i think you are confused.

just code protect the whole chip. what's the problem with that ? when you need to update your code, just update the code on
the whole chip. code protecting the flash does not make it OTP. you can program it as many times as you like. you just can't
READ it with the code protect on.

you've written a lot of rant here basically to say that you can't use a bootloader and code protect at the same time, that's
the bottom line.

solution : don't use a bootloader on production parts, just use ICSP. easy.


dld


On Tue, 17 Jul 2001 15:10:30 -0400, Ron Anthony wrote:

>Hello all, who else uses the 16F877?  Who else is totally disgusted that the
>77A revision is delayed until next March at the earliest???  The 77 is a
>flash chip that if you leave it flash upgradable, you must leave it
>completely NAKED for copying.  Is this as ridiculous as I think it is?  If
>you code protect the flash it becomes OTP memory that can never be updated
>without wiping out the entire chip, which means your bootloader code is GONE
>and can't be used to update the flash.
>
>Your options are:
>
>1.      leave the chip 100% naked and copyable, downloadable, and 100% reverse
>engineer-able
>2.      code protect 50% of the ram, making that half OTP style memory, you can't
>use a bootloader on it, and the other half is still naked
>3.      protect only 256 bytes, which prevents whole chip copying, but all that
>is missing is 256 bytes from someone's downloaded code, leaving them only
>256 bytes to recreate.
>
>None of these options are good.  The 77A revision is much more secure.  You
>can, supposedly, disable in circuit serial programming, and can also disable
>interal flash reads.  The chip is secure, the only liability being someone
>watching your comms when you update by way of the bootloader.  But, this can
>be solved with encryption algorithms that decode the flash update after
>getting blocks of data into the chip where it can't be watched.
>
>Microchip has the last 2 years to get out the 77A revision.  What the heck
>is taking this multi-billion dollar company from getting this chip out?  Do
>they have no respect whatsoever for the very code that runs in their chip?
>Most times that's the only non-commodity part of a device!!!!!
>
>Who else is utterly disappointed about this?  Now I have the tortured
>decision to make on a very large production run.  Should I leave the chip
>naked and throw caution to the wind?  Should I code protect half the chip
>with what code can be reasonably considered "stable" ???  Or should I leave
>the whole chip naked but for the 256 bytes they allow?  Not hard to get
>around that, having everything but 256 bytes.
>
>Any thoughts?  Advise?  Words of wisdom????
>
>Thanks All.
>Ron Anthony, a very disappointed buyer
>
>--
>http://www.piclist.com hint: The list server can filter out subtopics
>(like ads or off topics) for you. See http://www.piclist.com/#topics
>
>

--
http://www.piclist.com hint: The list server can filter out subtopics
(like ads or off topics) for you. See http://www.piclist.com/#topics