Re: [avr-gcc-list] UISP erased mega128 id bytes

[Iztok Zupet]

On Saturday 20 July 2002 22:04, Marek Michalkiewicz wrote:
> > This happend after about 20-th erase/program cycle, using the dapa mode.
> > The ID bytes are restored back after uploading the fuse segment, but
> > again ruined by a program attempt to flash segment, while the eeprom
> > segment still behaves normaly.
>
> It may be that you found a bug in the chip.  Is this a sample or normal
> production part?
>
> I had a similar problem long time ago with one of the ATmega163 samples -
> reported to address@hidden and they told me it can happen in samples, but
> normal production parts have the signature bytes locked.  Still it seems
> strange to me that these bytes are flash memory and not just hardwired...
>
> Marek

The chips were sold to us as normal parts. Series is 0220. 

Seems that the chip ID is not locked, since it can be written back with entire 
fuse section. But the erase function erases tham again.

The fuse section i use is:

S00700006675736545 
S10800002FC1AEFFFF5B 
S5030001FB 
S9030000FC 

Regards
iz
avr-gcc-list at http://avr1.org