avrdude 8.1 release discussion

[mcuee]

Just wondering if we want to have another release, say by end of June 2025, or even earlier.

avrdude 8.0 was released on Aug 25, 2024 and seems to be pretty stable.

[mcuee]

Bugs to be sorted out:
??

Enhancement nice to have:
??

[mcuee]

We seem to have quite some good stuff after 8.0 release already, especially related to PICKit 5 (and PICkit 4 in PIC mode) and new parts. It is also nice to fix the long standing avrdoper bug.

https://github.com/avrdudes/avrdude/blob/main/NEWS

[MCUdude]

A new PICkit programmer has just been released! It's called the PICkit Basic and retails at 29 USD. To me, it looks like a SNAP with a case and USB C connector. @MX682X, I'd bet that this is fairly similar to the PICkit5.

[MCUdude]

It seems like it's possible to switch between the "MPLAB" firmware and some sort of CMSIS-DAP firmware:
https://pypi.org/project/pycmsisdapswitcher/

Do you have any details to share @xedbg? 🙂

[xedbg]

Don't get too excited in the short-term.
Yes - its about the new hardware tool has been released this week (PICkit basic) - it is indeed more or less an upgraded Snap in an enclosure.
Natively it runs the "gen4" proprietary firmware, but the switcher enables it to switch to CMSIS-DAP mode. Essentially this is to enable support for third party IDEs to work with ARM devices, so it does not [yet] support the AVR vendor commands or anything special. It is a CMSIS-DAP v2 implementation (so bulk, not hid). The switcher also works for PKOB4 kits (Curiosity Ultra / Pro, but not EDBG).

[MX682X]

If it is indeed running the pickit command script it should be easy to add support, in the best case just by adding the pids (plus extending the avr mode / pic mode switch). Has anyone ordered one already?
As it is out of stock this will end up as developement by proxy.

Edit: just checked the switcher docs, why do they have 7 (8) pids for this debugger?!
If the use of CDC changes the PID this will open a big can of worms.

[MX682X]

on a side note - I can try to implement a preliminary support for the pickit basic if I get a suggestion on the naming (as in for the -c option).
maybe just pickit? or pk_basic?
@stefanrueger how would you call the new programmer?

[stefanrueger]

@MX682X Very cool! My preference usually is to be as close to the product name as possible, e.g., pickit_basic. Using the id list it's quite easy (even for the user in their private .avrduderc) to maintain personal nicknames.

[stefanrueger]

I'd be in favour of a 8.1 release end of May/start of June. I won't have bandwidth in July at all so wouldn't be available for fixes in case something goes wrong with the 8.1 release. So perhaps freezing features/PRs start of May?

[MCUdude]

I'm fine with this, but I don't have as much time for testing as I did when we prepared 8.0. We have a four-week-old girl we have to take care of, along with the two other kids we have. But I'll do my best! A list of parts and programmers you want to have tested is always helpful. Is our test script up to date?

[mraardvark]

We have a four-week-old girl

Congrats on the new arrival! 🎉

[stefanrueger]

Heartfelt Congratulations, Hans!

@MCUdude As to the tests, could you sample/source one of the AVR32SDnn parts? That might be worthwhile testing. We may have overlooked details in the data sheets. And, all the new -c programmers with existing physical programmers and (if you have it) the PICKit Basic programmer. I am sure, however, that @MX682X will have done a great job testing the modes and techniques deployed with the new -c programmers, so it's more of an independent sanity spot check. As to regression testing I am hoping that @mcuee might find some time to spot test OS/programmer/part combos. We will never be able to test all combinations though. I am still interested in the comprehensive ALL memory backups/restores with the test8 script. All in all my gut feeling is that we haven't made many changes in critical parts of the AVRDUDE suite so I am not worried too much about things breaking for 8.1.

[mcuee]

Congratulations to Hans!

@stefanrueger

Yes it is good to have a release and I should be able to carry out some regression testing.

[stefanrueger]

I'd be happy with a PR freeze soon, but would want the following looked at before we do so:

• linuxgpio programmer do not release RESET line on failure. #1972: Should be in hand, @MCUdude?
• How does hvupdi_variant 3 work? (AVR32SDnn) #1957/Cater for new UPDI_ENABLE_RESET_HS method #1967: Should we just close these? Is there something we can/should do?
• Backup Arduino Nano Every #1961: Still unconfirmed; can you reproduce this @mcuee / @MCUdude?
• Upload avrdude-gui artifacts for Windows #1946: What should we do about this?
• DEFAULT_SPI_PORT appears to be undefined #1923: Should we change the C code to use /dev/spidev0.0:/dev/gpiochip0 if DEFAULT_SPI_PORT is undefined? I don't know how to replace that variable in cmake or automake...

Other than that this should be ready for testing and, then, v8.1.

[MX682X]

the github release notes were exactly what I had in mind

[mcuee]

@MX682X

Also, while testing I noticed that only the mingw version of avrdude supports pickit5 - this should be added to the release notes.
The MSVC version can't open the usb port.

Just wondering if you have debugged the issue or not. This may be due to the known limitation.

• avrdude MSVC build problem with USB composite device (avrdude-libusb library issue) #968

That is a bug with the dependency. Unfortunately it is not easy to fix without the help from @mariusgreuel.

• avrdude-libusb issue with USB Composite Device libusb#5

Wiki updated:

1. https://github.com/avrdudes/avrdude/wiki/Getting-Nightly-Builds-for-AVRDUDE#windows-static-linking-to-dependancies
2. https://github.com/avrdudes/avrdude/wiki/Building-AVRDUDE-for-Windows-using-Visual-Studio#know-limitations-of-visual-studio-build

[MX682X]

Just wondering if you have debugged the issue or not. This may be due to the known limitation.

No I haven't, because I already guessed it was part of the known limitations.

[mcuee]

@MX682X

Just wondering if your test results of PICKit 5 are similar to my test results of PICKit 4 in MPLAB (PIC) mode or not. Thanks.

