Re: [RFC PATCHv2 0/4] Support for OTP memory

[Jamie Iles <jamie@jamieiles.com>]

On Thu, Mar 24, 2011 at 02:36:42PM -0400, Mike Frysinger wrote:
> On Thu, Mar 24, 2011 at 14:32, Jamie Iles wrote:
> > On Thu, Mar 24, 2011 at 01:56:17PM -0400, Mike Frysinger wrote:
> >> just an overview note here ... the Blackfin OTP is actually IP we
> >> bought from someone (not sure if i can say more, although i dont know
> >> the details myself ... i'd have to go ask around), so much of the
> >> feedback i give about the structure of it most likely (hopefully) be
> >> applicable to other people who have purchased this IP.  if/when anyone
> >> who is also using this IP steps forward, we could even see about
> >> generalizing the bfin-otp driver.
> >
> > Yes, we have a similar thing here - a block of OTP from an IP vendor and
> > a wrapper to provide an AXI interface.  As far as I know there are only
> > a few OTP vendors so there may well be plenty of common code in
> > different OTP implementations.
> 
> do you guys wrap the interface ?  the Blackfin processors have an
> on-chip ROM which provides higher level functions like
> read/write/lock/etc...  the Linux driver uses those rather than
> programming the memory mapped OTP interface itself since this
> interface is not publicly documented.

I'm not 100% on the exact details but I understand that the macro 
provides some non-AXI registers and signals that we wrap up into some 
AXI registers.  The wrapper doesn't handle all of the nasties of the 
macro but it does handle some of the redundancy and region splitting.

Jamie