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.

Linux/AM4372: 8250 UART driver issue

Part Number: AM4372

Tool/software: Linux

I've developed a board with the AM4372 and we are working to use a SCADA protocol along the serial port.  I'm having an issue where every so often the first byte of a message would get repeated and the last byte of the message would be missing.  We put a scope on it to determine this functionality.  I also put kernel print statements out at the point before it calls serial_out to see if there was something wrong in the transmit buffer.  The bytes match up to what is expected.  I'm assuming at this point it is writing to the register when it calls serial_out so I don't know where else I can dig into this problem further.  At this point I am assuming that there is something with the UART that is causing this.  I need some guidance on what my options are with this board and what I can do to fix this problem.  We are communicating over 9600 8n1.

  • Just some more information. I'm using the latest SDK and it seems to be that this issue came about when the 485 implementation was added for the 8250 driver.
  • Which serial driver are you using? You should be using the 8250_omap.c driver. And can you attach your dts file?

    Steve K.
  • Are you still seeing the issue?

    Steve K.
  • Steve,

    So I was able to get around this by using the latest version 3 SDK and modifying the serial omap driver to implement the RTS signal.  I still have not gotten the latest SDK with the latest linux version in there to work correctly.  I think its something related to the latest linux omap driver when they implemented the 485 stuff.  I don't see the issue before that but I do see the issue with the latest version.  Right now I am ok and able to move forward with my project but at some point I would like to update to the latest linux kernel if I can but I don't have the time right now to dive into what changes were made to the linux serial driver to see where the issue is.  I'm convinced its a linux issue related to these changes for 485.