Re[2]: [avrdude-dev] ATmega162 support + one minor issue

[Stephan Eisvogel]

On Sunday, September 7, 2003, 1:36:09 AM, Brian Dean wrote:
> Technically, this is a feature.

Brian, thank you for this insight.

> Glad you got your part working!  Thanks for the update to the config
> file.  It will get added some point soon.

Yeah and Atmel misguided me again, datasheet claimed a WRONG efuse bit
pattern  and  I  blindly  typed  that in, not checking the actual bits
present in that byte. You have to change efuse to:

--
    memory "efuse"
        size            = 1;
        read            = "0 1 0 1  0 0 0 0  0 0 0 0  1 0 0 0",
                          "x x x x  x x x x  o o o o  o o o o";

        write           = "1 0 1 0  1 1 0 0  1 0 1 0  0 1 0 0",
                          "x x x x  x x x x  1 1 1 i  i i i 1";
      ;
--

This  looks  correct  now  and  also  works for my part (the four bits
control brown-out trigger level and the ATmega161 compatibility mode).
One  curiosity to note is bit 5, which the manual claims is "reserved"
and "unprogrammed" hence always 1. You can set it to 0, in contrast to
bits  0,  6  and  7  which  are  always 1 indeed. This really warrants
further  investigation so tomorrow will be a good time to dust off the
HNO3  acid  bottle  and  put  the scanning electron microscope and the
micro prober to good use and find out! (just kidding... I really don't
want to know. :o)

In  case  this  wasn't  the last "issue" with ATmega162 I'll send more
fixes as soon as I found and squashed them.

Nite,
Stephan