On Wed, May 19, 2010 at 10:10:58PM +1000, Michael Ellerman wrote:
> > > The "checks the scope" requires an RTAS call, which takes a global lock
> > > (and you add another) - these aren't going to be used for anything
> > > performance critical I hope?
> >
> > Nope it shouldn't be performance crit
On Tue, 2010-05-18 at 23:27 -0500, Sonny Rao wrote:
> On Tue, 18 May 2010 23:37:31 +1000, Michael Ellerman wrote:
> >
> > On Mon, 2010-05-17 at 22:53 +1000, Mark Nelson wrote:
> > > This patch adds support for handling IO Event interrupts which come
> > > through at the /event-sources/ibm,io-event
On Wed, 2010-05-19 at 20:24 +1000, Mark Nelson wrote:
> On Tuesday 18 May 2010 23:37:31 Michael Ellerman wrote:
> > On Mon, 2010-05-17 at 22:53 +1000, Mark Nelson wrote:
> >
> > > + /* check to see if we've already registered this function with
> > > + * this scope. If we have, don't register it
On Wednesday 19 May 2010 14:27:44 Sonny Rao wrote:
> On Tue, 18 May 2010 23:37:31 +1000, Michael Ellerman wrote:
> >
> > On Mon, 2010-05-17 at 22:53 +1000, Mark Nelson wrote:
> > > This patch adds support for handling IO Event interrupts which come
> > > through at the /event-sources/ibm,io-events
On Tuesday 18 May 2010 23:37:31 Michael Ellerman wrote:
> On Mon, 2010-05-17 at 22:53 +1000, Mark Nelson wrote:
> > This patch adds support for handling IO Event interrupts which come
> > through at the /event-sources/ibm,io-events device tree node.
>
> Hi Mark,
>
> You'll have to explain to me o
On Tue, 18 May 2010 23:37:31 +1000, Michael Ellerman wrote:
>
> On Mon, 2010-05-17 at 22:53 +1000, Mark Nelson wrote:
> > This patch adds support for handling IO Event interrupts which come
> > through at the /event-sources/ibm,io-events device tree node.
>
> Hi Mark,
>
> You'll have to explain
On Mon, 2010-05-17 at 22:53 +1000, Mark Nelson wrote:
> This patch adds support for handling IO Event interrupts which come
> through at the /event-sources/ibm,io-events device tree node.
Hi Mark,
You'll have to explain to me offline sometime how it is we ran out of
interrupts and started needing
This patch adds support for handling IO Event interrupts which come
through at the /event-sources/ibm,io-events device tree node.
There is one ibm,io-events interrupt, but this interrupt might be used
for multiple I/O devices, each with their own separate driver. So, we
create a platform interrupt
