[CentOS] Centos 5: e1000 question

Mogens Kjaer mk at crc.dk
Mon Jun 25 08:47:50 UTC 2007


I have a machine running Centos 5, x86_64.

It has an Intel NIC:

# lspci
15:00.0 Ethernet controller: Intel Corporation 82572EI Gigabit Ethernet 
Controller (Fiber) (rev 06)

The machine has two dualcore cpus.

If I look at /proc/interrupts I get:

            CPU0       CPU1       CPU2       CPU3
   0:  144873228  220741831  173615012  220733357    IO-APIC-edge  timer
   4:     269120    1203485    1964914    1528555    IO-APIC-edge  serial
   7:          0          0          0          0    IO-APIC-edge  parport0
   8:          0          0          0          0    IO-APIC-edge  rtc
   9:          0          0          0          0   IO-APIC-level  acpi
  14:        137    3218764    3143397     430444    IO-APIC-edge  ide0
  90:    3403812    5504380     553784    1273225   IO-APIC-level  aic79xx
  98:  460440049          0          0          0         PCI-MSI  eth0
169:     210489    9211230   26272710   31733980   IO-APIC-level 
uhci_hcd:usb2, uhci_hcd:usb3, aacraid
177:          0          0          0          0   IO-APIC-level 
uhci_hcd:usb4
185:          2          0          0          0   IO-APIC-level 
uhci_hcd:usb1, ehci_hcd:usb5
NMI:        359        102        105        115
LOC:  744651234  743298229  744651090  743298086
ERR:          0
MIS:          0

My question is: Why are the eth0 interrupts distributed
to CPU0 only?

Mogens
-- 
Mogens Kjaer, Carlsberg A/S, Computer Department
Gamle Carlsberg Vej 10, DK-2500 Valby, Denmark
Phone: +45 33 27 53 25, Fax: +45 33 27 47 08
Email: mk at crc.dk Homepage: http://www.crc.dk



More information about the CentOS mailing list