Windows mingw64 version is working.

PS C:\work\avr\avrdude_bin> .\avrdude.exe -C .\avrdude.conf -c pickit4_mplab_isp -p m328p -U .\m328pfull.hex -v
Avrdude version 8.0-20250529 (98784e02)
Copyright see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

System wide configuration file is C:\work\avr\avrdude_bin\avrdude.conf

Using port            : usb
Using programmer      : pickit4_mplab_isp
Usbdev_open(): found MPLAB? PICkit 4, serno: BUR200973052
AVR part              : ATmega328P
Programming modes     : SPM, ISP, HVPP, debugWIRE
Programmer type       : pickit5
Description           : MPLAB(R) PICkit 4 in ISP Mode
Application version   : 02.04.22
Serial number         : BUR20097xxxx
Target Vdd: 4.72 V, target current: 0 mA
External voltage detected: will not supply power

AVR device initialized and ready to accept instructions
Device signature = 1E 95 0F (ATmega328P, ATA6614Q, LGT8F328P)
Auto-erasing chip as flash memory needs programming (-U flash:w:...)
specify the -D option to disable this feature
Target successfully erased
Erased chip
Reading 924 bytes for flash from input file m328pfull.hex
in 1 section [0, 0x39b]: 1 page and 100 pad bytes
Writing 924 bytes to flash
Writing | ################################################## | 100% 0.64 s
Reading | ################################################## | 100% 0.34 s
924 bytes of flash verified

Avrdude done.  Thank you.

MSVC version is not working.

PS C:\work\avr\avrdude_bin> .\avrdude_msvc.exe -C .\avrdude_msvc.conf -c pickit4_mplab_isp -p m328p -U .\m328pfull.hex -v
Avrdude_msvc version 8.0-20250529 (98784e02)
Copyright see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

System wide configuration file is C:\work\avr\avrdude_bin\avrdude_msvc.conf

Using port            : usb
Using programmer      : pickit4_mplab_isp
Usbdev_open(): found MPLAB« PICkit 4, serno: BUR200973052
AVR part              : ATmega328P
Programming modes     : SPM, ISP, HVPP, debugWIRE
Programmer type       : pickit5
Description           : MPLAB(R) PICkit 4 in ISP Mode
Application version   : 02.04.22
Serial number         : BUR20097xxxx
Error: reading data memory failed
Target Vdd: 4.71 V, target current: 0 mA
External voltage detected: will not supply power

AVR device initialized and ready to accept instructions
Device signature = 1E 95 0F (ATmega328P, ATA6614Q, LGT8F328P)
Auto-erasing chip as flash memory needs programming (-U flash:w:...)
specify the -D option to disable this feature
Target successfully erased
Erased chip
Reading 924 bytes for flash from input file m328pfull.hex
in 1 section [0, 0x39b]: 1 page and 100 pad bytes
Writing 924 bytes to flash
Writing | -------------------------------------------------- | 0% 0.00 s
Error: reading data memory failed
Error: reading from programmer failed
Error: sending script done message failed
Error: reading data memory failed
Error: reading from programmer failed
Error: sending script done message failed
Error: wrote -138 out of 109 bytes, err = No error
Error: sending script failed
Error: sending script with download failed
Error: invalid script pointer passed
Error: sending script with download failed
 *** failed
Error: unable to write flash (rc = -1)
Error: wrote -138 out of 26 bytes, err = No error
Error: sending script failed
Error: sending script failed
Error: could not end programming, aborting
Error: wrote -138 out of 29 bytes, err = No error
Error: sending script failed
Error: sending script failed

Avrdude_msvc done.  Thank you.

[MX682X]

Tested with the basic for a bit more variation, iirc Pickit 5 behaved the same (Windows 10): Mingw64 works, Visual Studio can't open libusb devices (I tried different drivers with zadig at some point, with and without admin, no success)

Mingw64

'C:\msys64\home\usr\avrdude\build_mingw64\src\avrdude.exe' -C '.\avrdude.conf' -c pickit_basic_isp -p m32u4 -vvv
Avrdude version 8.0-20250529 (98784e02)
Copyright see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

System wide configuration file is C:\msys64\home\usr\avrdude\build_mingw64\src\avrdude.conf

Using port            : usb
Using programmer      : pickit_basic_isp
Pickit5_open("usb")
Usbdev_open(): found MPLAB? PICkit Basic ICD, serno: 020126001RYN000034
AVR part              : ATmega32U4
Programming modes     : SPM, ISP, HVPP, JTAG

Memory           Size  Pg size
------------------------------
eeprom           1024        4
flash           32768      128
lfuse               1        1
hfuse               1        1
efuse               1        1
lock                1        1
prodsig/sigrow     24        1
sernum             10        1
io                224        1
sram             2560        1
signature           3        1
calibration         1        1

Variants          Package  F max   T range        V range
----------------------------------------------------------------
ATmega32U4-AU     TQFP44   16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]
ATmega32U4-AUR    TQFP44   16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]
ATmega32U4-MU     QFN44    16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]
ATmega32U4-MUR    QFN44    16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]
ATmega32U4RC-AU   TQFP44   16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]
ATmega32U4RC-AUR  QFP44    16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]
ATmega32U4RC-MU   QFN44    16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]
ATmega32U4RC-MUR  QFN44    16 MHz  [-40 C, 85 C]  [2.7 V, 5.5 V]

Programmer type       : pickit5
Description           : MPLAB(R) PICkit Basic in ISP Mode
Pickit5_get_fw_info()
Application version   : 00.00.81
Serial number         : 020126001RYN000034
Pickit5_initialize()
Found scripts at namepos 83
Pickit5_set_sck_period()
Pickit5_send_script(scr_len: 5, param_len: 4, data_len: 0)
Pickit5_program_enable()
Pickit5_send_script(scr_len: 47, param_len: 0, data_len: 0)
Pickit5_read_dev_id()
Pickit5_send_script(scr_len: 66, param_len: 0, data_len: 0)

