5.0 and 5.0.1 misconfigures network card?

[skydan]

Knoppix 5.x is failing to set up the network connection on one of my machines here.
Networking works with 3.3, 3.7, 3.9 and 4.0 (LinuxTAG DVD).

Machine is BX based motherboard (ATC-6220), PII/400, 256Mb. NIC is 3com 3c900B.
DHCP served by firewall (Smoothwall 2.0) connected to DLink 604T. Various other
machines on network running Debian. They can ping each other and use the Internet
and everything, so it's not that before you ask.

With netcardconfig I can configure an IP address for eth0 but it the card is non reponsive
(cannot ping other machines and they cannot ping it). It shows up in ifconfig though.

vortex_diag (from Donald Becker's site) shows the card is using the correct ethernet
port (it's a combo card).

Running "ispci -v" shows knoppix has detected network card. It looks normal.
Base address=0x6800 which it has always been in w9x and previous knoppixs (knoppixen?).

Running "ifconfig -a" shows eth0 as a detected interface but base address is something
else. In fact it shows a different address everytime it boots. So far I've had 0x2000,
0x6000, 0x8000, 0xc000, 0xe000 from both 5.0 DVD and 5.0.1 CD.

I believe this is why it's not working? It's something new in 5.0 as 3.x and 4.0 worked.

On another note, the soundcard is fanally working! It never work pre-5.0. The card is
Soundmonster II (AU8830). It's always been detected on the boot screen, just never
worked from KDE. Might this even be related to the network card problem?
(Now I think about it I should go back and check its base address?)

I'm still learning about device configuration stuff so this is about as far as I can go.
Is this a bug in 5.0?
Is there a way to set the base address manually or otherwise get it working?

Oh, I have another similar machine, also BX base (AX6B+) PIII/800. Also 3c900B NIC
but Radeon QD and SBLive (vs Elsa Erazor III and SM2). The on-board SCSI confounds
SuSE Live DVDs and various disk utils, but the 5.0 DVD and 5.0.1 CD work flawlessly on it.
The mystery deepens. Any ideas?

Thanks.

[Zrebel]

This sounds like a post for the Networking forum, but I will try to help anyways. What is the info returned from ifconfig? Can you ping 127.0.0.1 ? does the IP address you get look like the addresses of the other machines on your network? i.e. knoppixbox has 192.168.0.5 and the others are like 192.168.0.2 or 192.168.0.23. If knoppixbox is 192.168.0.2 and others are 192.164... then they cannot talk because you are on wrong subnet. (the exact numbers are irrelevent. the first, second or(inclusive) third numbers must be the same depending on what class of address your network is using.

[skydan]

Thanks for the reply. I wondered whether it was networking or booting but I thing it's a boot config problem hence this forum. I didn't try loopback. I'll try it now... Ping to loopback ok.

I understand IP addrs and subnet masks etc. So, no. It's not the choice of IP addr. (FYI the network uses 192.168.7.xx and separates it from the red side of the firewall which is 192.168.1.xx). Knoppix boot up shows it detected eth0 and was dhcp'ing in the background as normal. Running ifconfig, however, didn't display eth0. "ifconfig -a" showed eth0 but not configuraed with an IP addr. In case DHCP failed for some reason I can manually configure eth0. But it was unresponsive. The base address (io addr whatever) shown by ifconfig differed from the base addr shown by "lspci -v". I know the value from lspci was the correct one and ifconfig was incorrect. In fact, on every reboot the ifconfig one changed (see original post). I didn't capture the output of lspci and ifconfig (no network and would mean finding a floopy) and thought I could explain it well enough.

Anyway here is the output of various commands so you can see if you concur with my diagnosis:

lspci -v

shows io port of ethernet controller is 0x6800 (highlighted with >>>)



0000:00:00.0 Host bridge: Intel Corporation 440BX/ZX/DX - 82443BX/ZX/DX Host bridge (rev 03)
Flags: bus master, medium devsel, latency 64
Memory at e0000000 (32-bit, prefetchable) [size=128M]
Capabilities: [a0] AGP version 1.0

0000:00:01.0 PCI bridge: Intel Corporation 440BX/ZX/DX - 82443BX/ZX/DX AGP bridge (rev 03) (prog-if 00 [Normal decode])
Flags: bus master, 66MHz, medium devsel, latency 64
Bus: primary=00, secondary=01, subordinate=01, sec-latency=64
I/O behind bridge: 0000e000-0000efff
Memory behind bridge: de000000-dfffffff
Prefetchable memory behind bridge: ae000000-afffffff

