What I am trying to achieve is more or less exactly what I think Pranav is trying to do - read the latch without modifying the contents and store it in some arbitrary location

Why do you guys fight using the proven / standard method of a shadow register in software? In all the years I have worked with microcontrollers no other method has come up as more effective.

The only place where actually reading the latch comes in as the method is when debugging a program written without shadowing (like most).

Erik