Run Arduino code on MSP430

Tawhid;

I've had a number of problems trying to use Energia with the MSP432.  This is a 32-bit Harvard architecture platform vs. the MSP430's 16-bit Von Neumann.  TI has a document on porting code that might be of some help: http://www.ti.com/lit/an/slaa656b/slaa656b.pdf

The hardware abstraction for the two is very different, and even TI states that CCS is not completely accounting for ISR's (the MSP430 has an abstraction that automatically addresses memory registers, and the MSP432 requires a two-step process).  A couple of pages later they state that the MSP432 "...is compatible with the Energia platform..."  Not sure how it can be both ways?

Here are a few lines from that doc that might be of some help (the highlights are mine):

A few key highlights include:

• Interrupt vector table

– MSP430 starts at 0xFFFF with Reset Vector and growing downwards.

– MSP432 starts at 0x0000 with Reset Vector and growing upwards.

• Flash, RAM, and other memory placements:

– MSP430 memory components share 64KB space (for 16-bit CPU devices) or 1MB space (for 20-

bit-supported CPUX), and addresses for each component potentially vary per device.

– MSP432 uses the abundant 32-bit address space to reserve separate memory locations for

different components.

• MSP432 introduces the bit-band feature to compensate for the Read-Modify-Write nature of the

Cortex-M instruction set.

• Individual bit manipulation in SRAM or in peripheral space can take advantage of bit-banding to reduce

the Read-Modify-Write operation down to a single instruction. DriverLib APIs take advantage of bitbanding

when manipulating peripheral registers. When writing your own register-access code, use the

following defines to leverage bit-banding when possible.

This doesn't exactly answer your question, but I can say I've also had the same type of problem with the MSP432; and had to resort to bit-banging to get things to work.

Tawhid;

I've had a number of problems trying to use Energia with the MSP432.  This is a 32-bit Harvard architecture platform vs. the MSP430's 16-bit Von Neumann.  TI has a document on porting code that might be of some help: http://www.ti.com/lit/an/slaa656b/slaa656b.pdf

The hardware abstraction for the two is very different, and even TI states that CCS is not completely accounting for ISR's (the MSP430 has an abstraction that automatically addresses memory registers, and the MSP432 requires a two-step process).  A couple of pages later they state that the MSP432 "...is compatible with the Energia platform..."  Not sure how it can be both ways?

Here are a few lines from that doc that might be of some help (the highlights are mine):

A few key highlights include:

• Interrupt vector table

– MSP430 starts at 0xFFFF with Reset Vector and growing downwards.

– MSP432 starts at 0x0000 with Reset Vector and growing upwards.

• Flash, RAM, and other memory placements:

– MSP430 memory components share 64KB space (for 16-bit CPU devices) or 1MB space (for 20-

bit-supported CPUX), and addresses for each component potentially vary per device.

– MSP432 uses the abundant 32-bit address space to reserve separate memory locations for

different components.

• MSP432 introduces the bit-band feature to compensate for the Read-Modify-Write nature of the

Cortex-M instruction set.

• Individual bit manipulation in SRAM or in peripheral space can take advantage of bit-banding to reduce

the Read-Modify-Write operation down to a single instruction. DriverLib APIs take advantage of bitbanding

when manipulating peripheral registers. When writing your own register-access code, use the

following defines to leverage bit-banding when possible.

This doesn't exactly answer your question, but I can say I've also had the same type of problem with the MSP432; and had to resort to bit-banging to get things to work.