Can all the interrupt functionality be utilised by using only one interrupt pin (eg INT1) and leaving the other one not connected? In UG 214 I did not find any information stating that two pins are mandatory for using interrupts.
With two interrupts you can segregate events like input format changes and hot plug detect changes so there are two different interrupt handlers for these events. Each handler would only have look at it's associated subset of interrupt sources to determine what to do resulting in faster system response times.
As an example suppose you had only CP_UNLOCK interrupt mapped to INT2. Upon an CP unlock event you would immediately know you've loss CP lock and you could turn off an external audio path almost instantly without reading the register bit over the I2C bus. If all the interrupt source were mapped to INT1 then when a CP unlock event occurred, you'd have to read all the possible interrupt source register over I2C and see what changed. There would be a longer lag between the CP unlock event and actually turning off the audio path.
The options and flexibilities are up to you.
Both INT pins have the same masking structure therefore only on pin is really needed.
The quickest way to verify this is, in the index of the software user guide, If you search for "MB1" you will always fine a matching "MB2" indicating the underlying mask structures are the same.
Two INT pins allow you to assign a few high priority interrupts to one pin while the other handles all the rest. If too many interrupts are mapped to one pin it takes a while to search all possible interrupt sources over the I2C bus to see what actually caused the interrupt.
Thanks for the reply. In UG 214 section 16.2.4 it is mentioned that the decoder cannot handle two active interrupts at the same time. What is the point therefore in assigning interrupts between the two pins priority-wise?
Retrieving data ...