AVR device initialized and ready to accept instructions
Avr_signature(pickit_basic_isp, m32u4)
Reading |                                                    | 0% 0.00 s
Avr_read_mem(pickit_basic_isp, m32u4, signature, NULL)
Pickit5_read_dev_id()
Pickit5_send_script(scr_len: 66, param_len: 0, data_len: 0)
Reading | ################################################## | 100% 0.02 s
Device signature = 1E 95 87 (ATmega32U4)
Pickit5_program_disable()
Pickit5_send_script(scr_len: 2, param_len: 0, data_len: 0)
Pickit5_close()

Avrdude done.  Thank you.

and VS output

'C:\Users\usr\source\repos\avrdude\build_msvc\src\RelWithDebInfo\avrdude.exe' -C '.\..\avrdude.conf' -c pickit_basic_isp -p m32u4 -vvv
Avrdude version 8.0-20250529 (98784e02)
Copyright see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

System wide configuration file is C:\Users\usr\source\repos\avrdude\build_msvc\src\avrdude.conf

Using port            : usb
Using programmer      : pickit_basic_isp
Pickit5_open("usb")
Error pickit5_open() pickit5.c 754: no device found matching VID 0x04d8 and PID list: 0x9055
Avrdude done.  Thank you.

[stefanrueger]

OK, my plan is to merge #1976, #1977 and #1981 once @MCUdude or @mcuee confirm #1981 is OK. After that development for 8.1 would be frozen except for fixes for encountered problems during testing. I suggest closing #1957 and #1967 (no movement) and #1980 (won't fix). Still planning to release 8.1 end of May or start of June. Would that work for you @mcuee, @MCUdude, @dl8dtl and @mariusgreuel?

Not planning on considering Issue #1982 for v8.1: that issue might turn into a PR once it can be reproduced and the suggested fix (moving an assignment into a loop in a central posix serial receive function) is confirmed to not cause regressions. As ser_recv() is a pretty central function we could jeopardise a whole release by hastily moving on that, so suggest to treat that later.

[mcuee]

Looks like a good plan. Thanks.

[mcuee]

Closed #1957 and #1967 for now. We can always reopen later when necessary.

• How does hvupdi_variant 3 work? (AVR32SDnn) #1957
• Cater for new UPDI_ENABLE_RESET_HS method #1967

Let's see if @dl8dtl has any objection to close #1980 as "won't fix".

• terminal mode "d" command is ambiguous #1980

And I agree to put #1982 after 8.1 release.
Edit: #1982 has been merged.

• ser_posix.c: ser_recv times out for low baud rates (e.g 300) #1982

[mcuee]

@MCUdude

Please take a look at the following new WiKi page for autotools build instructions and see if you have some comments to make it bettter.
https://github.com/avrdudes/avrdude/wiki/General-Build-Instructions-Using-Auto%E2%80%90tools

[stefanrueger]

Suggesting to freeze the code now for v8.1 testing. Have added #1982 into the mix as low-risk.

Good luck with testing

[mcuee]

@dl8dtl

With regard to #1946, just wondering what is your proposal to move forward.

• Upload avrdude-gui artifacts for Windows #1946

I can prepare a shot write up on the building instruction in the comments here (build instruction and screenshots).

I have documented the build process and provide the screenshots in the above ticket.

I have also mentioned the potential issues the user may encounter.

[mcuee]

@stefanrueger

I have closed #1653 as it is still supported by Homebrew since they do not have good replacement for libelf for macOS users.

• libelf deprecation on macOS #1653

[mcuee]

Testing towards 8.1 release, last update on 11-June-2025.

Updated on 21-May-2025

Building avrdude git under all kinds of OS and reports the compiler warnings. Done.

• Compiler warnings #1988

@stefanrueger has done the "fixes" in PR #1989.

• Supress compiler warnings #1989

Simple testing of different programmers and bootloaders.

Mainly under Windows and Linux.

1. usbasp --> Done
2. urboot/urclock --> Done
   Including the test of the new feature in PR Autogenerate urboot bootloaders #1974.

• Autogenerate urboot bootloaders #1974

3. Optiboot with -c arduino and `-c urclock) --> Done
4. Arduino Nano Every --> Done
5. AVRISPmkii --> Done
6. Wiring bootloader --> Done

Updated on 1-June-2025

7. PICKit 4 in AVR mode, ISP for ATmega328P (Windows mingw64/MSVC mode both work fine)
8. PICKit 4 in PIC mode, ISP for ATmega328P (Windows MSVC build does not work)
9. Microchip ATtiny810 xPlained Pro

Updated on 2-June-2025

10. AVR JTAG mkII ISP mode with ATmega328P.

Minor finding on the test8 script help texts, fixed in 231bb28

Updated on 4-June-2025

11. STK600 ISP mode with ATmega328P.

Updated on 7-June-2025

12. linuxspi on Raspberry Pi 400 -- working fine
13. linuxgpio becomes not stable on my Rasperry Pi 400 but I think it is not an issue with avrdude, but rather connection stability issues

• linuxgpio do not work properly #2007

Minor compiler warning from Raspberry Pi OS gcc 12.2.0 fixed in git commit b031bd0.

Updated on 8-June-2025

14. Microchip AVR64EA48 Curiority Nano Kit -c pkobn_updi.
15. Microchip ATtiny416 Xplained Nano kit
16. serialupdi -- need to see how to use the -a option for test8.

Updated on 9-June-2025

17. serialupdi test8 done

18. Testing under FreeBSD: usbasp, urboot and AVRISP mkII.

Known issue: serialupdi issue with CH340).

• avrdude serialupdi issues under FreeBSD 14.0 with CH340 USB to TTL converter #1639

Two new issues -- not a regression, therefore not a blocker for avrdude 8.1 release.

