Why Trace_Printf doesNOT work?

[aresmarser]

Hi Sir,

I am working on NHS3100 with LPCXpresso.

I imported the project archive release_mra2_10_1_nhs3100.zip and tested the example projects such as "app_example_dp_i2c" to study I2C driver usage.

But here I met the problem that Trace_Printf doesNOT work.

It looks like the function "Trace_Printf" blocking the routine running.

I tried to disable "Trace_Printf" by adding a "return" statement in the begin of the function body as below:

---------

void Trace_Printf(const char *format, ...)
{

  return;

// ...

}

---------

and after that, the phenomenon disappears.

Could you help to review and advice how I should do or config something to let the original project "app_example_dp_i2c" run normally?

Many thanks,

Arna

[aresmarser]

Hi Dries Moors,

Thank you for your reply.

I read the function "Trace_Printf" again and debug by breakpoint.

It looks the default "#define TRACE_DATAPIPE_I2C0 1".

So the function call the below code snippet:

---

        #if TRACE_DATAPIPE_I2C0 == 1
    I2CIO_Tx((const uint8_t*)Trace_formattedString, stringSize);
        #endif

---

1. As my understand, this should require a I2C peripheral to printf the message?

2. So I should undefine "TRACE_DATAPIPE_I2C0" and define "TRACE_DATAPIPE_SEMIHOSTING" to value "1" if I have no I2C device for printf pipe?

Thanks,

Arna

[driesmoors · Accepted solution]

Hi,

1. Yes, correct.

2. Enabling will redirect the datapipe to the SWD debugger interface - provided you link in the semihosting functionality by using the Redlib (semihost) library. So, yes.

Best,
Dries.

[driesmoors]

Hi, Is this problem resolved at your side? Looking at the other posts, I assume it is? Keep in mind that semihosting in combination with the Deep Power Down and Power-off modes will not work: the moment you lose the SWD debug connection, any semihosting print will halt the core indefinitely.