Hi Kevin,

[...]
>>> >> Any objection to making and testing that change?

There is a difference in the way the OMAP and Davinci debug macros work.
 The OMAP uncompress passes in a code (OMAPxUARTy).  The debug macros
then figure out the phys and virt bases of the requested UART.

On Davinci, we unfortunately don't have a consistent memory map across
the family, and here we need to pass in base addresses from uncompress.
 For this, we have been passing both phys and virt bases, so that the
debug macros don't have to rely on IO_ADDRESS() translation internally.
 This flexibility is necessary for tnetv107x, where IO_ADDRESS() does
not translate properly.

Regards
Cyril.
_______________________________________________
Davinci-linux-open-source mailing list
[email protected]
http://linux.davincidsp.com/mailman/listinfo/davinci-linux-open-source

Reply via email to