• avrdude issues with ATtiny416 Xplained Nano under FreeBSD 14.2 Release #2010
• avrdude issues with pkobn_updi under FreeBSD 14.2 Release #2011

19. Testing under NetBSD -- usbasp, urboot, serialupdi

Known OS limitations -- USB HID based programmer will not work

20. Testing under OpenBSD -- usbasp, urboot, serialupdi, AVRISP mkII

Known OS limitations -- USB HID based programmer will not work

New issue -- not a regression, therefore not a blocker for avrdude 8.1 release.

• avrdude issue with urboot autobaud under OpenBSD 7.7 for Arduino Uno clone with ATmega16U2 #2012

Updated on 11-June-2025

More tests under FreeBSD/OpenBSD/NetBSDs, with stk600, AVR JTAGICE mkII and urboot.

[mcuee]

No issues with CH340 based serialupdi under OpenBSD 7.7 Release.

[root@openbsdn4200 tools]# ./test-avrdude -d0 -v -p "-c serialupdi -P /dev/ttyU0 -p avr64dd32"
Testing /usr/local/bin/avrdude 8.0-20250608 (de2d1ad5)
Prepare "-c serialupdi -P /dev/ttyU0 -p avr64dd32" and press 'enter' or 'space' to continue. Press any other key to skip
���   5.285 s: fuse access: clear, set and read eesave fuse bit
���   4.257 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
���   3.443 s: chip erase
���   7.047 s: flash -U write/verify holes_rjmp_loops_65536B.hex
���   5.967 s: flash -T write/verify holes_rjmp_loops_65536B.hex
���   3.298 s: eeprom check whether programmer can flip 0s to 1s
���   5.427 s: eeprom -U write/verify holes_pack_my_box_256B.hex
���   7.605 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_256B.hex
���   4.935 s: chip erase and spot check flash is actually erased
���   3.446 s: spot check eeprom is erased, too
���   4.647 s: usersig -T/-U write/read random_data_32B.bin

[mcuee]

No issues with usbasp under OpenBSD 7.7 Release as well.

But there seems to be an issue for urclock/urboot under OpenBSD 7.7.

• avrdude issue with urboot autobaud under OpenBSD 7.7 for Arduino Uno clone with ATmega16U2 #2012

[mcuee@openbsdn4200 tools]$ sudo su
openbsdn4200# bash

[root@openbsdn4200 tools]# pwd
/home/mcuee/build/avr/avrdude_81_rc/tools

[root@openbsdn4200 tools]# ./test-avrdude -d0 -p "-c usbasp -p m328p" -v
Testing /usr/local/bin/avrdude 8.0-20250608 (de2d1ad5)
Prepare "-c usbasp -p m328p" and press 'enter' or 'space' to continue. Press any other key to skip
���   0.351 s: fuse access: clear, set and read eesave fuse bit
���   0.317 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
���   0.531 s: chip erase
���   2.634 s: flash -U write/verify holes_rjmp_loops_32768B.hex
���   1.169 s: flash -T write/verify holes_rjmp_loops_32768B.hex
���   0.406 s: eeprom check whether programmer can flip 0s to 1s
���   3.011 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
���   5.813 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex
���   1.392 s: chip erase and spot check flash is actually erased
���   0.397 s: spot check eeprom is erased, too

[root@openbsdn4200 tools]# avrdude -c usbasp -p m328p -U urboot:16mhz_autobaud_ce_ee -qq && echo OK
OK

[root@openbsdn4200 tools]# ./test-avrdude -d0 -v -p "-c urclock -P /dev/ttyU1 -p m328p"
Testing /usr/local/bin/avrdude 8.0-20250608 (de2d1ad5)
Prepare "-c urclock -P /dev/ttyU1 -p m328p" and press 'enter' or 'space' to continue. Press any other key to skip
���   2.721 s: chip erase
 8.0-20250608 (de2d1ad5)
���   5.820 s: flash -U write/verify holes_rjmp_loops_32768B.hex (failed command below)
$ avrdude -qq -c urclock -P /dev/ttyU1 -p m328p -Uflash:w:./test_files/holes_rjmp_loops_32768B.hex
Warning: attempt 1 of 10: not in sync
Warning: attempt 2 of 10: not in sync
Warning: attempt 3 of 10: not in sync
Warning: attempt 4 of 10: not in sync
Warning: attempt 5 of 10: not in sync
Warning: attempt 6 of 10: not in sync
Warning: attempt 7 of 10: not in sync
Warning: attempt 8 of 10: not in sync
Warning: attempt 9 of 10: not in sync
Warning: attempt 10 of 10: not in sync
Warning: programmer is not responding; try -x strict and/or vary -x delay=100
Error: unable to open port /dev/ttyU1 for programmer urclock
Error: sp_list_ports() failed!
���   7.181 s: flash -U write/verify rjmp_loops_for_bootloaders_32768B.hex
���   2.897 s: flash -T write/verify holes_rjmp_loops_32768B.hex
���   1.727 s: eeprom check whether programmer can flip 0s to 1s
���   2.958 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
���   5.820 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex (failed command below)
$ avrdude -qq -c urclock -P /dev/ttyU1 -p m328p -T "write eeprom ./test_files/holes_the_five_boxing_wizards_1024B.hex:a" -T flush -T "write eeprom ./test_files/holes_pack_my_box_1024B.hex:a"
Warning: attempt 1 of 10: not in sync
Warning: attempt 2 of 10: not in sync
Warning: attempt 3 of 10: not in sync
Warning: attempt 4 of 10: not in sync
Warning: attempt 5 of 10: not in sync
Warning: attempt 6 of 10: not in sync
Warning: attempt 7 of 10: not in sync
Warning: attempt 8 of 10: not in sync
Warning: attempt 9 of 10: not in sync
Warning: attempt 10 of 10: not in sync
Warning: programmer is not responding; try -x strict and/or vary -x delay=100
Error: unable to open port /dev/ttyU1 for programmer urclock
Error: sp_list_ports() failed!
���   8.331 s: eeprom -T write/verify lorem_ipsum_1024B.srec
���   3.427 s: chip erase and spot check flash is actually erased