0000:00:07.0 ISA bridge: Intel Corporation 82371AB/EB/MB PIIX4 ISA (rev 02)
Flags: bus master, medium devsel, latency 0

0000:00:07.1 IDE interface: Intel Corporation 82371AB/EB/MB PIIX4 IDE (rev 01) (prog-if 80 [Master])
Flags: bus master, medium devsel, latency 64
I/O ports at f000 [size=16]

0000:00:07.2 USB Controller: Intel Corporation 82371AB/EB/MB PIIX4 USB (rev 01) (prog-if 00 [UHCI])
Flags: medium devsel
I/O ports at 6400 [size=32]

0000:00:07.3 Bridge: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 02)
Flags: medium devsel, IRQ 9

0000:00:0b.0 Ethernet controller: 3Com Corporation 3c900B-TPC Etherlink XL [Cyclone] (rev 04)
Subsystem: 3Com Corporation 3c900B-TPC Etherlink XL [Cyclone]
Flags: bus master, medium devsel, latency 64, IRQ 11
>>> I/O ports at 6800 [size=128]
Memory at e8040000 (32-bit, non-prefetchable) [size=128]
Expansion ROM at 20000000 [disabled] [size=128K]
Capabilities: [dc] Power Management version 1

0000:00:0c.0 Multimedia audio controller: Aureal Semiconductor Vortex 2 (rev fe)
Subsystem: Diamond Multimedia Systems Monster Sound II
Flags: bus master, medium devsel, latency 64, IRQ 10
Memory at e8000000 (32-bit, non-prefetchable) [size=256K]
I/O ports at 6c00 [size=8]
I/O ports at 7000 [size=8]
Capabilities: [dc] Power Management version 1

0000:01:00.0 VGA compatible controller: nVidia Corporation NV5 [RIVA TNT2/TNT2 Pro] (rev 11) (prog-if 00 [VGA])
Subsystem: Elsa AG Erazor III
Flags: bus master, 66MHz, medium devsel, latency 64, IRQ 9
Memory at de000000 (32-bit, non-prefetchable) [size=16M]
Memory at ae000000 (32-bit, prefetchable) [size=32M]
Expansion ROM at df000000 [disabled] [size=64K]
Capabilities: [60] Power Management version 1
Capabilities: [44] AGP version 2.0


ifconfig -a

shows base addr (io port) of eth0 is 0xc000 (highlighted with >>>)



eth0 Link encap:Ethernet HWaddr 00:10:5A:F7:2C:9B
BROADCAST MULTICAST MTU:1500 Metric:1
RX packets:0 errors:0 dropped:0 overruns:0 frame:0
TX packets:1 errors:0 dropped:0 overruns:0 carrier:1
collisions:0 txqueuelen:1000
RX bytes:0 (0.0 b) TX bytes:342 (342.0 b)
>>> Interrupt:11 Base address:0xc000

lo Link encap:Local Loopback
inet addr:127.0.0.1 Mask:255.0.0.0
UP LOOPBACK RUNNING MTU:16436 Metric:1
RX packets:2 errors:0 dropped:0 overruns:0 frame:0
TX packets:2 errors:0 dropped:0 overruns:0 carrier:0
collisions:0 txqueuelen:0
RX bytes:100 (100.0 b) TX bytes:100 (100.0 b)


lsmod

shows correct driver module is loaded (3c59x)



Module Size Used by
af_packet 24584 0
dm_mod 50328 0
fuse 35088 0
capability 7176 0
commoncap 9088 1 capability
snd_au8830 58912 0
joydev 11840 0
gameport 15240 2 snd_au8830
snd_ac97_codec 87712 1 snd_au8830
snd_pcm_oss 49696 0
snd_mixer_oss 18432 1 snd_pcm_oss
snd_pcm 77572 3 snd_au8830,snd_ac97_codec,snd_pcm_oss
parport_pc 39108 0
parport 33736 1 parport_pc
8250_pnp 12416 0
8250 47348 1 8250_pnp
serial_core 20864 1 8250
snd_timer 23684 1 snd_pcm
snd_page_alloc 11912 1 snd_pcm
snd_ac97_bus 5888 1 snd_ac97_codec
snd_mpu401_uart 9728 1 snd_au8830
snd_rawmidi 23328 1 snd_mpu401_uart
snd_seq_device 10636 1 snd_rawmidi
snd 46692 9 snd_au8830,snd_ac97_codec,snd_pcm_oss,snd_mixer_os s,snd_pcm,snd_timer,snd_mpu401_uart,snd_rawmidi,sn d_seq_device
intel_agp 22556 1
agpgart 31056 1 intel_agp
i2c_piix4 11664 0
i2c_core 20736 1 i2c_piix4
shpchp 43360 0
pci_hotplug 27460 1 shpchp
3c59x 44200 0
mii 8448 1 3c59x
soundcore 11360 1 snd
evdev 11392 0
usbhid 47584 0
pcmcia 34476 0
firmware_class 11776 1 pcmcia
yenta_socket 26764 0
rsrc_nonstatic 14464 1 yenta_socket
pcmcia_core 37528 3 pcmcia,yenta_socket,rsrc_nonstatic
apm 21996 1
unionfs 62108 1
cloop 19872 2
sbp2 24452 0
ohci1394 33716 0
ieee1394 288216 2 sbp2,ohci1394
usb_storage 70848 0
ohci_hcd 22020 0
uhci_hcd 32912 0
ehci_hcd 33160 0
usbcore 111492 6 usbhid,usb_storage,ohci_hcd,uhci_hcd,ehci_hcd


