Bug in AVR8X XPlained Mini FW preventing Avrdude connection

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

Bug in AVR8X XPlained Mini FW preventing Avrdude connection

JanEgil.Ruud
Hi all,

During recent work we have discovered a bug in the mEDBG FW, used on AVR8X XPlained Mini kits (with UPDI), that is preventing Avrdude from connecting. The error is fixed in mEDBG UPDI FW version 1.12 (and later).

The error message you get is "avrdude: jtag3_edbg_recv(): Inconsistent fragment number; expect 1, got 0", but what is causing the issue is the CMSIS-DAP commands sent in jtag3_edbg_prepare() and jtag3_edbg_signoff(). These command are superfluous on mini kits and could be skipped by adding a test if programmer id ends with "mini_updi".

But, there are not a lot of AVR8X XPlained Mini kits in circulation, and even fewer Avrdude users with AVR8X XPlained Mini kits. I'm a bit reluctant to implement the workaround in Avrdude since Avrdude really does nothing wrong, and everybody with a AVR8X XPlained Mini kits also has Atmel Studio which will upgrade the FW once it's out (since Avrdude hasn't had support for AVR8X XPlained Mini kits until now).

Any views on this? If not I'll leave it at this. At least you are aware of it now.

-Jan Egil
_______________________________________________
avrdude-dev mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/avrdude-dev
Reply | Threaded
Open this post in threaded view
|

Re: Bug in AVR8X XPlained Mini FW preventing Avrdude connection

Joerg Wunsch
As [hidden email] wrote:

> Any views on this? If not I'll leave it at this. At least you are
> aware of it now.

I agree to leave it as it is.  The number of Xplained Mini users
with UPDI devices is by now probably neglicible, and once their
firmware is upgraded, nobody would notice anymore.

> The error message you get is "avrdude: jtag3_edbg_recv():
> Inconsistent fragment number; expect 1, got 0", but what is causing
> the issue is the CMSIS-DAP commands sent in jtag3_edbg_prepare() and
> jtag3_edbg_signoff(). These command are superfluous on mini kits and
> could be skipped by adding a test if programmer id ends with
> "mini_updi".

As a stopgap measure, maybe ignore the failure status in these
commands if -F is given?  This option is already used to skip a number
of consistency/plausibility checks otherwise, and "just proceed".

--
cheers, Joerg               .-.-.   --... ...--   -.. .  DL8DTL

http://www.sax.de/~joerg/
Never trust an operating system you don't have sources for. ;-)

_______________________________________________
avrdude-dev mailing list
[hidden email]
https://lists.nongnu.org/mailman/listinfo/avrdude-dev