em interrupt storm

Barney Wolff barney at databus.com
Thu Nov 24 07:14:08 GMT 2005


I've observed the interrupt aliasing problem on an Asus A7M266-D
with 2 Athlon 2200-MPs, so it's not confined to the Intel chipset.
Here's the evidence (there is nothing connected to ehci0):
lab:/home/barney $ vmstat -i
interrupt                          total       rate
irq6: fdc0                            10          0
irq13: npx0                            1          0
irq14: ata0                     12380884          6
irq15: ata1                           58          0
irq16: ehci0                   212494994        109
irq17: pcm0                      4551325          2
irq18: ohci1                     4275754          2
irq19: fxp0 ohci0+                124765          0
irq20: em0                     210064498        108
irq21: ahc0                      4245565          2
irq22: ahc1                      4261858          2
cpu0: timer                   3877840775       1999
cpu1: timer                   3877813525       1999
Total                         8208054012       4233
lab:/home/barney $ uname -a
FreeBSD lab.databus.com 7.0-CURRENT FreeBSD 7.0-CURRENT #1: Thu Oct 20 16:16:12 EDT 2005     toor at lab.databus.com:/usr/obj/usr/src/sys/LAB  i386

It's been there for a long time - I just assumed it was a hw issue.

-- 
Barney Wolff         http://www.databus.com/bwresume.pdf
I never met a computer I didn't like.


More information about the freebsd-current mailing list