C2000 Launchipad's SCI(UART) works oddly

This is my SCI initializing code. Please give me some advice.

void scia_init()
{

    CLK_enableSciaClock(myClk);

    // 1 stop bit,  No loopback
    // No parity,8 char bits,
    // async mode, idle-line protocol
    SCI_disableParity(mySci);
    SCI_setNumStopBits(mySci, SCI_NumStopBits_One);
    SCI_setCharLength(mySci, SCI_CharLength_8_Bits);

    // enable TX, RX, internal SCICLK,
    // Disable RX ERR, SLEEP, TXWAKE
    SCI_enableTx(mySci);
    SCI_enableRx(mySci);
    SCI_enableTxInt(mySci);
    SCI_enableRxInt(mySci);

#if (CPU_FRQ_60MHZ)
    SCI_setBaudRate(mySci, SCI_BaudRate_9_6_kBaud);
#elif (CPU_FRQ_50MHZ)
    SCI_setBaudRate(mySci, (SCI_BaudRate_e)13);
#elif (CPU_FRQ_40MHZ)
    SCI_setBaudRate(mySci, (SCI_BaudRate_e)10);
#endif

    SCI_enableFifoEnh(mySci);
    SCI_resetTxFifo(mySci);
    SCI_clearTxFifoInt(mySci);
    SCI_resetChannels(mySci);
    SCI_setTxFifoIntLevel(mySci, SCI_FifoLevel_Empty);
//    SCI_enableTxFifoInt(mySci);


    SCI_resetRxFifo(mySci);
    SCI_clearRxFifoInt(mySci);
    SCI_setRxFifoIntLevel(mySci, SCI_FifoLevel_Empty);
//    SCI_enableRxFifoInt(mySci);


    SCI_setPriority(mySci, SCI_Priority_FreeRun);
    SCI_enable(mySci);

    return;
}

Hi,

Which C2000 controller are you using? And the code that you're using is a custom one or a sample?

Regards,

Gautam

Thank you for your interest.

I have started from Launchpad demo sample.

Can you explain about C2000 controller? I just using C2000 launchpad, LAUNCHXL-F28027 and CCS v5.4.

Regards.

Hwang

Hi,

 Can you explain about C2000 controller?

I would first like you to go through SCI basics:

http://www.ti.com/lit/ug/sprugh1c/sprugh1c.pdf

Regards,

Gautam

I read it. But I can't know the controller.

However, I can clarify the problem.

By the Guide,

RXRDY should set when SCIRXBUF get a data. But my code doesn't work that.

This is why I'm very confused. Can you give me some advice?

Regards,

Hwang.

Hi,

The above doc is F28027 compatible. If RXRDY is not setting even after your Bluetooth module has sent a data then  there seems to be only one issue; baudrate matching! I guess your BM will be set to 9600bps as default; so please go through your BM's specs and initialize F28027's baudrate properly.

Regards,

Gautam

I don't use the BM yet.

I'm testing board's SCI(UART) function by PC(hyper terminal).

The board can Tx data well. But Rx is not.

I can observe the SCIRXBUF get the right data.

But RXRDY never sets.

In other words, Tx works well but Rx is not. In this case, the baudrate can be problem?

Sorry to bother you.

But I can't understand this problem.

Regards,

Hwang

Congrats!

Cheers,

Gautam

I'm not sure if this is a related problem but I see a potential issue with the SCI_setTxFifoIntLevel() function which seems to be incorrectly modifying the SCIFFTX register rather than the SCIFFRX register. The line that concerns me is hilighted in yellow below:

void SCI_setRxFifoIntLevel(SCI_Handle sciHandle, const SCI_FifoLevel_e fifoLevel)
{
    SCI_Obj *sci = (SCI_Obj *)sciHandle;


    // clear the value
    sci->SCIFFTX &= (~SCI_SCIFFRX_IL_BITS);  // <--- This should be SCIFFRX I believe

    // set the bits
    sci->SCIFFRX |= fifoLevel;

    return;
} // end of SCI_setRxFifoIntLevel() function

Gautam, could this potential explain any issues with Tx and Rx not both working correctly in FIFO interrupt mode?  

Cheers,

Gord

Hello Gordon,

Good finding! ^^

Indeed it is a bug (TI member should note this in their bug file for next release).

If SCI_setRxFifoIntLevel is called after SCI_setTxFifoIntLevel, then SCIFFTXIL will be cleared and usually never been set again because these functions usually are called only once in initialization.

However in Hwang case, the TXFFIL bits are 0, so i think this doesn't influence the entire code.

Best regards,

Maria