One or more AVRDUDE "-c urclock -P /dev/ttyU1 -p m328p" tests failed. Do you want to retry this particular test? (y/n): n

[mcuee]

More tests under FreeBSD 14.2 -- no issues with AVRISP mkII, no issues with urboot.

mcuee@freebsd14:~/build/avrdude_81_rc/tools $ ./test-avrdude -v -p "-c avrispmkii -p m328pb"
Testing /usr/local/bin/avrdude 8.0-20250608 (de2d1ad5)
Prepare "-c avrispmkii -p m328pb" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.322 s: fuse access: clear, set and read eesave fuse bit
✅   0.299 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   0.384 s: chip erase
✅   0.739 s: flash -U write/verify holes_rjmp_loops_32768B.hex
✅   0.455 s: flash -T write/verify holes_rjmp_loops_32768B.hex
✅   0.307 s: eeprom check whether programmer can flip 0s to 1s
✅   0.559 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
✅   0.877 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex
✅   0.539 s: chip erase and spot check flash is actually erased
✅   0.337 s: spot check eeprom is erased, too

mcuee@freebsd14:~/build/avrdude_81_rc/tools $ avrdude -c avrispmkii -p m328pb -U urboot:16mhz_autobaud_ce_ee -qq && echo OK
OK

mcuee@freebsd14:~/build/avrdude_81_rc/tools $ ./test-avrdude -v -p "-c urclock -P /dev/ttyU0 -p m328pb"
Testing /usr/local/bin/avrdude 8.0-20250608 (de2d1ad5)
Prepare "-c urclock -P /dev/ttyU0 -p m328pb" and press 'enter' or 'space' to continue. Press any other key to skip
✅   1.564 s: chip erase
 8.0-20250608 (de2d1ad5)
✅   2.999 s: flash -U write/verify holes_rjmp_loops_32768B.hex
✅   1.575 s: flash -T write/verify holes_rjmp_loops_32768B.hex
✅   1.010 s: eeprom check whether programmer can flip 0s to 1s
✅   2.116 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
✅   3.689 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex
✅   2.179 s: chip erase and spot check flash is actually erased

[mcuee]

More tests under FreeBSD 14.2 -- no issues with AVR JTAGICE mkII clone (programmer only).

mcuee@freebsd14:~/build/avrdude_81_rc/tools $ sudo ./test-avrdude -v -t. -p "-c jtag2isp -p m328p"
Testing /usr/local/bin/avrdude 8.0-20250610 (9d7a893d)
Prepare "-c jtag2isp -p m328p" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.361 s: fuse access: clear, set and read eesave fuse bit
✅   0.335 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   0.416 s: chip erase
✅   3.388 s: flash -U write/verify holes_rjmp_loops_32768B.hex
✅   1.686 s: flash -T write/verify holes_rjmp_loops_32768B.hex
✅   0.359 s: eeprom check whether programmer can flip 0s to 1s
✅   0.926 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
✅   1.696 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex
✅   1.767 s: chip erase and spot check flash is actually erased
✅   0.506 s: spot check eeprom is erased, too

[mcuee]

More tests under FreeBSD 14.2 -- no issues with STK600 clone (programmer only).

mcuee@freebsd14:~/build/avrdude_81_rc/tools $ sudo ./test-avrdude -v -t. -p "-c stk600 -p m328p"
Testing /usr/local/bin/avrdude 8.0-20250610 (9d7a893d)
Prepare "-c stk600 -p m328p" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.360 s: fuse access: clear, set and read eesave fuse bit
✅   0.337 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   0.519 s: chip erase
✅   2.594 s: flash -U write/verify holes_rjmp_loops_32768B.hex
✅   1.248 s: flash -T write/verify holes_rjmp_loops_32768B.hex
✅   0.352 s: eeprom check whether programmer can flip 0s to 1s
✅   0.713 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
✅   1.185 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex
✅   1.430 s: chip erase and spot check flash is actually erased
✅   0.404 s: spot check eeprom is erased, too

[mcuee]

@stefanrueger @MCUdude @dl8dtl

I guess we are more or less done with the PRs and consider avrdude 8.1 reaching the feature-freeze phase. So we can start the test cycle.

[stefanrueger]

start the test cycle

I have concentrated on checking new features on a AVRDUDE level. I do not really have a lot of parts and/or programmers with me while on extended travel, so am playing mostly with -c dryrun. I have detected two bugs which I am going to fix. This makes @mcuee and me the joint leaders (2 problems found each) in the score board for pre-release testing 😄 @dl8dtl and @MCUdude: you may want to put the skates on!

[MCUdude]

@MCUdude As to the tests, could you sample/source one of the AVR32SDnn parts?

I'll see what I can do!

[mcuee]

As to regression testing I am hoping that @mcuee might find some time to spot test OS/programmer/part combos. We will never be able to test all combinations though. I am still interested in the comprehensive ALL memory backups/restores with the test8 script.

@stefanrueger

Yes I am carrying out some simple tests of various programmers under Windows and Linux. I will carry out a few sanity tests under FreeBSD/OpenBSD/NetBSD as well.

I will try to use test8 script as well.

[MCUdude]

@MCUdude As to the tests, could you sample/source one of the AVR32SDnn parts?

AVR-SD parts and a PKbasic are on their way! They'll be here by the end of the week

[MCUdude]

First AVR32SD28 test. Works with an up-to-date PICkit Basic, but not with a SerialUPDI. @dbuchwald is this something you can look into? It appears that Pymcuprog has been updated to support NVMv6.

