e1000 kills me
 

Hi all,

First, I thank everyone who writes on this forum, it is a great source of information.

Don't even know for sure if it is a hardware issue. Anyway, recently I bought a new mobo Gigabyte GA-8KNXP and installed RH9. Everything was good and setup passed without any troubles. I compiled a last official stable kernel 2.4.21 in order to have a support for ICH5 chipset Serial ATA. After that I began to configure NIC, this Gigabyte mobo has an Intel® PRO/1000 integrated card (82547EI) and I enabled support for it as a kernel module (e1000).

Now everytime I try to modprobe e1000 system freezes completely and nothing can be done except resetting the system. Sometimes for some unexplainable reason it goes through but when I try to ifup eth0 (modules.conf contains alias eth0 e1000) it freezes again. I tried different approaches recompiled kernel for about 20 times tried two other e1000 drivers from Intel 5.0.43 and 5.1.11 result stays absolutely the same system hangs right after device is being initialized. There are few lines in /var/log/messages after modprobe e1000 but nothing informative and it ends with /etc/hotplug/net.agent: invoke ifup eth0.

When all my desperate attempts failed I added another NIC to the system and it worked like it should. Got eth0 up and pinged it OK. So, I guess after all it is a hardware issue.

How can I determine the source of that problem? Can it be NIC malfunction and how can I check this?

BTW when I searched Internet for the answer I've found out that many people have problems with that type of NICs, ehh, wish I knew that before purchasing that mobo.

Adding lspci, /var/log/messages and ifcfg-eth0 also I forgot to mention that kermel built with SMP and I tried non SMP kernel too with no luck ;).

Any advices and help are greatly appreciated.

================================
#lspci -v

00:00.0 Host bridge: Intel Corp.: Unknown device 2578 (rev 02)
Subsystem: Giga-byte Technology: Unknown device 2578
Flags: bus master, fast devsel, latency 0
Memory at e8000000 (32-bit, prefetchable) [size=128M]
Capabilities: [e4] #09 [2106]
Capabilities: [a0] AGP version 3.0

00:01.0 PCI bridge: Intel Corp.: Unknown device 2579 (rev 02) (prog-if 00 [Normal decode])
Flags: bus master, 66Mhz, fast devsel, latency 64
Bus: primary=00, secondary=01, subordinate=01, sec-latency=32
Memory behind bridge: f2000000-f3ffffff
Prefetchable memory behind bridge: f0000000-f1ffffff

00:03.0 PCI bridge: Intel Corp.: Unknown device 257b (rev 02) (prog-if 00 [Normal decode])
Flags: bus master, 66Mhz, fast devsel, latency 32
Bus: primary=00, secondary=02, subordinate=02, sec-latency=0
I/O behind bridge: 0000a000-0000afff
Memory behind bridge: f6000000-f60fffff

00:1e.0 PCI bridge: Intel Corp. 82801BA/CA/DB PCI Bridge (rev c2) (prog-if 00 [Normal decode])
Flags: bus master, fast devsel, latency 0
Bus: primary=00, secondary=03, subordinate=03, sec-latency=32
I/O behind bridge: 00008000-00009fff
Memory behind bridge: f4000000-f5ffffff

00:1f.0 ISA bridge: Intel Corp. 82801EB ISA Bridge (LPC) (rev 02)
Flags: bus master, medium devsel, latency 0

00:1f.2 IDE interface: Intel Corp.: Unknown device 24d1 (rev 02) (prog-if 8a [Master SecP PriP])
Subsystem: Giga-byte Technology: Unknown device 24d1
Flags: bus master, 66Mhz, medium devsel, latency 0, IRQ 18
I/O ports at <unassigned>
I/O ports at <unassigned>
I/O ports at <unassigned>
I/O ports at <unassigned>
I/O ports at f000 [size=16]

00:1f.3 SMBus: Intel Corp. 82801EB SMBus (rev 02)
Subsystem: Giga-byte Technology: Unknown device 24d2
Flags: medium devsel, IRQ 17
I/O ports at 1400 [size=32]

01:00.0 VGA compatible controller: nVidia Corporation NV5 [RIVA TNT2/TNT2 Pro] (rev 11) (prog-if 00 [VGA])
Subsystem: Asustek Computer, Inc. AGP-V3800 SGRAM
Flags: bus master, 66Mhz, medium devsel, latency 32, IRQ 16
Memory at f2000000 (32-bit, non-prefetchable) [size=16M]
Memory at f0000000 (32-bit, prefetchable) [size=32M]
Expansion ROM at <unassigned> [disabled] [size=64K]
Capabilities: [60] Power Management version 1
Capabilities: [44] AGP version 2.0

