[E1000-devel] e1000 linux driver problems

[popescu c. <z3r...@gm...>]

Hi,

After some time of googling and reading i can't find a solution to high CPU
usage on high amount of packets.
My server:

admin linux # lspci
00:00.0 Host bridge: Broadcom CMIC-LE Host Bridge (GC-LE chipset) (rev 32)
00:00.1 Host bridge: Broadcom CMIC-LE Host Bridge (GC-LE chipset)
00:00.2 Host bridge: Broadcom CMIC-LE Host Bridge (GC-LE chipset)
00:03.0 VGA compatible controller: ATI Technologies Inc Rage XL (rev 27)
00:04.0 RAID bus controller: Compaq Computer Corporation Smart Array 5i/532
(rev 01)
00:05.0 System peripheral: Compaq Computer Corporation Integrated Lights Out
Controller (rev 01)
00:05.2 System peripheral: Compaq Computer Corporation Integrated Lights
Out  Processor (rev 01)
00:0f.0 ISA bridge: Broadcom CSB5 South Bridge (rev 93)
00:0f.1 IDE interface: Broadcom CSB5 IDE Controller (rev 93)
00:0f.2 USB Controller: Broadcom OSB4/CSB5 OHCI USB Controller (rev 05)
00:0f.3 Host bridge: Broadcom CSB5 LPC bridge
00:11.0 Host bridge: Broadcom CIOB-X2 PCI-X I/O Bridge (rev 05)
00:11.2 Host bridge: Broadcom CIOB-X2 PCI-X I/O Bridge (rev 05)
01:01.0 Ethernet controller: Intel Corporation 82543GC Gigabit Ethernet
Controller (Copper) (rev 02)
01:02.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5703X Gigabit
Ethernet (rev 02)
04:01.0 Ethernet controller: Intel Corporation 82543GC Gigabit Ethernet
Controller (Copper) (rev 02)
04:02.0 Ethernet controller: Broadcom Corporation NetXtreme BCM5703X Gigabit
Ethernet (rev 02)

The CPUs are:
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 2
model name      : Intel(R) Xeon(TM) CPU 2.80GHz
stepping        : 7
cpu MHz         : 2799.297
cache size      : 512 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 1
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe cid xtpr
bogomips        : 5601.90
clflush size    : 64

In total 2 CPUs with 2 cores each ( 4 cores in total )

( i suspect the network cards, maybe they are to old, anyway i am
configuring a new server, the same like above but with 2x  01:01.0 Ethernet
controller: Intel Corporation 82545GM Gigabit Ethernet Controller (rev 04),
till then i wanna know what is wrong with the actual one )

More information:

admin linux # ethtool -g eth1
Ring parameters for eth1:
Pre-set maximums:
RX:             256
RX Mini:        0
RX Jumbo:       0
TX:             256
Current hardware settings:
RX:             256
RX Mini:        0
RX Jumbo:       0
TX:             256

(eth1 is one of the intel devices connected to a 1gb high performance
switch)

admin linux # ethtool -S eth1
NIC statistics:
     rx_packets: 47413772
     tx_packets: 36393608
     rx_bytes: 39115951199
     tx_bytes: 25299823783
     rx_broadcast: 60490
     tx_broadcast: 111399
     rx_multicast: 78
     multicast: 78
     rx_no_buffer_count: 37186
     rx_missed_errors: 1735
     tx_restart_queue: 18
     rx_long_byte_count: 39115951199
     rx_csum_offload_good: 47321361
     rx_csum_offload_errors: 78

This are the fields witch have valors not equal with 0, there were more
fields but with 0.

admin linux # cat /proc/interrupts
           CPU0       CPU1       CPU2          CPU3
  0:         90          0          0          0      IO-APIC-edge
timer
  1:         22          0        903          0      IO-APIC-edge
i8042
  6:          3          0          0          0      IO-APIC-edge
floppy
  8:          2          0          0          0      IO-APIC-edge      rtc
  9:          0          0          0          0      IO-APIC-fasteoi   acpi
 12:        113          0          0          0      IO-APIC-edge
i8042
 14:      33838     382488     230378      45544      IO-APIC-edge      ide0
 16:       1116      22101      20589       6867      IO-APIC-fasteoi
cciss0
 17:      77866   46287144    2212068    1405787      IO-APIC-fasteoi   eth2
 19:    1255717    7308389  242756519   28949634      IO-APIC-fasteoi   eth0
 20:        100    7306079   34092280  274598160      IO-APIC-fasteoi   eth1
NMI:          0          0          0          0
LOC:    6100900    4647630    2510178    1482718
ERR:          0
MIS:          0

On high traffic the cpus that are assigned for eth0 and eth1 are used from
80% to 100% resulting in over 150ms response time isted of 1ms from a
computer connected to the same switch as the server.

I am using Gentoo 2007.0 always up to date, running services on this server:
bgpd ( 4200 routes in total ), named, dhcp ( about 800 macs ) and iptables (
1700+ rules ).
I don't think my problem is from this services, i suspect a high amount of
packages sent by virused computers connected to my local network or
something like this.
I am using irqbalance daemon from irqbalance.org insted of the kernel
irqbalance, the e1000 driver is 7.6.9 installed with NAPI enabled and the
module was loaded with modprobe e1000 InterruptThrottleRate=20000,20000

Any kind of help is kingly welcomed.
Thank you