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.

UART1 vs UART3 for console and periph booting on OMAP35x

Is there a particular reason that on the OMAP35x EVM, UART1 is used for the console and UART3 is used for peripheral booting?  Is there any reason why UART3 can't be used for console and for periph booting?  The product we''re designing uses the other UART ports (1+2) for other peripherals so they're unavailable for use as the console port.  (And the only UART that the OMAP35x supports for peripheral booting is UART3.)  I'm concerned I may be overlooking something in assuming I can support both console and periph booting on UART3.

Thanks,

twebb

  • There isn't any hardware restriction, it would mainly be a modification to the operating system and associated bootloaders to direct them to use UART3.

    What Operating System are you targeting?

    For instance, the WinCE BSP and bootloader uses UART3 for the console output.

  • I have wondered this myself, and as Brandon mentions there is no hardware limitation that is causing this. I am guessing that it was hooked up this way to allow you to monitor the terminal output of the booting kernel/U-Boot simultaneously while allowing you to boot load the board without having to swap cables, of course though this would require two UARTs on a PC or two PCs, the boot loading utility typically takes full control of the UART port so you could not see any output from the board during/immediately after the serial boot process unless you had two UARTS exposed.

  • I'm using Linux.  Started with SDK-0.9.5, then 1.0.0, and have now moved to linux-omap-2.6.27-omap1 for various reasons.  Assuming that our board's sys_boot[5:0] inputs are strapped like the default settings for SW4 of the EVM (110110 = peripheral booting:UART3:OneNAND), and we were using UART3 for console also, wouldn't this result in the "ASIC_ID" sequence being sent out to the console at every cold boot?  Are you aware of any side effects to that?

    Thanks.

     

  • You are correct that you would see the RBL trying to contact a host for serial booting, but as long as your host terminal program was ignoring it there should be no side effects, since this would be coming out of UART3 anyway with the boot mode, you just would not have seen it.