02:01.0 Ethernet controller: Intel Corp.: Unknown device 1019
Subsystem: Intel Corp.: Unknown device 1019
Flags: bus master, 66Mhz, medium devsel, latency 0, IRQ 18
Memory at f6000000 (32-bit, non-prefetchable) [size=128K]
I/O ports at a000 [size=32]
Capabilities: [dc] Power Management version 2

=======================================
/var/log/messages


Jul 13 21:06:36 localhost kernel: Intel(R) PRO/1000 Network Driver - version 5.1.11
Jul 13 21:06:36 localhost kernel: Copyright (c) 1999-2003 Intel Corporation.
Jul 13 21:06:36 localhost kernel: eth0: Intel(R) PRO/1000 Network Connection
Jul 13 21:06:36 localhost kernel: AutoNeg advertising 10/HD
Jul 13 21:06:36 localhost /etc/hotplug/net.agent: invoke ifup eth0

or just

Jul 14 00:16:29 localhost kernel: Intel(R) PRO/1000 Network Driver - version 5.0.43-k1
Jul 14 00:16:29 localhost kernel: Copyright (c) 1999-2003 Intel Corporation.
Jul 14 00:16:29 localhost kernel: eth0: Intel(R) PRO/1000 Network Connection

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

I tried some passing some parameters to e1000 module.

=======================================
/etc/sysconfig/ifcfg-eth0

DEVICE=eth0
ONBOOT=yes
BOOTPROTO=static
IPADDR=10.1.1.1
NETMASK=255.255.255.0
GATEWAY=10.255.255.254

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

It has been couple of days since I started to wage against e1000 and now I ask local hardware gurus to help me. What can I do to make that beast work or at least how can I get more information about the problem :scratch: ? Thank you.

gyrus,
How about we go for the obvious: have you recompiled your kernel with the e1000 support in the kernel rather than as a module?

Oops. Just checked and I see that, due to licensing, the e1000 is not able to be compiled into the kernel. Bummer.
But there due appear to be several forums where folks talk about e1000 support. I googled for "linux e1000 intel network card" and found refs, though on RH 7.3, not 9.X.
Interesting that a vanilla 2.4.21 also has problems; rules out RH specific patches. I'll snoop around as well and see if I find anything; I am considering buying the same MoBo.

gyrus,

I note that on Intel's official page regarding support for this card under Linux, they mention kernel support upto and including 2.4.18:

http://www.intel.com/support/network...inux/e1000.htm

So perhaps your kernel version is problematic after all.

Here's a couple of places from your lspci output which I suspect could cause you problems:AFAIK, Linux has problems with IRQs above 16. Try disabling APIC in the BIOS, or if that is not possible, include 'noapic' in the 'append =' line in lilo.conf. If, for example, your /etc/lilo.conf file has a line that says:

append = "quiet devfs=mount hdc=ide-scsi acpi=off"

then edit to read:

append = "quiet noapic devfs=mount hdc=ide-scsi acpi=off"

Then, in a console, as root, run the command:

lilo

This will make the changes to your bootloader that turns off APIC in the hardware configuration. Then re-boot your computer. Your lspci output should then see no IRQs above 16. Hopefully, this will make your hardware behave correctly and maybe fix your problem.

<Edit> For correct hardware detection, you also need to make sure that Plug N Play OS is disabled in the BIOS.

e1000 kills me
 

Great answer ... it helps me solve my problem, with
e1000 card not working.

got everything in order, insmod driver installed,
/sbin/route looks good, /sbin/ifconfig show eth0 eth1 eth2
active with correct IP, /etc/modules.conf show all aliases,
/etc/sysconfig/network-scripts/ifcfg-eth_ all seems inorder,
/sbin/ip -s neigh looks normal. 'ethereal' detect cable
connect and disconnect correctly. 'neat' work just fine.
All firewall off.

All of that, and I couldn't ping anything beyond the local
address. Just kills me !!!!

I read geoff_f's post and double check IRQ. Sure enough
lspci -vv show the card is using IRQ 42/43. So I stick in
the noapic boot option in /etc/grub.conf and hard reset the
machine ... boom ... it works.

Thanks geoff_f, you're king !!!!!

ETcute :-)

Ignore... I didn't realize this thread is three years old.

You're welcome etcute. It's gratifying to know that even after three years, someone can still benefit from the help given here on LQ. Many thanks for the feedback.