High interrupt CPU in top...

[Peter Harrison]

On 28 October 2014 23:19, Dutch Ingraham <stoa at gmx.us> wrote:

> On 10/28/14 18:37, Peter Harrison wrote:
>
> >
> > Here's vmstat-i with Xorg running:
> >
> >  interrupt                          total       rate
> > irq1: atkbd0                         139          1
> > irq9: acpi0                          835          7
> > irq12: psm0                         1746         15
> > irq16: uhci3                    20622511     177780
> > irq20: hpet0 uhci0                118696       1023
> > irq21: uhci1                          11          0
> > irq23: ehci0                           2          0
> > irq257: hdac0                       3495         30
> > irq258: iwn0                       13406        115
> > irq259: ahci0                       8639         74
> > irq260: vgapci0                     1979         17
> > Total                           20771459     179064
> >
> > And without Xorg running:
> >
> > interrupt                          total       rate
> > irq1: atkbd0                         150          3
> > irq9: acpi0                          378          9
> > irq20: hpet0 uhci0                  8303        202
> > irq21: uhci1                          11          0
> > irq23: ehci0                           2          0
> > irq257: hdac0                         61          1
> > irq258: iwn0                        1536         37
> > irq259: ahci0                       1634         39
> > Total                              12075        294
> >
> > So X is the problem? What can I do to fix it?
> >
> > Cheers,
> >
> >
> >
> > Peter Harrison.
> > _______________________________________________
> > freebsd-questions at freebsd.org mailing list
> > http://lists.freebsd.org/mailman/listinfo/freebsd-questions
> > To unsubscribe, send any mail to "
> freebsd-questions-unsubscribe at freebsd.org"
> >
>
>
> Try adding the kernel tunable "hw.drm.msi=0" (without quotes) to your
> /boot/loader.conf file and reboot.  Check vmstat -i with xorg running.
> If it works, fine - that addition will persist.  Of course, if it
> doesn't work, remove that line.
>
>
That looks like it did the trick - thanks!


Peter Harrison.