PRINTF interferes with LPI2C3 on RT1050?

[jackking]

Hello,

I have run into a bit of an anomaly during my tests.

I am using the SDK example for the LPI2C driver (b2b_master_transfer)

Configuring the master to use LPI2C3 instead of LPI2C1, it seems that if I put a PRINTF statement before the transfer it causes the bus to leave the idle state and the transfer will hang.

// A print statement here will cause the lpi2c3 bus to leave the idle state
PRINTF("TEST\n");    

/* Send master non-blocking data to slave */    
reVal = LPI2C_MasterTransferNonBlocking(EXAMPLE_I2C_MASTER, &g_m_handle, &masterXfer);

// the reVal check will fail if the bus is not idle   
PRINTF("handle state:0x%02X \n", g_m_handle.state); 

/*  Reset master completion flag to false. */    
g_MasterCompletionFlag = false;   
 
if (reVal != kStatus_Success)    
{        
  return -1;    
}‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍‍

Any idea how the PRINTF is impacting LPI2C3?

This doesn't happen when using LPI2C1...

Thanks!

[danielchen]

Hi Jack:

I downloaded the latest SDK_2.5.0_EVK-MIMXRT1060, run the demo lpi2c_interrupt_b2b_transfer_master,

modified the code to use lpi2c3 as below,

#define EXAMPLE_I2C_MASTER_BASE (LPI2C3_BASE)

and added a PRINTF as you did.

 I only have one board at hand, so I don't connect to slave. But I don't see the transfer hang issue till now.  Could you let me know how to reproduce your issue? I need to connect to  a slave board?

Regards

Daniel

[jackking]

Thanks for testing that!  I think it was a result of a bad solder joint on my development board.  After touching it up, the problem went away.

Thanks