Patch from abbotti@mev.co.uk (Ian Abbott):

Patch from abbotti@mev.co.uk (Ian Abbott):

Micheal Head reported problems with the PCI230 generating interrupts constantly
in his system.

Looking at the source code, that could happen if the AI interrupt handler
(pci230_handle_ai) detects a FIFO overrun error.  In that case, it ends up
turning the ADC interrupt source back on in the board's INT_SCE by mistake after
cancelling the transfer, resulting in continuous unhandled interrupts.

I have a patch to fix the above and initialize the INT_SCE register before
hooking up the interrupt with request_irq.