If I'm not misled, intrd is a perl script that monitors kstats to discover
and correct imbalances of interrupt distribution between devices and
CPUs; you should be able to look at it to get some idea of what it's doing or 
why.

It looks to me just from the comments in the code that it's quite aware that
there may be situations that look imbalanced and yet where its changes may
not result in any improvement.  Which is not to say that it's perfectly tuned
yet.  

Maybe intrstat(1m) might be of use to see the sort of thing it appears
to be shuffling, namely the assignment of particular device interrupts to 
particular
CPUs (if my interpretation of what I've looked at is correct).

If someone in the know happens to be watching, perhaps they'd have some
advice for how to find out more about whether it's doing anything useful
in a particular situation.

Since you mention patents, there's a quite verbose description in the
patent application of what it _may_ be doing:

http://www.freepatentsonline.com/y2007/0043347.html
--
This message posted from opensolaris.org
_______________________________________________
opensolaris-discuss mailing list
[email protected]

Reply via email to