cat /proc/ioports

I looked around some more and found that /proc/ioports looked malformed? It's different from the one Debian (sarge) shows.
It looks like header lines are missing/malformed.



0000-001f : dma1
0020-0021 : pic1
0040-0043 : timer0
0050-0053 : timer1
0060-006f : keyboard
0070-0077 : rtc
0080-008f : dma page reg
00a0-00a1 : pic2
00c0-00df : dma2
00f0-00ff : fpu
0170-0177 : ide1
01f0-01f7 : ide0
02f8-02ff : serial
0320-0323 : ter_protocol
0350-0353 : ter_protocol
0376-0376 : ide1
03bc-03be : parport0
03c0-03df : vesafb
03f2-03f5 : floppy
03f6-03f6 : ide0
03f7-03f7 : floppy DIR
03f8-03ff : serial
0cf8-0cff : PCI conf1
4000-403f : 0000:00:07.3
5000-500f : 0000:00:07.3
5000-5007 : piix4_smbus
6400-641f : 0000:00:07.2
6800-687f : 0000:00:0b.0
6c00-6c07 : 0000:00:0c.0
6c00-6c07 : au8830
7000-7007 : 0000:00:0c.0
7000-7007 : au8830
e000-efff : PCI Bus #01
f000-f00f : 0000:00:07.1
f000-f007 : ide0
f008-f00f : ide1


So, is this a knoppix configuration bug? And any way around it?

[Zrebel]

'ifconfig eth0 io_addr' address

will set the start address of IO space, have you tried that to match the values? seems odd problem, I will watch for changes on my system after next reboot.

[skydan]

Thanks. Didn't know you could do that (should read the man pages more closely).
Tried "ifconfig eth0 io_addr 0x6800" and the error "io_addr: SIOCSIFMAP: Operation not supported" came back.
Same whether knoppix or root. I figure that this error is telling me that either the module has to be compiled to accept
this command or that it is not applicable after booting has completed (more likely).

It is a strange problem, and only on this machine. Might it be a tricksy BIOS? I tried the various BIOS use/ignore cheatcode
options and none had any effect. Only with 5.x too. Though <5.0 the sound card didn't work and I feel it's part of the same
problem. Any pointers as to how Knoppix does it's config magic? Then I can delve a bit deeper and try and find the cause.

[Zrebel]

The network card configure script is found at /etc/init.d/networking. All it does is call ifup wilth the --all option, maybe using the --force is required... hehe. It could also be a udev issue, try a 'noudev' boot.

[Zrebel]

It turns out you can not change the IO address or IRQ for a PCI device with io_addr. I'm thinking that the option can only be used on the first ifup call during init maybe? How did noudev workout?

[skydan]

Tried the noudev. Seemed to make things worse... NIC same as usual but now no sound and trying to start a terminal throws an error (should have written it down). Alt-F1 to the console works though.

Tried some more testing with the 5.0.1 CD on other machines. When I said that it worked on the other machines in my first post, it turns out that isn't quite true. The NIC works but the base address shown by ifconfig is incorrect. Now that I know what I'm looking for it's easy to spot! I probably checked using 3.9 and never looked further if the NIC worked on 5.0.

Anyway, I tried 5.0.1 again on the PIII BX machine and on the firewall box (K6-2, I forget the motherboard now but has VIA chipset, 2x 3com NICs). Both show incorrect base addrs from ifconfig yet the network works quite happily. There were a few quirks though (listed below). Looking through the "dmesg" output threw up an interesting result though. On <5.0 when the NIC is being configured the driver (I guess but I'm not sure what is printing this value) echoes the base address, irq etc but as a 16bit number ie 0x6800. 5.0 and 5.0.1 now display a 32bit number (clearly hex but without the 0x), with values that tie up with the address shown in ifconfig. I rebooted a few times to see how the values changed.

