Re: [avr-chat] ATMEL AVR UART Bootloader for AVR-GCC/avr-libc on ATmega16 with avrdude

[Alex Eremeenkov]

Hello Joerg,
> Alex Eremeenkov <address@hidden> wrote:
>
>   
> > All can find board without problems. But they cant update
> > application (problems with writing, reading flash,etc).
> >     
>
> I have fetched Martin's bootloader code, built and installed it
> according to his instructions (mostly using defaults except for the
> clock frequency, as my STK500 generates only 3.68 MHz), and have no
> problems using it:
>
> % ./avrdude -C avrdude.conf -c avr109 -P /dev/cuad0 -b115200 \
> -p atmega16 -U demo.bin
>
> Connecting to programmer: .
> Found programmer: Id = "AVRBOOT"; type = S
>     Software Version = 0.8; No Hardware Version given.
> Programmer supports auto addr increment.
> Programmer supports buffered memory access with buffersize=128 bytes.
>
> Programmer supports the following devices:
>     Device code: 0x75
>
> avrdude: AVR device initialized and ready to accept instructions
>
> Reading | ################################################## | 100% 0.01s
>
> avrdude: Device signature = 0x1e9403
> avrdude: NOTE: FLASH memory has been specified, an erase cycle will be performed
>          To disable this feature, specify the -D option.
> avrdude: erasing chip
> avrdude: reading input file "demo.bin"
> avrdude: input file demo.bin auto detected as raw binary
> avrdude: writing flash (342 bytes):
>
> Writing | ################################################## | 100% 0.06s
>
>
>
> avrdude: 342 bytes of flash written
> avrdude: verifying flash memory against demo.bin:
> avrdude: load data flash data from input file demo/demo.bin:
> avrdude: input file demo.bin auto detected as raw binary
> avrdude: input file demo.bin contains 342 bytes
> avrdude: reading on-chip flash data:
>
> Reading | ################################################## | 100% 0.04s
>
>
>
> avrdude: verifying ...
> avrdude: 342 bytes of flash verified
>
> avrdude done.  Thank you.
>
>
> This is the SVN version of AVRDUDE, although I doubt it makes much of
> a difference compared to 5.8 for the AVR109 (aka. butterfly)
> programmer type.
>
>   

I haven't  board near now, so can't check that i got from latest avrdude.
But first difference that i see it's a that your bootloader return 
correct buffer size for memory access with better h/w and s/w version.
Maybe my problem something near at this issue. So will investigate 
around this.

> > Programmer supports the following devices:
> >     Device code: 0x80
> >     Device code: 0x75
> >     
>
> That's suspicious.  A bootloader should never announce support for
> more than a single MCU type.
>   

How i understand from playing with avrdude and bootloader device code 
must match with MCU code given to avrdude.
So in my case, it success with code 0x75, other one just ignored.

> > avrdude.exe: current erase-rewrite cycle count is 808530488 (if being
> > tracked)
> >     
>
> That means the last four EEPROM cells read back as 0x38, 0x32, 0x31,
> 0x30 (or ASCII "8210").  Is your EEPROM filled with data?
>
>   
Yes. It's contain of my EEPROM at update process.
Also this is don't matter now - i don't give command for avrdude to 
rewrite EEPROM yet, because I stuck on flash updating process.

Thank you for a help.
Can i ask you to run your avrdude again with additional -vvvv option and 
sent result full log to my email? Maybe it can help me to compare my 
incorrect updating process with your correct one.


--
Best regards,
Alex Eremeenkov