• State Resolved
  • Locked Locked
  • Replies 8 replies
  • Subscribers 33 subscribers
  • Views 842 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

LX4F and I2C1 initialization

mmaaxx
Intellectual 280 points

Hi, I'm trying to setup PA6 and 7 as I2C1 and have a very hard time :(

here is a code that I'm using (an example from stellarisware - slave_receive_int.c, but adopted for I2C1):

[code]
// Setup clock and interrupts
ROM_FPUEnable();
ROM_FPULazyStackingEnable();

ROM_SysCtlClockSet(SYSCTL_SYSDIV_1 | SYSCTL_USE_PLL | SYSCTL_XTAL_16MHZ | SYSCTL_OSC_MAIN);

// I2C
ROM_SysCtlPeripheralEnable(SYSCTL_PERIPH_I2C1);

// Configure peripherals that are being used
ROM_SysCtlPeripheralEnable(SYSCTL_PERIPH_GPIOA); // Display, I2C


ROM_GPIOPinConfigure(GPIO_PA6_I2C1SCL);
ROM_GPIOPinConfigure(GPIO_PA7_I2C1SDA);
ROM_GPIOPinTypeI2C(GPIO_PORTA_BASE, GPIO_PIN_6 | GPIO_PIN_7);

ROM_I2CMasterInitExpClk(I2C1_MASTER_BASE, SysCtlClockGet(), false);

ROM_I2CMasterIntEnable(I2C1_MASTER_BASE);
ROM_I2CSlaveEnable(I2C1_SLAVE_BASE);
ROM_I2CSlaveInit(I2C1_SLAVE_BASE, 0x11);
ROM_I2CSlaveIntEnable(I2C1_SLAVE_BASE);

ROM_IntEnable(INT_I2C1);
[/code]

I have 2 pull-ups and scope on PA6 and 7. 

And right after "ROM_GPIOPinTypeI2C(GPIO_PORTA_BASE, GPIO_PIN_6 | GPIO_PIN_7);" line of code executes I2C-Clock goes low, and remains low. 

Any suggestions?

  • 0
    TI__Mastermind 18905 points

    Can you tell me some background information:

    Which LM4Fxxx part are you using?

    Are you working on an evaluation kit, or your own custom design?

    Lela

  • 0 in reply to Lela Garofolo
    Intellectual 280 points

    I'm trying the code on Both EKS-LM4F232 Eval Kit, and custom design board (based on LX4F120E). Trying I2C1 on both boards. 

    Thank you Lela!

  • 0
    Guru 117855 points

    Quickly - several items jump out:

            GPIOPinTypeI2CSCL(GPIO_PORTA_BASE, GPIO_PIN_6);   //   I2CSCL required for M4 parts

    SYSCTL_SYSDIV_1 and your use of SYSCTL_USE_PLL - - Just how fast do you think your MCU will then run?

    Would you not be better served by using an I2C in Master - rather than Slave mode?  (at least for preliminary check out)

    6 months or so past - I posted first known, complete set-up/config "working" code for basic M4 I2C operation.  Forum search should reveal...

  • 0 in reply to cb1_mobile
    Guru 17650 points

    cb1_mobile said:
    6 months or so past - I posted first known, complete set-up/config "working" code for basic SSI operation.  Forum search should reveal...

    ... and there's also a working example API in the Code Sharing Section ...

  • 0 in reply to cb1_mobile
    Intellectual 280 points

    Thank you sir! 

    GPIOPinTypeI2CSCL did the trick. Point, about Divider and PLL, taken, I assume it wasn't really running at 200Mhz. I copied rest of the initialization from the driver I did for LM3 years ago, that's why it also inits Slave Mode. 

    Too bad that StellarisWare doesn't contain proper I2C init sample for M4 parts :/

  • 0 in reply to mmaaxx
    Guru 117855 points

    My pleasure - my friend, and thank you.  You were almost there - and good for you re: PLL and likely divide by 4.  (for 50MHz opn.)

    At some point - vendor staff will address that, other issues - for now your best bet is to post here - and use the forum search function.  Much "breaking news" here - which cannot/has not yet made it into, "official release... "

    Final point - (for you/others) do note that ROM version of, "GPIOPinTypeI2CSCL" had not yet made it into "normal" ROM code (w/in MCU) - thus my quite deliberate use of Flash version - for that function.  (and MAP call would also work...)

  • 0 in reply to cb1_mobile
    Prodigy 40 points

    If you really want to use the ROM functions, ROM_GPIOPinTypePWM(GPIO_PORTA_BASE, GPIO_PIN_6) will do the same thing as GPIOPinTypeI2CSCL(GPIO_PORTA_BASE, GPIO_PIN_6).

  • 0 in reply to Walter Benjamin
    Guru 117855 points

    While counter intuitive - and w/the caveat that you confine this to Blizzard class & beyond - you appear correct.

    GPIO.C reveals only this difference: (below is encoded w/in GPIOPinTypeI2CSCL() - Ver 9453:)

     if(CLASS_IS_SANDSTORM || CLASS_IS_FURY || CLASS_IS_DUSTDEVIL ||
           CLASS_IS_TEMPEST || CLASS_IS_FIRESTORM)
        {
            GPIOPadConfigSet(ulPort, ucPins, GPIO_STRENGTH_2MA, GPIO_PIN_TYPE_OD);
        }

    That code block does not appear w/in GPIOPinTypePWM().   Otherwise - code is line for line duplicative.

    As TI's Dave W. has herein noted - even a "single usage" of non-ROM function - causes the full addition of that code's base w/in Flash. 

    Your language can be strengthened via, "If you really want to exclusively use the ROM I2C functions..."

    Curious as to your motivation for this discovery - yours is the first such mention to my knowledge...  (as I spend some limited time here - other ARM/tech forums...)  Not too bad a, "first-time" post (really quite extraordinary!....)

    Update: unknown yet is whether the apparent "clash" between GPIOPinType() (as PWM - under your discovery) and GPIOPinConfigure() (as I2C) may someway/somehow "disrupt" any of the IDEs or any related Stellaris codings...