Hi Ben! One of the reasons why the 2410 NAND access is so slow, is the fact that the NAND controller is broken so it can't work together with the DMA controller.
After having studied the 2440 documentation in detail last week (to implement 2440 and large-page-size NAND steppingstone code for u-boot) it occurred to me that this bug is not mentioned in the 2440 docs. Does this now mean that they just forgot to document this bug, or is the 2440 NAND controller really interoperable with DMA? Do you know anything about this? If you know or assume it has been fixed, then openmoko should probably have a go at adding DMA support to the linux drivers/nand/s3c2410.c code... Cheers, -- - Harald Welte <[EMAIL PROTECTED]> http://openmoko.org/ ============================================================================ Software for the world's first truly open Free Software mobile phone