AVR32SD28 + PICkit Basic

$ ./test8 -r1 pickit_basic_updi avr32sd28
Target successfully erased
Reading 33972 bytes for multiple memories from input file bak-pickit_basic_updi-avr32sd28.hex
Writing 256 bytes to eeprom
Writing | ################################################## | 100% 4.30 s 
Reading | ################################################## | 100% 0.20 s 
256 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 5.91 s 
Reading | ################################################## | 100% 2.21 s 
32768 bytes of flash verified
Writing 16 bytes to fuses, 16 bytes written, 16 verified
Writing 1 byte (0x00) to fuse0/wdtcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse1/bodcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse5/syscfg0, 1 byte written, 1 verified
Writing 1 byte (0x48) to fuse6/syscfg1, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse7/codesize, 1 byte written, 1 verified
Writing 1 byte (0x07) to fuse8/bootsize, 1 byte written, 1 verified
Writing 2 bytes (0x03 00) to fusea/pdicfg, 2 bytes written, 2 verified
Writing 4 bytes (0x5C C5 C5 5C) to lock, 4 bytes written, 4 verified
Writing 256 bytes to bootrow
Writing | ################################################## | 100% 0.15 s 
Reading | ################################################## | 100% 0.03 s 
Warning: bootrow verification mismatch
  device 0xff != input 0x40 at addr 0x0000 (error)
Error: bootrow verification mismatch
Writing 512 bytes to userrow/usersig
Writing | ################################################## | 100% 0.13 s 
Reading | ################################################## | 100% 0.03 s 
512 bytes of userrow/usersig verified

Avrdude done.  Thank you.
test8: avr32sd28 initialised
Hans-MacBook-Pro:tools hans$ avrdude -cpickit_basic_updi -pavr32sd28 -t
avrdude> read
Syntax: read <mem> <addr> <len> # Entire region
        read <mem> <addr>       # Start at <addr>
        read <mem>              # Continue with memory where left off
        read                    # Continue with most recently shown <mem>
Function: display memory section as hex dump

Both the <addr> and <len> can be negative numbers; a negative <addr> starts
an interval from that many bytes below the memory size; a negative <len> ends
the interval at that many bytes below the memory size.

The latter two versions of the command page through the memory with a page
size of the last used effective length (256 bytes default)
avrdude> part
AVR32SD28 with programming modes SPM, UPDI

Memory            Size  Pg size    Offset
-----------------------------------------
eeprom             256       32    0x1400 
flash            32768      512  0x800000 
fuses               16        1    0x1050 
fuse0/wdtcfg         1        1    0x1050 
fuse1/bodcfg         1        1    0x1051 
fuse5/syscfg0        1        1    0x1055 
fuse6/syscfg1        1        1    0x1056 
fuse7/codesize       1        1    0x1057 
fuse8/bootsize       1        1    0x1058 
fusea/pdicfg         2        1    0x105a 
lock                 4        1    0x1040 
prodsig/sigrow     128      128    0x1080 
signature            3        1    0x1080 
tempsense            4        1    0x1084 
sernum              16        1    0x1090 
bootrow            256       64    0x1100 
userrow/usersig    512      512    0x1200 
io                4160        1         0 
sram              4096      256    0x7000 
sib                 32       32         0 

Variants              Package  F max   T range         V range       
---------------------------------------------------------------------
AVR32SD28-SSOP/SPDIP  DIP28    20 MHz  [-40 C, 125 C]  [1.8 V, 5.5 V]
AVR32SD28-VQFN        QFP28    20 MHz  [-40 C, 125 C]  [1.8 V, 5.5 V]
avrdude> read bootrow
Reading | ################################################## | 100% 0.03 s 
0000  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0010  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0020  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0030  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0040  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0050  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0060  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0070  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0080  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0090  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00a0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00b0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00c0  40 40 40 40 40 20 40 20  20 20 20 20 40 40 20 20  |@@@@@ @     @@  |
00d0  40 40 20 40 40 20 40 40  20 20 40 40 20 20 40 20  |@@ @@ @@  @@  @ |
00e0  20 20 40 20 20 40 40 20  20 20 40 20 40 20 40 40  |  @  @@   @ @ @@|
00f0  20 20 40 20 20 20 40 20  20 62 6f 6f 74 72 6f 77  |  @   @  bootrow|

avrdude> q

Avrdude done.  Thank you.

$ ./test8 pickit_basic_updi avr32sd28

Processing -U ALL:r:bak-pickit_basic_updi-avr32sd28.hex:I
Reading multiple memories ...
 - eeprom          | ################################################## | 100% 0.20 s 
 - flash           | ################################################## | 100% 2.21 s 
 - fuses           | ################################################## | 100% 0.03 s 
 - fuse0/wdtcfg    | ################################################## | 100% 0.00 s 
 - fuse1/bodcfg    | ################################################## | 100% 0.00 s 
 - fuse5/syscfg0   | ################################################## | 100% 0.00 s 
 - fuse6/syscfg1   | ################################################## | 100% 0.00 s 
 - fuse7/codesize  | ################################################## | 100% 0.00 s 
 - fuse8/bootsize  | ################################################## | 100% 0.00 s 
 - fusea/pdicfg    | ################################################## | 100% 0.00 s 
 - lock            | ################################################## | 100% 0.01 s 
 - prodsig/sigrow  | ################################################## | 100% 0.06 s 
 - signature       | ################################################## | 100% 0.00 s 
 - tempsense       | ################################################## | 100% 0.06 s 
 - sernum          | ################################################## | 100% 0.00 s 
 - bootrow         | ################################################## | 100% 0.03 s 
 - userrow/usersig | ################################################## | 100% 0.03 s 
 - sib             | ################################################## | 100% 0.00 s 
Writing 34003 bytes to output file bak-pickit_basic_updi-avr32sd28.hex

Processing -T fact reset
performing chip erase; discarded pending writes to flash, EEPROM and bootrow
Target successfully erased
Caching | ################################################## | 100% 0.20 s 
Caching | ################################################## | 100% 0.00 s 
Caching | ################################################## | 100% 0.03 s 
Caching | ################################################## | 100% 0.04 s 
Synching cache to device ... done
after the next reset the part will have F_CPU = 4.000 MHz

