Re[3]: [avrdude-dev] AVR910 Programming speed

[Theodore A. Roth]

On Mon, 7 Apr 2003, Bertolt Mildner wrote:

:) Thanks!
:)
:) But your changes caused verification to fail!

Yes, I can see that. Thanks for catching it.

:)
:) The problem is in avr910_paged_load().
:) Replacing the line #595:
:)
:)      avr910_recv(pgm, &m->buf[addr * rd_size], rd_size);
:)
:) with
:)
:)      for (run = rd_size - 1; run >= 0; run--) {
:)        avr910_recv(pgm, &m->buf[(addr * rd_size) + run], 1);
:)      }
:)
:) fixes it for me. Byte pairs (words) are received in reverse order.

Well, that's what I get for not splitting writing separate functions
for flash and eeprom.

I'm not too keen on the use of a loop here. It kind of obfuscates what
is going on. Can you try the attached patch and see if it fixes it? (I
don't have my avr910 board here at work.)

:)
:) Programming time is now down to about 2:15, which is much better than
:) before but still not very close to the 1:10 with AVR Studio.
:)
:) I really wonder what AVR Studio does different!?

I don't have a clue...

Ted Roth

avr910-msb-paged.diff
Description: Text document