On Tue, Nov 26, 2013 at 06:35:35PM +0530, Sourav Poddar wrote: > On Tuesday 26 November 2013 06:31 PM, Mark Brown wrote: > >Is this really generic enough to put here rather than in the driver > >data? > The thought behind this was that if a master controller supports > memory mapped operation and slave wants to use memcpy in his > driver(bypassing spi ), then we should have a mean to communicate > between the > master and the slave. So, 'spi_master' seems to be the place for me > for this data, > which could be parsed in the slave side also. I would have expected that the address would be returned by the function that maps the buffer in rather than stored permanently in the master.
Attachment:
signature.asc
Description: Digital signature