Processing -U ALL:w:bak-pickit_basic_updi-avr32sd28.hex:I
Reading 33972 bytes for multiple memories from input file bak-pickit_basic_updi-avr32sd28.hex
Writing 256 bytes to eeprom
Writing | ################################################## | 100% 4.30 s 
Reading | ################################################## | 100% 0.20 s 
256 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 5.91 s 
Reading | ################################################## | 100% 2.22 s 
32768 bytes of flash verified
Writing 16 bytes to fuses, 16 bytes written, 16 verified
Writing 1 byte (0x00) to fuse0/wdtcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse1/bodcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse5/syscfg0, 1 byte written, 1 verified
Writing 1 byte (0x48) to fuse6/syscfg1, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse7/codesize, 1 byte written, 1 verified
Writing 1 byte (0x07) to fuse8/bootsize, 1 byte written, 1 verified
Writing 2 bytes (0x03 00) to fusea/pdicfg, 2 bytes written, 2 verified
Writing 4 bytes (0x5C C5 C5 5C) to lock, 4 bytes written, 4 verified
Writing 256 bytes to bootrow
Writing | ################################################## | 100% 0.15 s 
Reading | ################################################## | 100% 0.03 s 
256 bytes of bootrow verified
Writing 512 bytes to userrow/usersig
Writing | ################################################## | 100% 0.13 s 
Reading | ################################################## | 100% 0.03 s 
512 bytes of userrow/usersig verified

Processing -U ALL:v:bak-pickit_basic_updi-avr32sd28.hex:I
Verifying 33972 bytes of multiple memories against input file bak-pickit_basic_updi-avr32sd28.hex
Reading | ################################################## | 100% 0.20 s 
256 bytes of eeprom verified
Reading | ################################################## | 100% 2.21 s 
32768 bytes of flash verified
16 bytes of fuses verified
1 byte of fuse0/wdtcfg verified
1 byte of fuse1/bodcfg verified
1 byte of fuse5/syscfg0 verified
1 byte of fuse6/syscfg1 verified
1 byte of fuse7/codesize verified
1 byte of fuse8/bootsize verified
2 bytes of fusea/pdicfg verified
4 bytes of lock verified
Reading | ################################################## | 100% 0.06 s 
128 bytes of prodsig/sigrow verified
3 bytes of signature verified
4 bytes of tempsense verified
16 bytes of sernum verified
Reading | ################################################## | 100% 0.02 s 
256 bytes of bootrow verified
Reading | ################################################## | 100% 0.03 s 
512 bytes of userrow/usersig verified
32 bytes of sib verified

Avrdude done.  Thank you.

AVR32SD28 + PICkit4

$ ./test8 -r2 pickit4_updi avr32sd28
Reading 33972 bytes for multiple memories from input file bak-pickit4_updi-avr32sd28.hex
Writing 256 bytes to eeprom
Writing | ################################################## | 100% 7.65 s 
Reading | ################################################## | 100% 1.75 s 
256 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 11.70 s 
Reading | ################################################## | 100% 5.93 s 
32768 bytes of flash verified
Writing 16 bytes to fuses, 16 bytes written, 16 verified
Writing 1 byte (0x00) to fuse0/wdtcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse1/bodcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse5/syscfg0, 1 byte written, 1 verified
Writing 1 byte (0x48) to fuse6/syscfg1, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse7/codesize, 1 byte written, 1 verified
Writing 1 byte (0x0F) to fuse8/bootsize, 1 byte written, 1 verified
Writing 2 bytes (0x03 00) to fusea/pdicfg, 2 bytes written, 2 verified
Writing 4 bytes (0x5C C5 C5 5C) to lock, 4 bytes written, 4 verified
Writing 256 bytes to bootrow
Writing | ################################################## | 100% 0.24 s 
Reading | ################################################## | 100% 0.07 s 
Warning: bootrow verification mismatch
  device 0x00 != input 0x20 at addr 0x00c1 (error)
Error: bootrow verification mismatch
Writing 512 bytes to userrow/usersig
Writing | ################################################## | 100% 0.25 s 
Reading | ################################################## | 100% 0.18 s 
512 bytes of userrow/usersig verified

Avrdude done.  Thank you.
test8: avr32sd28 initialised
Hans-MacBook-Pro:tools hans$ avrdude -cpickit4_updi -pavr32sd28 -T 'read bootrow'
Reading | ################################################## | 100% 0.07 s 
0000  40 20 40 40 40 20 40 40  40 20 40 20 20 40 20 40  |@ @@@ @@@ @  @ @|
0010  40 40 40 40 20 20 20 40  40 20 20 20 20 40 40 20  |@@@@   @@    @@ |
0020  40 20 40 20 40 40 20 20  20 40 20 20 20 20 20 40  |@ @ @@   @     @|
0030  20 40 20 20 40 20 40 40  40 20 40 40 40 20 20 20  | @  @ @@@ @@@   |
0040  20 40 40 40 40 20 40 40  20 20 40 20 40 40 40 20  | @@@@ @@  @ @@@ |
0050  20 40 20 20 20 40 40 20  40 40 40 20 20 40 40 20  | @   @@ @@@  @@ |
0060  20 20 20 40 40 20 40 40  40 40 40 40 20 20 40 40  |   @@ @@@@@@  @@|
0070  20 40 40 20 20 40 40 40  20 20 20 20 40 40 40 20  | @@  @@@    @@@ |
0080  20 20 20 20 40 20 40 40  40 40 40 20 20 40 20 40  |    @ @@@@@  @ @|
0090  40 40 40 20 40 20 40 20  40 20 20 20 20 20 40 20  |@@@ @ @ @     @ |
00a0  40 40 20 20 40 40 20 20  20 40 20 40 20 40 40 20  |@@  @@   @ @ @@ |
00b0  20 40 20 40 20 20 40 40  40 20 20 20 20 40 40 20  | @ @  @@@    @@ |
00c0  40 00 40 40 00 00 40 20  20 20 00 00 00 40 00 20  |@.@@..@   ...@. |
00d0  00 40 20 00 00 20 00 00  20 20 00 00 00 00 00 20  |.@ .. ..  ..... |
00e0  20 00 00 00 00 40 00 00  20 20 00 20 00 20 00 40  | ....@..  . . .@|
00f0  20 20 40 00 00 20 00 20  20 62 6f 6f 74 72 6f 77  |  @.. .  bootrow|