So, a snippet of dmesg for 5.0.1 CD on the PII machine shows:


PCI: Found IRQ 11 for device 0000:00:0b.0
3c59x: Donald Becker and others. www .scyld.com/network/vortex.html
0000:00:0b.0: 3Com PCI 3c900 Cyclone 10Mbps TPC at d0d3e000. Vers LK1.1.19

The "at d0d3e000" used to be "at 0x6800" and in this case the base address in ifconfig is 0xe000 ie the lower 16 bits of d0d3e000.

On the PIII machine (3c900B) the 32bit value shown was d0dba000. ifconfig shows a000. IO addr should be 0x6800.
NIC works, DHCP everything. Has sound. No quirks apparent.

The K6 machine has two 3com cards: eth0=3c900, eth1=3c905B.
eth0 base addr should be 0xe800. 32bit value shown is 0001e800. Was same for all reboots.
eth1 base addr should be 0xec00. 32bit value shown has been c8c90000 and c8cac000. Probably show other values for more reboots.
eth1 picked up dhcp from modem and worked normally. Manual coniguration of eth0 showed it working.
Also on this machine the boot sometimes, that is not everytime, crashes near the end. KDE throws up an error and there are no desktop icons or Konqueror window. Sound card (SB AWE64 - ok, firewall doesn't need soundcard I just couldn't be hassled removing it!) is detected but does not work. Works ok on <5.0.

You may have noticed all the machines have 3com cards so it could be a problem with the driver or something. Unfortunately they are all I have so cannot test with another NIC. But considering other quirks like desktop crashes and non-working soundcard, or now working in the case of the PII machine, makes me think it is some subtle bug in the boot/config script. I don't know what displays this value so can't say if it's in the driver or the config scripts. From a programming perspective it looks like an address is being passed rather than the value of that address but that's probably too simplistic (though typos can happen). When I get some time I'll have a look at those boot scripts you mentioned.

Here's the full dmesg output from the PII machine (the one with non-working NIC) if it helps:


Linux version 2.6.15 (root@Knoppix) (gcc version 4.0.3 20060128 (prerelease) (Debian 4.0.2-8)) #7 SMP PREEMPT Sun Feb 19 23:35:17 CET 2006
BIOS-provided physical RAM map:
BIOS-e820: 0000000000000000 - 000000000009fc00 (usable)
BIOS-e820: 000000000009fc00 - 00000000000a0000 (reserved)
BIOS-e820: 00000000000f0000 - 0000000000100000 (reserved)
BIOS-e820: 0000000000100000 - 0000000010000000 (usable)
BIOS-e820: 00000000ffff0000 - 0000000100000000 (reserved)
0MB HIGHMEM available.
256MB LOWMEM available.
On node 0 totalpages: 65536
DMA zone: 4096 pages, LIFO batch:0
DMA32 zone: 0 pages, LIFO batch:0
Normal zone: 61440 pages, LIFO batch:15
HighMem zone: 0 pages, LIFO batch:0
DMI 2.0 present.
ACPI: Unable to locate RSDP
Allocating PCI resources starting at 20000000 (gap: 10000000:efff0000)
Built 1 zonelists
Kernel command line: ramdisk_size=100000 init=/etc/init lang=uk apm=power-off vga=791 initrd=minirt.gz nomce quiet BOOT_IMAGE=knoppix BOOT_IMAGE=linux
Local APIC disabled by BIOS -- you can enable it with "lapic"
mapped APIC to ffffd000 (01342000)
Initializing CPU#0
PID hash table entries: 2048 (order: 11, 32768 bytes)
Detected 400.981 MHz processor.
Using tsc for high-res timesource
Console: colour dummy device 80x25
Dentry cache hash table entries: 65536 (order: 6, 262144 bytes)
Inode-cache hash table entries: 32768 (order: 5, 131072 bytes)
Memory: 253472k/262144k available (2120k kernel code, 8128k reserved, 776k data, 296k init, 0k highmem)
Checking if this processor honours the WP bit even in supervisor mode... Ok.
Calibrating delay using timer specific routine.. 803.38 BogoMIPS (lpj=1606771)
Security Framework v1.0.0 initialized
SELinux: Disabled at boot.
Mount-cache hash table entries: 512
CPU: After generic identify, caps: 0183f9ff 00000000 00000000 00000000 00000000 00000000 00000000
CPU: After vendor identify, caps: 0183f9ff 00000000 00000000 00000000 00000000 00000000 00000000
CPU: L1 I cache: 16K, L1 D cache: 16K
CPU: L2 cache: 512K
CPU: After all inits, caps: 0183f9ff 00000000 00000000 00000040 00000000 00000000 00000000
mtrr: v2.0 (20020519)
Enabling fast FPU save and restore... done.
Checking 'hlt' instruction... OK.
Checking for popad bug... OK.
CPU0: Intel Pentium II (Deschutes) stepping 01
SMP motherboard not detected.
Local APIC not detected. Using dummy APIC emulation.
Brought up 1 CPUs
checking if image is initramfs...it isn't (no cpio magic); looks like an initrd
Freeing initrd memory: 813k freed
NET: Registered protocol family 16
EISA bus registered
PCI: PCI BIOS revision 2.10 entry at 0xfaf20, last bus=1
PCI: Using configuration type 1
ACPI: Subsystem revision 20050902
ACPI: Interpreter disabled.
Linux Plug and Play Support v0.97 (c) Adam Belay
pnp: PnP ACPI: disabled
PnPBIOS: Scanning system for PnP BIOS support...
PnPBIOS: Found PnP BIOS installation structure at 0xc00fbae0
PnPBIOS: PnP BIOS version 1.0, entry 0xf0000:0xbb08, dseg 0xf0000
PnPBIOS: 15 nodes reported by PnP BIOS; 15 recorded by driver
SCSI subsystem initialized
PCI: Probing PCI hardware
PCI: Probing PCI hardware (bus 00)
PCI quirk: region 4000-403f claimed by PIIX4 ACPI
PCI quirk: region 5000-500f claimed by PIIX4 SMB
Boot video device is 0000:01:00.0
PCI: Using IRQ router PIIX/ICH [8086/7110] at 0000:00:07.0
PCI: Bridge: 0000:00:01.0
IO window: e000-efff
MEM window: de000000-dfffffff
PREFETCH window: ae000000-afffffff
audit: initializing netlink socket (disabled)
audit(1153631017.840:1): initialized
Total HugeTLB memory allocated, 0
VFS: Disk quotas dquot_6.5.1
Dquot-cache hash table entries: 1024 (order 0, 4096 bytes)
Initializing Cryptographic API
io scheduler noop registered
io scheduler anticipatory registered
io scheduler deadline registered
io scheduler cfq registered
Limiting direct PCI/PCI transfers.
vesafb: framebuffer at 0xae000000, mapped to 0xd0880000, using 3072k, total 4096k
vesafb: mode is 1024x768x16, linelength=2048, pages=0
vesafb: protected mode interface info at c000:042b
vesafb: scrolling: redraw
vesafb: Truecolor: size=0:5:6:5, shift=0:11:5:0
vesafb: Mode is VGA compatible
Console: switching to colour frame buffer device 128x48
fb0: VESA VGA frame buffer device
isapnp: Scanning for PnP cards...
isapnp: No Plug & Play device found
Real Time Clock Driver v1.12
PNP: PS/2 Controller [PNP0303,PNP0f13] at 0x60,0x64 irq 1,12
serio: i8042 AUX port at 0x60,0x64 irq 12
serio: i8042 KBD port at 0x60,0x64 irq 1
Floppy drive(s): fd0 is 1.44M
FDC 0 is a post-1991 82077
RAMDISK driver initialized: 16 RAM disks of 100000K size 1024 blocksize
Uniform Multi-Platform E-IDE driver Revision: 7.00alpha2
ide: Assuming 33MHz system bus speed for PIO modes; override with idebus=xx
PIIX4: IDE controller at PCI slot 0000:00:07.1
