From Bugzilla Helper: User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.6) Gecko/20040510 Description of problem: The SMC Ether Power 10Mbs NIC is detected and configured to use tulip.o and works properly under FC1. Under FC2 it is detected and configured to use tulip.ko and fails: Bringing up interface eth0: tulip device eth0 does not seem to be present, delaying initialization. When eth0 is modprobed no eth0 entry is added to the links. [root@twinkie root]# ip link 1: lo: <LOOPBACK,UP> mtu 16436 qdisc noqueue link/loopback 00:00:00:00:00:00 brd 00:00:00:00:00:00 2: sit0: <NOARP> mtu 1480 qdisc noop link/sit 0.0.0.0 brd 0.0.0.0 FC2 has several tulip driver files de2104x.ko, de4x5.ko ... The NIC uses the DEC 21041 chip. I tried the de2104x.ko driver. It fails with a kernel panic. The de4x5.ko driver seems to work. [root@twinkie root]# /etc/rc.d/init.d/network restart Shutting down loopback interface: [ OK ] Setting network parameters: [ OK ] Bringing up loopback interface: [ OK ] Bringing up interface eth0: tulip device eth0 does not seem to be present, delaying initialization. [FAILED] [root@twinkie root]# vi /etc/modprobe.conf changed "alias eth0 tulip" to "alias eth0 de4x5" [root@twinkie root]# /etc/rc.d/init.d/network restart Shutting down loopback interface: [ OK ] Setting network parameters: [ OK ] Bringing up loopback interface: [ OK ] Bringing up interface eth0: [ OK ] root@twinkie root]# lspci -v ..... 00:0e.0 Ethernet controller: Digital Equipment Corporation DECchip 21041 [Tulip Pass 3] (rev 21) Flags: bus master, medium devsel, latency 32, IRQ 5 I/O ports at d400 Memory at dc140000 (32-bit, non-prefetchable) [size=128] ..... EITHER the wrong driver is being configured OR the tulip.ko driver has a problem with this NIC Version-Release number of selected component (if applicable): How reproducible: Always Steps to Reproduce: 1. Do a complete install of FC2 2. Reboot 3. Actual Results: Bringing up interface eth0: tulip device eth0 does not seem to be present, delaying initialization. [FAILED] Expected Results: Bringing up interface eth0: [ OK ] Additional info:
I reported the same on the LKML, subject was: "DEC 2104x driver confusion with 2.6.5+" Machine parameters I installed FC2/i386: Pentium 166MMX, PCI mainboard USB Two SMC branded ethernet card: # lspci ... 00:0f.0 Ethernet controller: Digital Equipment Corporation DECchip 21041 Tulip Pass 3] (rev 11) 00:10.0 Ethernet controller: Digital Equipment Corporation DECchip 21041 [Tulip Pass 3] (rev 11) # lspci -n -vvv -s 00:0f.0 00:0f.0 Class 0200: 1011:0014 (rev 11) 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: 96 Interrupt: pin A routed to IRQ 10 Region 0: I/O ports at fc00 Region 1: Memory at fedffc00 (32-bit, non-prefetchable) [size=128] # lspci -n -vvv -s 00:10.000:10.0 Class 0200: 1011:0014 (rev 11) 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: 96 Interrupt: pin A routed to IRQ 11 Region 0: I/O ports at f880 Region 1: Memory at fedff800 (32-bit, non-prefetchable) [size=128] One and a half year ago I installed RH9 on this machine that acts as a firewall. RH9 tried to drive those with "tulip_old.o" but id didn't work. "tulip.o" needed setting up media type in modules.conf but then it worked 7x24 until the harddisk died. FC2 loaded "tulip.ko" but it did not work. I tried "modprobe de2104x.ko" but it oopsed approximately 2 minutes later with this: Jun 6 17:49:48 localhost kernel: irq 11: nobody cared! (screaming interrupt?) Jun 6 17:49:48 localhost kernel: Call Trace: Jun 6 17:49:48 localhost kernel: [<c0106ee5>] __report_bad_irq+0x2b/0x67 Jun 6 17:49:48 localhost kernel: [<c0106f7d>] note_interrupt+0x43/0x66 Jun 6 17:49:48 localhost kernel: [<c0107143>] do_IRQ+0x109/0x169 Jun 6 17:49:48 localhost kernel: [<c0287338>] common_interrupt+0x18/0x20 Jun 6 17:49:48 localhost kernel: [<c011c254>] __do_softirq+0x2c/0x73 Jun 6 17:49:48 localhost kernel: [<c0107708>] do_softirq+0x46/0x4d