[PATCH 4/4] ppc32 CPM_UART: Fixed odd address translations

2006-04-26 Thread Marcelo Tosatti
Ouch. I would convert the warning printk() to a BUG(), since such an address error indicates deep trouble anyway... And add a "likely()" indication around the range check. Looks great otherwise! On Tue, 2006-04-25 at 20:26 +0400, Vitaly Bordug wrote: > Current address translation methods can

[PATCH 4/4] ppc32 CPM_UART: Fixed odd address translations

2006-04-25 Thread Vitaly Bordug
Current address translation methods can produce wrong results, because virt_to_bus and vice versa may not produce correct offsets on dma-allocated memory. The right way is, while tracking both phys and virt address of the window that has been allocated for boffer descriptors, and use those numbers