PIIX4: chipset revision 1
PIIX4: not 100% native mode: will probe irqs later
ide0: BM-DMA at 0xf000-0xf007, BIOS settings: hda:pio, hdb:pio
ide1: BM-DMA at 0xf008-0xf00f, BIOS settings: hdc:pio, hdd:pio
Probing IDE interface ide0...
hda: QUANTUM FIREBALL CR13.0A, ATA DISK drive
hdb: TOSHIBA DVD-ROM SD-M1212, ATAPI CD/DVD-ROM drive
ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Probing IDE interface ide1...
hdc: YAMAHA CRW4416E, ATAPI CD/DVD-ROM drive
ide1 at 0x170-0x177,0x376 on irq 15
hda: max request size: 128KiB
hda: 25429824 sectors (13020 MB) w/418KiB Cache, CHS=25228/16/63
hda: cache flushes not supported
hda: hda1 hda2 < hda5 hda6 >
hdb: ATAPI 32X DVD-ROM drive, 256kB Cache
Uniform CD-ROM driver Revision: 3.20
hdc: ATAPI 16X CD-ROM CD-R/RW drive, 2048kB Cache
libata version 1.20 loaded.
mice: PS/2 mouse device common for all mice
EISA: Probing bus 0 at eisa.0
Cannot allocate resource for EISA slot 4
Cannot allocate resource for EISA slot 5
Cannot allocate resource for EISA slot 6
Cannot allocate resource for EISA slot 7
EISA: Detected 0 cards.
NET: Registered protocol family 2
IP route cache hash table entries: 4096 (order: 2, 16384 bytes)
TCP established hash table entries: 16384 (order: 5, 196608 bytes)
TCP bind hash table entries: 16384 (order: 5, 196608 bytes)
TCP: Hash tables configured (established 16384 bind 16384)
TCP reno registered
NET: Registered protocol family 1
NET: Registered protocol family 15
Using IPI No-Shortcut mode
RAMDISK: Compressed image found at block 0
VFS: Mounted root (ext2 filesystem).
input: AT Translated Set 2 keyboard as /class/input/input0
seagate: ST0x/TMC-8xx not detected.
Failed initialization of WD-7000 SCSI card!
usbcore: registered new driver usbfs
usbcore: registered new driver hub
USB Universal Host Controller Interface driver v2.3
PCI: No IRQ known for interrupt pin D of device 0000:00:07.2. Please try using pci=biosirq.
uhci_hcd 0000:00:07.2: Found HC with no IRQ. Check BIOS/PCI 0000:00:07.2 setup!
uhci_hcd 0000:00:07.2: init 0000:00:07.2 fail, -19
ohci_hcd: 2005 April 22 USB 1.1 'Open' Host Controller (OHCI) Driver (PCI)
Initializing USB Mass Storage driver...
usbcore: registered new driver usb-storage
USB Mass Storage support registered.
logips2pp: Detected unknown logitech mouse model 1
ieee1394: Initialized config rom entry `ip1394'
sbp2: $Rev: 1306 $ Ben Collins <bcollins@debian.org>
ieee1394: sbp2: Driver forced to serialize I/O (serialize_io=1)
ieee1394: sbp2: Try serialize_io=0 for better performance
input: PS/2 Logitech Mouse as /class/input/input1
Unable to identify CD-ROM format.
VFS: Can't find an ext2 filesystem on dev hda.
ISO 9660 Extensions: Microsoft Joliet Level 3
ISO 9660 Extensions: RRIP_1991A
cloop: Initializing cloop v2.03
cloop: loaded (max 8 devices)
cloop: /cdrom/KNOPPIX/KNOPPIX: 75359 blocks, 65536 bytes/block, largest block is 65562 bytes.
ISO 9660 Extensions: RRIP_1991A
cloop: losetup_file: 50406 blocks, 65536 bytes/block, largest block is 65562 bytes.
ISO 9660 Extensions: RRIP_1991A
Registering unionfs 20060221-0341
Freeing unused kernel memory: 296k freed
apm: BIOS version 1.2 Flags 0x07 (Driver version 1.16ac)
usbcore: registered new driver hiddev
usbcore: registered new driver usbhid
drivers/usb/input/hid-core.c: v2.6:USB HID core driver
pci_hotplug: PCI Hot Plug PCI Core version: 0.5
shpchp: Standard Hot Plug PCI Controller Driver version: 0.4
piix4_smbus 0000:00:07.3: Found 0000:00:07.3 device
PCI: Found IRQ 11 for device 0000:00:0b.0
3c59x: Donald Becker and others. www .scyld.com/network/vortex.html
0000:00:0b.0: 3Com PCI 3c900 Cyclone 10Mbps TPC at d0d3e000. Vers LK1.1.19
Linux agpgart interface v0.101 (c) Dave Jones
agpgart: Detected an Intel 440BX Chipset.
agpgart: AGP aperture is 128M @ 0xe0000000
Serial: 8250/16550 driver $Revision: 1.90 $ 4 ports, IRQ sharing disabled
serial8250: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
serial8250: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
00:0b: ttyS0 at I/O 0x3f8 (irq = 4) is a 16550A
00:0e: ttyS1 at I/O 0x2f8 (irq = 3) is a 16550A
parport: PnPBIOS parport detected.
parport0: PC-style at 0x3bc, irq 7 [PCSPP,TRISTATE]
PCI: Found IRQ 10 for device 0000:00:0c.0
Vortex: init.... <6>done.
gameport: AU88x0 Gameport is pci0000:00:0c.0/gameport0, speed 1612kHz
Capability LSM initialized
fuse init (API version 7.3)
device-mapper: 4.4.0-ioctl (2005-01-12) initialised: dm-devel @redhat.com
PCI: Found IRQ 11 for device 0000:00:0b.0
NET: Registered protocol family 17
NET: Registered protocol family 10
lo: Disabled Privacy Extensions
ADDRCONF(NETDEV_UP): eth0: link is not ready
IPv6 over IPv4 tunneling driver
vortex: IRQ fifo error

