OK, so for myself and others out there I wanted to clarify I am correct on how I am reading datasheets such as the one in the image below. The thing is the data sheet of this MCU says it is 16-bit registers but it looks to me like it has 32-bit addressing space.

The base address 5 hex values and if I were to just replace the offset with the last three hex values that would be 20-bit addressing which seems totally wrong to me. Whereas if I add the offset to the end of the base I get a 32-bit address, 4 bits per the 8 hex values. But again this is a 16-bit MCU so I am confused as to how to get a 16-bit address from what I am looking at.

If someone could should me what is wrong with the way I am reading this that would be great!

• Perhaps you're concatenating Base 0100 and Offset 0002 to get 01000002 instead of adding to get 0102. – user_1818839 Oct 5 '20 at 23:04
• Where do you get "5 hex values"? you don't count the 'h' at the end, that just eans to interpret the previous 4 characters as hexadecimal. It is only 4 hex values. – Kevin White Oct 6 '20 at 0:12

The base address 5 hex values ...

The base addresses are four hex digits long. The offsets are two.

Taking SFRRPCR:

0100h = 0x0100 = 0000 0001 0000 0000
+004h = 0x0004 = 0000 0000 0000 0100
-------------------
Sum    = 0000 0001 0000 0100 = 0104h = 0x0104


The base and offset values are simply added. For example, 012016 + 0A16 = 012A16.

Have a look into the header files that your compiler actually uses. The msp430fr2000_symbols.ld file for gcc says:

PROVIDE(PMMCTL0            = 0x0120);
PROVIDE(PMMCTL2            = 0x0124);
PROVIDE(PMMIFG             = 0x012A);
PROVIDE(PM5CTL0            = 0x0130);