[PATCH] Fix masking of interrupts for 52xx GPT IRQ.

[Benjamin Herrenschmidt]

On Fri, 2011-03-04 at 23:40 +0100, Henk Stegeman wrote:
> What if the unmask/mask functions are dropped and only interrupt
> enable/disable functions are provided?

No, that's backward. enable/disable are handled by the core, leave them
there.

> I.e: rename the old unmask/mask functions to enable/disable and
> register them as enable/disable? I did try that, and it works too, no
> spurious IRQ's and no missing IRQ's.
> 

Ben.