[skydan]

Tried noudev again on the PII machine and had the desktop crash again with signal 6 (SIGABRT). just before knoppix finishes booting. Last time was on K6 machine. Was testing "noudev" at the time but don't feel it's connected. Looking at /home/knoppix/Desktop shows the files used for the icons, just no icons shown. Rebooted and tried noudev again and no crash. With noudev and attempting to open terminal window says it is unable to open a PTY etc.

Here is the traceback of the desktop crash if it is useful: (or should this be sent to the knoppix bug list?)

Using host libthread_db library "/lib/tls/libthread_db.so.1".
[Thread debugging using libthread_db enabled]
[New Thread -1231890752 (LWP 1825)]
[KCrash handler]
#5 0xb7d4e7c7 in raise () from /lib/tls/libc.so.6
#6 0xb7d5006b in abort () from /lib/tls/libc.so.6
#7 0xb7f43654 in __gnu_cxx::__verbose_terminate_handler ()
from /usr/lib/libstdc++.so.6
#8 0xb7f41045 in std::set_unexpected () from /usr/lib/libstdc++.so.6
#9 0xb7f41082 in std::terminate () from /usr/lib/libstdc++.so.6
#10 0xb7f411ba in __cxa_throw () from /usr/lib/libstdc++.so.6
#11 0xb7f415fb in operator new () from /usr/lib/libstdc++.so.6
#12 0xb7f416cd in operator new[] () from /usr/lib/libstdc++.so.6
#13 0xb75dfe01 in QString::setLength () from /usr/lib/libqt-mt.so.3
#14 0xb75e83e3 in operator>> () from /usr/lib/libqt-mt.so.3
#15 0xb7b1a88a in KIO::Slave::createSlave () from /usr/lib/libkio.so.4
#16 0xb7b1baa5 in KIO::Scheduler::createSlave () from /usr/lib/libkio.so.4
#17 0xb7b3f317 in KIO::Scheduler::startJobDirect () from /usr/lib/libkio.so.4
#18 0xb7b3f72c in KIO::Scheduler::startStep () from /usr/lib/libkio.so.4
#19 0xb7b3f83c in KIO::Scheduler::qt_invoke () from /usr/lib/libkio.so.4
#20 0xb72c354b in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#21 0xb72c3fdc in QObject::activate_signal () from /usr/lib/libqt-mt.so.3
#22 0xb76572ae in QTimer::timeout () from /usr/lib/libqt-mt.so.3
#23 0xb72e84a1 in QTimer::event () from /usr/lib/libqt-mt.so.3
#24 0xb725987a in QApplication::internalNotify () from /usr/lib/libqt-mt.so.3
#25 0xb7259a76 in QApplication::notify () from /usr/lib/libqt-mt.so.3
#26 0xb793002e in KApplication::notify () from /usr/lib/libkdecore.so.4
#27 0xb71eb001 in QApplication::sendEvent () from /usr/lib/libqt-mt.so.3
#28 0xb724b305 in QEventLoop::activateTimers () from /usr/lib/libqt-mt.so.3
#29 0xb71fed2a in QEventLoop::processEvents () from /usr/lib/libqt-mt.so.3
#30 0xb7272255 in QEventLoop::enterLoop () from /usr/lib/libqt-mt.so.3
#31 0xb727217a in QEventLoop::exec () from /usr/lib/libqt-mt.so.3
#32 0xb725838d in QApplication::exec () from /usr/lib/libqt-mt.so.3
#33 0xb68c112f in kdemain () from /usr/lib/libkdeinit_kdesktop.so
#34 0xb7fbc524 in kdeinitmain () from /usr/lib/kde3/kdesktop.so
#35 0x0804e1d6 in ?? ()
#36 0x00000001 in ?? ()
#37 0x080d03e8 in ?? ()
#38 0x00000001 in ?? ()
#39 0x00000000 in ?? ()

[skydan]

Had a closer look at how knoppix autoconfigures the hardware using the 5.0.1 CD. Starting point was "/etc/init.d/knoppix-autoconfig" that basically does it all. The one command hwsetup does all the magic though the script shows you can specify "nohwsetup" from the boot prompt to switch to using "etc/modules" instead, whatever that is. I tried both and there was no change.

