Re: [avrdude-dev] avrdude problem with Attiny13 - works fine in AVRstudio

[Haberler Michael]

Am 04.01.2009 um 23:12 schrieb Joerg Wunsch:

> As Haberler Michael wrote:
>
> > - second run should program target in ISP mode, but fails as the
> > signature is read back as 0x000000 .
>
> Curious: can you proceed successfully when using the -F (force)
> option?  This will make AVRDUDE ignore the signature mismatch.

I tried that already - it wont sucessfully program the chip.

The problem seems to be that the read fuse operation is done  
differently in avrdude than in AVR Studio programming tool. I'm still  
decoding the AVRStudio protocol USB trace to see what causes the  
difference.

I have the suspicion that it's part of the "Shenzen feature set" of my  
MKII clone handling embedded ISP commands in the MKII stream. It is  
buried obviously in the second avrdude trace of my original mail  
during command sequence 5-7 where the signature is read and all bytes  
come back as zeroes. The KEE folks respond to mail quickly and try to  
be helpful though, so I would give it a stab.


A plea for help:

Can somebody with a bona fide (original ATMEL) JTAGICE MKII programmer  
and a debugWire capable AVR chip at hand do the following:

- set the chip to debugWire enabled
- run avrdude twice trying to program it like
avrdude -pt13 -cjtag2isp -Pusb -v -v -v -v -v  - 
Uflash:w:<somehexfile>.hex:a
avrdude -pt13 -cjtag2isp -Pusb -v -v -v -v -v  - 
Uflash:w:<somehexfile>.hex:a

and mail me the avrdude outputs?

that should help a lot in nailing the problem down. I dont think the  
nature of the hexfile actually matters. I would assume 'main(){}'  
compiled for your target at hand should do.


thanks in advance


-Michael



> --
> cheers, J"org               .-.-.   --... ...--   -.. .  DL8DTL
>
> http://www.sax.de/~joerg/                        NIC: JW11-RIPE
> Never trust an operating system you don't have sources for. ;-)