Re: Interrupt time inflation on Xen

[David Brownlee <abs%absd.org@localhost>]

On 22 March 2016 at 14:47, Thor Lancelot Simon <tls%panix.com@localhost> wrote:
> The controller is a ciss.  I observed the same problems with an arcmsr.
>
> I don't think this has anything to do with the controller.  At 1/2 the
> data and request rate, under Xen we use 60% CPU as interrupt time
> while with a native kernel, it's 1-3%.
>
> The same test's easy to reproduce with any Xen dom0, since all you have
> to do is dd from the raw disk into /dev/null.  Do you see >40% interrupt
> time and all your idle CPU go away?  That's the issue.

In case it helps, with a recent netbsd-7 dom0 kernel I see
- around 20% interrupt, 70% idle with dd if=/dev/wd1d of=/dev/null bs=64k
- around 4% interrupt, 80% idle with dd if=/dev/rwd1d of=/dev/null bs=64k

wd1 is an SSD, but I get ~ the same overhead with the spinning rust on wd0

System is a Dell Optiplex 990 SFF 3.4Ghz i7, xen4.5

I will note that I have found the machine laggy if I ran pkgsrc builds in dom0

Kernel is a generic plus npf - dmest at http://sync.absd.org/hermes.dmesg