To track down exactly where the information in dmesg was coming from I tried the following:
1. booted using "knoppix emergency" from the boot prompt to get a bare system without running any scripts.
2. mount -t proc /proc /proc to mount /proc.
3. mount -o remount,rw / to remount root as read-write.
4. dmesg shows no network card information. Only loopback shown in ifconfig -a.
5. modprobe -v 3c59x fires up the driver for the NIC and echoes:

insmod /lib/modules/2.6.17/kernel/drivers/net/mii.ko
insmod /lib/modules/2.6.17/kernel/drivers/net/3c59x.ko
3c59x: Donald Becker and others. www .scyld.com/network/vortex.html
6. dmesg now shows netword card information:

PCI: Found IRQ 11 for device 0000:00:0b.0
3c59x: Donald Becker and others. www .scyld.com/network/vortex.html
0000:00:0b.0: 3Com PCI 3c900 Cyclone 10Mbps TPC at d085a000. Vers LK1.1.19
and ifconfig -a shows eth0 with base address=0xa000

So the bad address would appear to be from modprobe, the 3c59x driver and/or the 2.6.17 kernel (and 2.6.15 from 5.0).
I figure that modprobe and the driver haven't changed much yet the kernel is bleeding(ish) edge? Am I correct in guessing
that the first and last lines shown in (6) are from the kernel? I'm not familiar with the intricacies of the kernel so I think this
is about as far as I can take it. Maybe a kernel hacker can expand on this a bit more.

[skydan]

Oh, something I forgot to ask. Only having 3Com cards I can't test if it is driver related or a problem in the kernel.
Does anyone else see anything like this in the dmesg output and from ifconfig even if the NIC appears to work?
If so is it something other than a 3Com NIC?

[skydan]

Update. On the PII BX machine, the one with the unresponsive NIC under 5.x, I flattened a windows partition, installed Debian woody kernel 2.2.20 and upgraded it to testing/etch kernel 2.6.16-17. This displays exactly the same results as knoppix, so it's not something that's been added by knoppix. Something new from dmesg is the error "ADDRCONF(NETDEV_UP): eth0: link is not ready". Also the address displayed by the 3c59x driver is always the same (c081c000) on every boot.

I see a lot of people have read this thread but no replies. Has no-one else has seen this or am I the only one foolish enough to attempt to use Linux with 3Com NICs?! Is it possible I have the only three machines that exhibit this behaviour? ;) Is the 3Com driver the only one to display this address? I'd almost buy another (non-3Com) NIC just to find some answers to these questions. Though I now believe it's a kernel and/or driver issue so probably really fall into the Knoppix remit...

Zrebel, Klaus also said it could be a udev issue when I mentioned it to him, it being new to 5.x. I had a look at the source code for 2.6.11-9 and 2.6.17-7 and the 3c59x driver has changed a bit, specifically in the area of the io base address which might explain it. The top 16 bits of the address reported by the driver look suspiciously like a memory address (lower 16 bits do look like io addr) and is within the range of addresses of the driver as shown by "cat /proc/kallsyms|grep 3c59x". I don't know the internal gubbins of Linux that well but the driver code looks ok. lspci, hwsetup, /proc/pci all report the same (correct) io address so it would seem like the PCI settings are being mangled somewhere between the kernel and the driver. modprobe?

As this is getting OT for knoppix where is the best place to report this? debian.org?

[skydan]

Update 2. Tested PII BX machine with Knoppix 4.0.2 CD (I should have realised I'd not tested this earlier). Odd results. The NIC fails to function exactly the same way as later kernels but the io base address shown by the 3c59x module and ifconfig is 0x6800, the normal expected address. 4.0.2 is kernel 2.6.12. Also tried Kanotix 04-2005 and Helix 1.7. Both appear to use the same 2.6.14-kanotix-9 kernel. NIC fails to functions as before and the address shown is the strange 32bit one. I tried Debian kernels 2.6.15-1-486, in case it was a 686 quirk, upto the latest 2.6.17-1-686. All fail the same way as previously described. I read that someone was having problems with drivers compiled using gcc 4 (I think it was this forum but can't find it again) so I recompiled the 2.6.17 kernel using gcc 3.3.6. Exact same results.

I'm beginning to think that something broke around 2.6.12 that stopped this machine working. After that something else caused the strange 32bit address thing but it isn't enough to stop my other machines from working. That is, the two things may not be necessarily connected.

I've taken this much further that I originally expected. I can't do much more unless I delve into the source code. 'Spose it's one way to get to know Linux better...
Still on-one else with something similar?