[Arm-netbook] CT-PC89E: DS2431 eeprom driver reverse-engineering PLEASE CHECK

[Luke Kenneth Casson Leighton]

http://piratepad.net/LPEw9nHASt

could someone please verify that i have the read_bit, read_byte,
write_bit and write_byte functions correct?

i'm particularly concerned about lines 179 and 269, which will
determine whether the bits are being stored / read MSB or LSB.

also absolutely absolutely crucial is to ensure that the GPIO Bank N
pin4 setpin and cfgpin sequence and timing is absolutely correct.
failure to get this right will result in corruption of the eeprom
data, it's as simple as that.

what seatron - the designers of the SO-DIMM - have done is to connect
the eeprom *directly* to a GPIO pin.  by setting it to an input or an
output, you can read or write data.  so basically this is I/O at its
mooost basic.

and _guess what_?  there's no irq stop/save/restore around all this,
so it's dead-easy to have the eeprom corrupted!

goood grieeeef.. but, ignoring that: i _really_ need someone to review
the sequence.

tia,

l.