How to deal with the unconfirmed github issues?

[mcuee]

Right now the number of issues have some down to a more healthy number of around 100.

There are still 23 unconfirmed issues, how shall we deal with them?
https://github.com/avrdudes/avrdude/issues?q=is%3Aopen+is%3Aissue+label%3Aunconfirmed

Among them, quite some are related to xmega.

[mcuee]

Ref: xmega relared open issues (Updated on 24-September-2022)

Unconfirmed issues
No more.

Confirmed as bug

• [bug #43839] unable to verify device signature or program boot memory on xmega16d4 from ELF file #365
• flip2: ATxmega32A4U DFU bootloader does not work even though Atmel FLIP 3.47 works fine. #1082 (on 21-Aug-2022)
• [bug #43577] Extended address not sent with avr109 #360 (on 24-Sept-2022)
• [bug #51117] Problems with extended address (>128K) and buffer size #454 (on 24-Sept-2022)

Enhancement request:

• [bug #29896] Add support for XMEGA specific AVR109 protocol extensions used by XBoot #200 (enhancement)
• [patch #9563] New feature: ability to reset FLIPv2 targets #733 (enhancement request, with patch)

Recently closed issues:

• [bug #29264] avrdude upload to xmega using stk500 protocol using byte addresses rather than word addresses #190 (should work with the right bootloader)
• [bug #27254] Cannot program ATXMega with AVRDUDE using AVR911 AVROSP mode #164
• [bug #33160] JTAGICE mkII in JTAG mode, target not started correctly #226
• [bug #34994] avrdude does not send PDI offsets when erasing XMEGA chip #243
• [bug #39033] Support for ATxmegaXXXA4U devices #289
• [bug #40112] 6.0.1 - 911 flash upload fails #304
• [bug #40493] Can't program boot section of xmega256a3 #310
• [bug #40518] Using application with AVR109 on Xmega does not work as expected #311
• [bug #40702] fix: enable AVRISPmkII to read prodsig #314
• *[bug #41878] AVR Dragon JTAG: ATxmega256A3BU hangs, when programming or reading device signature #337
• [bug #45826] ATMEL-ICE content mismatch on Linux and OS X #391 (also affect other AVR chips, closed as invalid on 19-Aug-2022).
• [bug #46843] avrdude=>6 fails to write from 128k-mark onwards, loops back to 0x0000 #404
• [bug #49387] ATXmega32E5 Flash section address wrong #435

[dl8dtl]

Thanks for all your (not only you but everyone involved) work dealing with all the old issues.

I've set #360 as an enhancement request, since it asks for a protocol addition.

I'm really not sure how relevant any AVR32 patches are, given that even Atmel/Microchip went away from them now (the EDBG chips were the last domain of UC3 devices). There's a patch, but I don't know of anyone who would be able to even verify it works.

I can try looking at the remaining unconfirmed items piecewise.

[mcuee]

Great that you can start to take a look at those unconfirmed items one by one.

And yes I think we can close the AVR32 related enhancement request based on your comments and the reality that Microchip will concentrate on ARM and PIC32 (even the MIPS based PIC32 is going lower priority) and stopped the development of AVR32 long ago.

[mcuee]

So I just closed the few AVR32 related feature requests.

• [bug #27501] [Feature Request]: Support AVR32 #166
• [bug #34767] Avrdude to enable AVR Dragon to program AVR32 parts #239
• [patch #9573] patch to add atmelice avr32 programming support to avrdude #735
• Extend Flip support for UC3A devices #977

[mcuee]

@dl8dtl
Do we really need to take care of the long obsoleted AT90S parts?

• [bug #49082] avrdude 6.3 fails with AT90S8515 #433 (AT90S8515)
• [bug #27207] very slow with JTAGICE MKII using jtag2isp #163 (AT90S2313)
• [bug #58251] AT90S8515 STK500v2 problem: only odd bytes are programmed / programming only works in terminal mode. #500 (AT90S8535)

[MCUdude]

I'd argue that we could drop support for them if we can get rid of legacy code written specifically for these parts. And if one have to use one of these 20+ year-old parts, you could always just use an older version of Avrdude.

[mcuee]

Right now the un-confirmed issues have reduced to 13.
https://github.com/avrdudes/avrdude/issues?q=is%3Aopen+is%3Aissue+label%3Aunconfirmed

I think quite some of them may not be relevant any more. I took the liberty and closed them. Users can always re-open if there are similar issues happening again.

• [bug #52653] device initialization often fails in FT232R syncbb mode, due to glitchy start sequence #464 (I tested ft232r and no such issue)
• [bug #43811] Writing a hex file using a ft2232H programmer fails when poll after a page write does not come back as expected. #364 (I tested FT2232H and no such issue).
• [bug #40145] 6.0.1 - STK500v2 fails on uploads to parts with pages greater than 256b #307 (I have not found such issue with STK500v2) --> it turns out the issue is real and replaced with avrftdi.c: paged write can fail at addr 0 #1073 and fixed by Correct flash paged write for avrftdi.c #1074.

[mcuee]

Linux gpio related issues:

• [bug #51320] linuxgpio and linuxspi problem: 0x00 written as 0xFF when byte count is less than 0x10 in hex file lines #455 (tested to be a real bug)
• [bug #51497] avrdude not writing high byte using linuxgpio when data byte count is less than 0x10 #458 (duplicate of 455)
• [bug #56686] Verification error at 0x0000, 0x00 != 0x0c #490 (I have tested it and no issue found -- so I closed it for now. Users can re-open if similar issues popped up).

[mcuee]

Right now the un-confirmed issues are related to xmega.
https://github.com/avrdudes/avrdude/issues?q=is%3Aopen+is%3Aissue+label%3Aunconfirmed

[mcuee]

Finally I am able to program the ATxmega32D4 breakout board using an AVRISP mkii clone. I am able to confirm that #365 is a bug and #243 is not valid (Closed).

[mcuee]

I closed #289 as invalid today. I just got an ATxmeag32A4U breakout board and there is no issue to flash the chip using an AVRISP mkii clone.

[mcuee]

I closed the very old issue #190 yesterday (27-Auguest-2022) since there are reports of successing using Arduino compatible bootloader for xmega.

[mcuee]

#391 was closed by @MCUdude two days ago.

[mcuee]

Adding new issue related to xmega with FLIP2 as a bug.

• flip2: ATxmega32A4U DFU bootloader does not work even though Atmel FLIP 3.47 works fine. #1082

[mcuee]

As of today 28-August-2022, there are 76 issues remaining, 38 are enhancement requests, 26 are bugs, 5 are unconfirmed and 10 are related to documentation (some issues may have more than one label).

[mcuee]

@MCUdude
Thanks for closing #310, #337 and #404 as invalid. Now we only have two unconfirmed issues, related to AVR109 bootloader for the ATxmega256A3 or similar xmega.

• [bug #43577] Extended address not sent with avr109 #360
• [bug #51117] Problems with extended address (>128K) and buffer size #454

Now that you have the ATxmega256A3BU Xplained board, please help to test the above two issues when you got a chance. Thanks,

[mcuee]

@MCUdude
Thanks to confirm that #360 and #454 are indeed bugs. Now we do not have any unconfirmed issues. Great.