[PATCH] MTD: LPC32xx SLC NAND driver

Roland Stigge stigge at antcom.de
Tue May 15 23:48:28 EST 2012


On 05/15/2012 03:31 PM, Artem Bityutskiy wrote:
>> As I understand loops_per_jiffy, this loop will take much longer
>> than the 100 ms you defined above?
> 
> Not sure about much, but longer. The idea is that this is about
> the error path so if we report -EIO with a slight delay - no
> problem.

Turned out that the condition (FIFO empty) is always true for me.
Keeping the check for safety reasons for now, doing the timeout with
msleep()s which should be (cpu-wise) "social" enough and are
unexpected anyway but do approximate the ms timeout more precisely.

Roland


More information about the devicetree-discuss mailing list