RELENG_7: interrupt eating whole cpu core
Dominic Fandrey
kamikaze at bsdforen.de
Wed Feb 6 18:33:57 UTC 2008
While ripping DVDs the irq14: ata0 eats ~95% of one of my cores. The channel
is exclusive to the drive.
# atacontrol list 64 /root
ATA channel 0:
Master: acd0 <Optiarc DVD RW AD-7560A/DH10> ATA/ATAPI revision 7
Slave: no device present
ATA channel 1:
Master: no device present
Slave: no device present
ATA channel 2:
Master: ad4 <ST9160821AS/3.BHE> Serial ATA v1.0
Slave: no device present
ATA channel 3:
Master: no device present
Slave: no device present
ATA channel 4:
Master: no device present
Slave: no device present
The number of interrupts looks reasonable to me.
# vmstat -i 0 /root
interrupt total rate
irq1: atkbd0 35171 1
irq9: acpi0 15729 0
irq12: psm0 10004 0
irq14: ata0 5148195 272
irq16: pcm0 uhci0 40783 2
irq17: uhci1+ 1446238 76
irq18: bge0 ehci0+ 38588 2
irq20: uhci2 ehci1 125447 6
irq21: uhci3 156783 8
cpu0: timer 37792079 1999
cpu1: timer 37784062 1999
Total 82593079 4370
I'm currently rebuilding world+kernel. I will report weather the behaviour has
changed. This is an HP 6510b GR695EA#ABD, if anyone thinks it might be
helpful, I can supply you with a dmesg and the output of pciconf -lv.
More information about the freebsd-stable
mailing list