Jun 6 17:49:48 localhost kernel: ======================= Jun 6 17:49:48 localhost kernel: [<c0107197>] do_IRQ+0x15d/0x169 Jun 6 17:49:48 localhost kernel: [<c0287338>] common_interrupt+0x18/0x20 Jun 6 17:49:48 localhost kernel: [<d091599f>] de_set_rx_mode+0xb/0xd [de2104x] Jun 6 17:49:48 localhost kernel: [<d0916339>] de_init_hw+0x68/0x6e [de2104x] Jun 6 17:49:48 localhost kernel: [<d09165e2>] de_open+0x60/0xdd [de2104x] Jun 6 17:49:48 localhost kernel: [<c023b589>] dev_open+0x5f/0xc6 Jun 6 17:49:48 localhost kernel: [<c023c582>] dev_change_flags+0x45/0xe7 Jun 6 17:49:48 localhost kernel: [<c026e437>] devinet_ioctl+0x296/0x50c Jun 6 17:49:48 localhost kernel: [<c026ff3f>] inet_ioctl+0x47/0x73 Jun 6 17:49:48 localhost kernel: [<c023557d>] sock_ioctl+0x25e/0x274 Jun 6 17:49:48 localhost kernel: [<c0150576>] sys_ioctl+0x207/0x23d Jun 6 17:49:48 localhost kernel: [<c0287147>] syscall_call+0x7/0xb Jun 6 17:49:48 localhost kernel: Jun 6 17:49:48 localhost kernel: handlers: Jun 6 17:49:48 localhost kernel: [<c021995d>] (usb_hcd_irq+0x0/0x4b) Jun 6 17:49:48 localhost kernel: Disabling IRQ #11 IRQ 11 is shared between the USB host chip and eth1, in one occasion from five try the machine locked up. "de4x5.ko" drives the cards nicely BUT ONLY AFTER LOADING AND REMOVING "de2104x.ko" Rebooting into single mode and loading "de4x5.ko" does not emit any messages into dmesg. loading "de2104x.ko" gives these messages: de2104x PCI Ethernet driver v0.7 (Mar 17, 2004) PCI: Enabling device 0000:00:0f.0 (0004 -> 0007) de0: SROM leaf offset 30, default media 10baseT auto de0: media block #0: 10baseT-FD de0: media block #1: BNC de0: media block #2: 10baseT-HD divert: allocating divert_blk for eth0 eth0: 21041 at 0xd0843c00, 00:e0:29:09:84:99, IRQ 10 PCI: Enabling device 0000:00:10.0 (0004 -> 0007) de1: SROM leaf offset 30, default media 10baseT auto de1: media block #0: 10baseT-FD de1: media block #1: BNC de1: media block #2: 10baseT-HD divert: allocating divert_blk for eth1 eth1: 21041 at 0xd08a9800, 00:e0:29:09:84:ab, IRQ 11 The "PCI: Enabling..." messages are suspicious, maybe these card are disabled, that's why "de4x5.ko" did not see them by itself. After removing "de2104x.ko" and loading "de4x5.ko" I have these messages: divert: freeing divert_blk for eth0 divert: freeing divert_blk for eth1 0000:00:0f.0: DC21041 at 0xfc00, h/w address 00:e0:29:09:84:99, and requires IRQ10 (provided by PCI BIOS). de4x5.c:V0.546 2001/02/22 davies.com divert: allocating divert_blk for eth0 0000:00:10.0: DC21041 at 0xf880, h/w address 00:e0:29:09:84:ab, and requires IRQ11 (provided by PCI BIOS). de4x5.c:V0.546 2001/02/22 davies.com divert: allocating divert_blk for eth1 eth0: media is TP. eth1: media is TP. From that point, both cards operate well, the machine is not restarted since: # uptime 09:03:35 up 5 days, 13:49, 1 user, load average: 0.00, 0.00, 0.00 I have put "modprobe de2104x ; rmmod de2104x" into /etc/rc.d/init.d/network at the beginning of the "start)" section and alias eth0 de4x5 alias eth1 de4x5 into /etc/modprobe.conf. The machine brings up eth0 and eth1 as it should.
Problem continues in Fedora 3
Same problem on #fedora with a "D-Link, System Inc DE-530+" card which uses the same chipset: 00:0e.0 Ethernet controller: Digital Equipment Corporation DECchip 21041 [Tulip Pass 3] (rev 21) Using the de4x5 driver instead of tulip works. I'm assuming this person was using the kernel that shipped with FC3 (2.6.9-1.667), since it was a new install, and was unable to get on the net yet to do updates.
Fedora Core 2 has now reached end of life, and no further updates will be provided by Red Hat. The Fedora legacy project will be producing further kernel updates for security problems only. If this bug has not been fixed in the latest Fedora Core 2 update kernel, please try to reproduce it under Fedora Core 3, and reopen if necessary, changing the product version accordingly. Thank you.