Re: [avrdude-dev] [bug #28066] dragon_jtag and atxmega128A1 fails chip erase

[Uwe Bonnes]

>>>>> "Joerg" == Joerg Wunsch <address@hidden> writes:

    Joerg> Update of bug #28066 (project avrdude):

    Joerg>                   Status: None => Fixed Assigned to: None =>
    Joerg> joerg_wunsch Open/Closed: Open => Closed

    Joerg>     _______________________________________________________

    Joerg> Follow-up Comment #2:

    Joerg> Finally, AVR067 has been updated to document the CMND_XMEGA_ERASE
    Joerg> command, so I rewrote the chip erase functionality for Xmega
    Joerg> parts.  There has been a hidden Xmega chip erase in
    Joerg> jtagmkII_paged_write() before (where the author of that code
    Joerg> simply didn't knew it's been a chip erase), so I reverted that
    Joerg> one back to the regular auto_erase functionality.

As of today, grepping through the SVN trunk sources
> find | xargs grep CMND_ERASEPAGE
./.svn/text-base/jtagmkII_private.h.svn-base:#define CMND_ERASEPAGE_SPM 0x0D
./jtagmkII_private.h:#define CMND_ERASEPAGE_SPM 0x0D

only brings up the definition of the command, but no usage.
This coincides with my observation, that always the -e option had to be
given when reprogramming the flash like
/avrdude -p atxmega256a3b -P usb -c dragon_jtag -y -e -U
flash:w:l_motctl.hex

Otherwise verify errors show up.

Is there any fundamental issue with using CMND_ERASEPAGE_SPM? Or should I
try to come up with a patch?

Bye
-- 
Uwe Bonnes                address@hidden

Institut fuer Kernphysik  Schlossgartenstrasse 9  64289 Darmstadt
--------- Tel. 06151 162516 -------- Fax. 06151 164321 ----------