Re: [PATCH 2/2] Initial generic hypertransport interrupt support.

[ebiederm@xmission.com (Eric W. Biederman)]

Dave Olson <olson@pathscale.com> writes:

> On Thu, 13 Jul 2006, Eric W. Biederman wrote:
> | > There's really nothing special at all about the interrupt
> | > setup, except in one very minor way.   The value of the HT interrupt
> | > destination address needs to be copied from HT config space, to
> | > an internal chip register (which is, can, and should be, handled by
> | > the driver init code).
> | 
> | The kernel changes the value at runtime, based upon user input.
> | I assume your mirror register needs to be updated after every change.
>
> Yes.  If the interrupt address changes, then we need a callback.
>
> | Since the kernel changes the value at runtime, and since a different
> | register needs to be written to, I can't quite use the generic code I
> | have written as is.  
>
> I imagine at least some other drivers would like to know when their interrupt
> configuration changes, also, so an interface where a driver can register
> a callback handler seems like the right generic answer, or more simply,
> a way for a driver to say it doesn't want it's interrupt handler
> migrated (which we would like anyway, for performance reasons).

As I recall that is "killall irqbalanced"

Eric