- if (happy_status & GREG_STAT_TXALL) {
- HMD(("TXALL "));
+ if (happy_status & GREG_STAT_HOSTTOTX) {
+ HMD(("HOSTTOTX "));
happy_meal_tx(hp);
}

@@ -2155,7 +2156,7 @@

if (!(happy_status & (GREG_STAT_ERRORS |
GREG_STAT_MIFIRQ |
- GREG_STAT_TXALL |
+ GREG_STAT_HOSTTOTX |
GREG_STAT_RXTOHOST)))
continue;

@@ -2172,8 +2173,8 @@
happy_meal_mif_interrupt(hp);
}

- if (happy_status & GREG_STAT_TXALL) {
- HMD(("TXALL "));
+ if (happy_status & GREG_STAT_HOSTTOTX) {
+ HMD(("HOSTTOTX "));
happy_meal_tx(hp);
}

The other just put a "udelay(1)" on line 1999 of sunhme.c.
With the udelay(1) it seemed like after a reboot it would take longer
for the timeout to happen. The other made no difference.

Very thankful for your time.
Jonas



----- Original Message -----
From: "John Z. Bohach" <[email protected]>
To: "PaNiC" <[email protected]>; <[email protected]>
Sent: Monday, March 06, 2006 6:33 PM
Subject: Re: Problem: NIC transmit timeouts


> On Monday 06 March 2006 04:28, PaNiC wrote:
>> 1. The problem is that the outbound interface in a Sun Enterprise 250
>> running maquerade gets transmit timeouts frequently.
>>
>> 2. I get the error "NETDEV WATCHDOG: eth0: transmit timed out" and a
>> couple of seconds later the interface jumps back up again. This
>
> I can't say what the cause of your particular NETDEV WATCHDOG timeout
> may
> be, but I had the same problem, and I root-caused it to the host bus
> <--> PCI bridge
> configuration. In particular, the multi-transaction timeout register
> in the bridge
> wasn't programmed, and heavy PCI traffic would cause aborts. Also,
> the
> ICH configuration register had to be programmed according to the
> manufacturer's
> recommendations.
>
> This was on Intel h/w, and the registers to which I refer are
> proprietary, so its a bit difficult to know what values to program
> where,
> but it might give you a place to start. On the other hand, some
> people have
> reported issues with their device driver causing some timeouts, but
> your symptoms
> seem to more closely resemble what I was seeing than those folks who
> had
> s/w issues.
>
> Regards,
> John
>
>
>


2006-03-07 16:48:03

by John Z. Bohach

[permalink] [raw]
Subject: Re: Problem: NIC transmit timeouts

On Tuesday 07 March 2006 04:08, PaNiC wrote:
> Thanks for your reply.
> I have an idea what you're talking about, but no more i'm afraid.
> I'm no programmer and i don't know how to try what you suggested.
> What i did try was applying some patches manually that i found in the
> mailing list archives.
>
> This is one of them:

I'll try to elaborate a little, but since my experience in dealing with the
NETDEV watchdog timeout issue is confined to x86 architectures,
some translation may be necessary...

I've seen two causes for this: 1) driver bug, 2) firmware bug. I have yet to
come across one where the kernel itself was at fault.

1) if its a driver bug, updating to the latest driver should have an effect,
hopefully a positive one. Generally, the driver maintainer is at least aware
of the problem, if this is the cause.

2) If its a firmware bug, you'll have to upgrade to the latest version of
the BIOS, uboot, rommon, or whatever your platform uses to boot the
OS.

Case 2) is the type of problem where many people can be using the exact
same kernel and exact same driver, and only a few experience any problems.
This is because its not the kernel or the driver, but the firmware that is at fault.

In particular, on the x86 platform where I've root-caused this issue, a different
version of the BIOS was present (I did an upgrade, and the upgrade failed to
implement the proper register programming, and that's when the problem showed
up). The manifestation is that heavy PCI traffic (not just ethernet traffic) would
overload the host-PCI bridge, and bad things started happening. In some cases,
just ethernet traffic alone was enough to overpower the bridge, but that dependended
on the link speed. 100 Mbs took a lot longer to NETDEV watchdog timeout than
1000 Mbs. Another fun aspect of case 2) is that since its PCI traffic related, its very hard
to recreate exactly, since there's usually other stuff on the PCI bus.

So if nobody seems to have a solution for your problem, it is possible that its
firmware-related, so try getting a different firmware. In particular, if you have any
experience at this, or know someone in your organization who does, try focusing on
the MTT (multi-transaction timeout) register, and the equivalent (if any) of the ICH
configuration register in the host-PCI bridge. These are both proprietary registers, and
are not present in the standard 64 byte PCI header, so its pretty hard to "guess" where they
are what they should be set to. If you have a similar platform that does not have any
NETDEV watchodog timeout issues, it may be possible to dump the PCI config. space
of the host-PCI bridge on that platform, and compare it with the failing platform, and start
"guessing" from there, but the firmware is the responsible entity that should have these
registers programmed correctly in the first place.

Sorry for being so verbose, I'm hoping that others googling in the future may save themselves
some time in isolating this pretty obscure cause...