Re: [avrdude-dev] con"fuse" bits

[Brian Dean]

On Sat, May 17, 2003 at 07:05:18PM -0700, Theodore A. Roth wrote:

> I'm wondering what the logic behind setting unused fuse bits to 0 before 
> display is?
> 
> This seems to be confusing behaviour. When you read the fuse bits from a 
> device, unused and unprogrammed bits are read as ones and programmed bits 
> are zeros. Looking at the printed bytes values and then trying to figure out 
> the values from the datasheet seems slightly counter intuitive to me.
> 
> >>> d hfuse
> avrdude: Send: V [56] X [58] . [08] . [00] . [08]   [20]
> avrdude: Recv: . [14]
> avrdude: Recv: . [ff]    <<<<<<<< read fuse byte as 0xff
> avrdude: Recv: . [10]
> 0000  07                 <<<<<<<< display fuse byte as 0x0f ???
>  
> avrdude> d lfuse
> >>> d lfuse
> avrdude: Send: V [56] P [50] . [00] . [00] . [00]   [20]
> avrdude: Recv: . [14]
> avrdude: Recv: . [d2]    <<<<<<<<< read as 0xd2
> avrdude: Recv: . [10]
> 0000  c2                 <<<<<<<<< display as 0xc2 ???
>  
> 
> If there is some compelling reason for this behaviour, it is not obvious to 
> me.

What chip are use working with?  With an ATmega128 here, only two bits
are defined for the extended fuse "byte".  When I display it, I get:

>>> d efuse 
avrdude: Send: V [56] P [50] . [08] . [00] . [00]   [20] 
avrdude: Recv: . [14] 
avrdude: Recv: . [ff] 
avrdude: Recv: . [10] 
0000  ff                                                |.               |

Unused bits are left alone.  Perhaps the instruction bits are not
encoded correctly for the chip you are using.

-Brian