Re: [avr-gcc-list] boot vector address for ATMega32 in datasheet correct ?

[Ewout Boks]

Erik Christiansen wrote:
> On Thu, Dec 18, 2003 at 09:45:55AM +0100, Ewout Boks wrote:
>
> > I do not understand this, as the ATMega32 has 32K of flash and address 
> > 0x37ff is exactly at 16K, ie halfway the application flash of the ATMega32.
>
>
> Is this a case of the avr using word adressing, and gnu tools remaining
> with the standard byte addressing? Here is what I did for the mega64,
> yesterday, for the same purpose:

I do suspect this. In the datasheet on page 253, table 100 the boot size 
is listed in words. The address type is not indicated, though. If using 
byte addressing (like the gnu toolchain) then my question is valid. If 
word addressing is used, then (word *) 0x3800 would be at (byte *)0x7000 
 and things are correct.



Merry Christmas and a good holiday!

Ewout


> MEMORY
> {     
>   text   (rx)   : ORIGIN = 0, LENGTH = 62K
>   boot   (rx)   : ORIGIN = 62K, LENGTH = 2K
>   data   (rw!x) : ORIGIN = 0x800100, LENGTH = 4K
>   eeprom (rw!x) : ORIGIN = 0x810000, LENGTH = 2K
> }
>
> ...
>
>   .boot :
>   {
>     *(.boot_vectors)
>     *(.boot)
>   } > boot
>
> The length of the boot (flash) sector is 1K words, so we set 2K bytes.
>
> Sorry if this is short, but my Christmas holiday started some hours ago,
> and I'm still chained to the desk.
>
> Erik