Checking one bit only of IFG2

Corin Browning said:

if any of these other bits are high it will get stuck in the line of code above

Are you sure about that? "&" is a bit-wise AND operation. First off, UCA0TXIFG = BIT1, not BIT0 (Assuming 2xxx device, since you don't specify).

So the outcome of the IFG2&UCA0TXIFG portion is either 0x00 (FALSE) or 0x02 (Not FALSE), so the logical NOT operation should invert that.

You can see the assembly is

putc:

bit.b 0x2,&IFG2
jnc putc

That looks correct. And the other bits in the register should not affect how this operates. I think you have something else going on. If it hangs, it's because the bit is never getting set.

Sorry about that, mixed up RX and TX flags for UCA0.  Rx is BIT0 tx is BIT1.  And for UCB0 rx is BIT2 and tx is BIT3.  Here is my initialization routine for the UART, maybe somebody can find something that would cause it to conflict with the SPI initialization.

/**
 * Initializes the UART for 9600 baud with a RX interrupt
 **/
void initUART(void) {

        P1DIR = BIT0;

        //Sets all the pins on Port 1 to I/O
        P1SEL = 0;
        P1SEL2 = 0;

        //For UART Steps see pg 424
        UCA0CTL1 = UCSWRST;


        //UCSWRST = 1;
        //int i = 0;
        //UCSWRST = 0x00;

        UCA0CTL0 = 0x0;


        UCA0CTL1 |= BIT6+BIT7;    //Take UART out of RESET (toggle 1st bit of UCA0CTL1)
                                    //Select clock source for UART (6th and 7th bit in UCA0CTL1)
                                    // (UCCSEL)  --- 2^7 = 0x80h & 2^0 = 0x01h => 0x81


        //Baud rate settings for 8MHz (9600bps)
        //Refer to table on page 435 in MSP430x2xxx document
        UCA0BR1 = 0x03;        //load UCA0BR with 833(dec)= 0x341h
        UCA0BR0 = 0x41;        //UCA0BR is UCA0BR0 and UCA0BR1 where BR1 is the high byte

        //UCOS16 = 0 and UCBRF0 = 0 and UCBRS0 = 2  UCBR0 = 833 (see table on page 435)
        UCA0MCTL = 0x04;    //Baud rate settings for 8MHz (9600bps)

        //Setting port pins
        P1DIR |= BIT2;

        P1SEL |= BIT2; // Primary peripheral module function is selected.
                          // With the following line commented out (//P1SEL2 = BIT2;)

        P1SEL2 |= BIT2; // Secondary peripheral module function is selected.
                             // (if P1SEL and P1SEL2 are set high)


        UCA0CTL1 &= (0xFF - UCSWRST);//Clearing RESET bit

        //UCA0CTL1 |= UCSSEL_2; //Select clock source for UART (6th and 7th bit in UCA0CTL1)

        //UCBRS = 0x02;
        //UCA0BRS0
        //UCA0CTL0 = 0x00;
        //UCA0MCTL = 0x00;
        //UCA0TXIE = i;
        //UTXEx = 1;


}

Corin Browning said:

    UCA0CTL1 |= BIT6+BIT7;    //Take UART out of RESET (toggle 1st bit of UCA0CTL1)

Corin Browning said:

        UCA0CTL1 &= (0xFF - UCSWRST);//Clearing RESET bit

You have 8MHz SMCLK? Sure? Because the divider you use sets the USCI up for 9600Bd from 8MHz SMCLK.

Corin Browning said:

 * Initializes the UART for 9600 baud with a RX interrupt

First off I want to express my gratitude to the TI community for all the help and suggestions, this is my first "major" C code MSP430 project and it's proven to be quite a challenge but the breakthroughs make all the head banging worth it.

One of the beauties of assembly was that everything was so in depth and you could account for all the registers/memory/PC/stack and manipulating individual bits was very straight forward. I think I've gotten a better handle on bit manipulation in C now, it's just a little more complicated in my opinion.  I've been doing C programming for years but not for uPs. It does seem however that C has an advantage because less lines of code are needed to get what you want done, you just have to make sure you understand what each line is really doing "behind the scenes" so to speak.

I've managed to get the UART working along with the SPI interface.  I'm still having some strange issues here and there but I am really glad to have both the SPI and UART interface functioning now.  The main challenge from here on is storing 128bits from the SPI buffer into memory (FLASH?) and analyzing that data for finding the header "0000000001" and from there finding the unique 38bit "serial #" contained within those 128bits.  Right now I am having difficulty passing the data in the SPI Rx buffer to a global variable (volatile) and using this data later on, ie to print to the terminal the word/byte the contents of the SPI Rx buffer.  I had read in my MSP430 textbook that the SPI interface can be set to receive 16bits instead of 8 for each segment of the transmitted data.  If only I could figure out how to store 8 words from the SPI into memory...

Jens-Michael Gross said:

    UCA0CTL1 |= BIT6+BIT7;    //Take UART out of RESET (toggle 1st bit of UCA0CTL1)

The comment doesn't match the command. This line does NOT take the UART out of reset.
Besides this, I strongly recommend using the specific defines rather than BITx values. It's really easy to set something completely wrong and never notice.
UCA0CTL1 |= UCSSEL_3; is much more telling.

[/quote]

Thanks, this was a misleading comment, as you saw the software reset bit was cleared later on in the UART_init function.  I would prefer to use mnemonics such as UCSSEL_3 for example but half the time I can't find the proper mnemonic that the compiler recognizes.  I was using CCS and I really liked the ability to press CTRL and click on a mnemonic and go directly to the header/file where it is defined/created, but now I am using IAR (more stable on my current system) and it doesn't seem to have this ability.

Jens-Michael Gross said:

        UCA0CTL1 &= (0xFF - UCSWRST);//Clearing RESET bit