Avrdude done.  Thank you.
Hans-MacBook-Pro:tools hans$ ./test8 -r3 pickit4_updi avr32sd28 
Reading 33972 bytes for multiple memories from input file bak-pickit4_updi-avr32sd28.hex
Writing 256 bytes to eeprom
Writing | ################################################## | 100% 7.65 s 
Reading | ################################################## | 100% 1.75 s 
256 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 11.70 s 
Reading | ################################################## | 100% 5.93 s 
32768 bytes of flash verified
Writing 16 bytes to fuses, 16 bytes written, 16 verified
Writing 1 byte (0x00) to fuse0/wdtcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse1/bodcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse5/syscfg0, 1 byte written, 1 verified
Writing 1 byte (0x48) to fuse6/syscfg1, 1 byte written, 1 verified
Writing 1 byte (0x30) to fuse7/codesize, 1 byte written, 1 verified
Writing 1 byte (0x07) to fuse8/bootsize, 1 byte written, 1 verified
Writing 2 bytes (0x03 00) to fusea/pdicfg, 2 bytes written, 2 verified
Writing 4 bytes (0x5C C5 C5 5C) to lock, 4 bytes written, 4 verified
Writing 256 bytes to bootrow
Writing | ################################################## | 100% 0.24 s 
Reading | ################################################## | 100% 0.07 s 
Warning: bootrow verification mismatch
  device 0x00 != input 0x20 at addr 0x0002 (error)
Error: bootrow verification mismatch
Writing 512 bytes to userrow/usersig
Writing | ################################################## | 100% 0.25 s 
Reading | ################################################## | 100% 0.18 s 
512 bytes of userrow/usersig verified

Avrdude done.  Thank you.
test8: avr32sd28 initialised
Hans-MacBook-Pro:tools hans$ avrdude -cpickit4_updi -pavr32sd28 -T 'read bootrow'
Reading | ################################################## | 100% 0.07 s 
0000  40 20 00 00 40 20 00 00  00 00 40 00 20 40 20 40  |@ ..@ ....@. @ @|
0010  00 40 40 40 00 20 00 40  40 00 20 00 00 00 00 20  |.@@@. .@@. .... |
0020  00 00 00 20 00 40 00 20  20 00 00 00 20 00 00 40  |... .@.  ... ..@|
0030  00 00 00 20 40 00 40 00  00 20 00 40 40 00 20 20  |... @.@.. .@@.  |
0040  00 40 00 00 00 00 40 40  20 00 00 20 40 40 00 20  |.@....@@ .. @@. |
0050  20 40 20 20 00 40 40 00  40 00 40 20 20 00 40 20  | @  .@@.@.@  .@ |
0060  00 20 00 00 40 00 40 40  00 00 00 40 00 00 40 40  |. ..@.@@...@..@@|
0070  20 40 00 00 20 40 40 40  00 00 00 00 40 00 00 20  | @.. @@@....@.. |
0080  00 20 20 20 40 00 00 00  40 00 40 00 20 40 20 40  |.   @...@.@. @ @|
0090  00 40 40 00 40 00 40 20  40 00 20 00 00 00 40 20  |.@@.@.@ @. ...@ |
00a0  00 40 00 00 00 00 00 20  00 40 20 40 20 40 00 00  |.@..... .@ @ @..|
00b0  20 00 00 40 00 20 00 40  40 20 20 00 00 40 00 20  | ..@. .@@  ..@. |
00c0  00 00 40 40 00 00 40 20  00 00 00 00 00 40 00 00  |..@@..@ .....@..|
00d0  00 40 20 00 00 00 00 00  20 20 00 00 00 00 00 20  |.@ .....  ..... |
00e0  00 00 00 00 00 00 00 00  20 00 00 20 00 00 00 40  |........ .. ...@|
00f0  20 20 40 00 00 20 00 20  20 62 6f 6f 74 72 6f 77  |  @.. .  bootrow|

Avrdude done.  Thank you.
Hans-MacBook-Pro:tools hans$ ./test8 pickit4_updi avr32sd28

Processing -U ALL:r:bak-pickit4_updi-avr32sd28.hex:I
Reading multiple memories ...
 - eeprom          | ################################################## | 100% 1.75 s 
 - flash           | ################################################## | 100% 5.93 s 
 - fuses           | ################################################## | 100% 0.11 s 
 - fuse0/wdtcfg    | ################################################## | 100% 0.01 s 
 - fuse1/bodcfg    | ################################################## | 100% 0.01 s 
 - fuse5/syscfg0   | ################################################## | 100% 0.01 s 
 - fuse6/syscfg1   | ################################################## | 100% 0.01 s 
 - fuse7/codesize  | ################################################## | 100% 0.01 s 
 - fuse8/bootsize  | ################################################## | 100% 0.01 s 
 - fusea/pdicfg    | ################################################## | 100% 0.01 s 
 - lock            | ################################################## | 100% 0.03 s 
 - prodsig/sigrow  | ################################################## | 100% 0.03 s 
 - signature       | ################################################## | 100% 0.01 s 
 - tempsense       | ################################################## | 100% 0.03 s 
 - sernum          | ################################################## | 100% 0.11 s 
 - bootrow         | ################################################## | 100% 0.07 s 
 - userrow/usersig | ################################################## | 100% 0.18 s 
 - sib             | ################################################## | 100% 0.00 s 
Writing 34003 bytes to output file bak-pickit4_updi-avr32sd28.hex

Processing -T fact reset
performing chip erase; discarded pending writes to flash, EEPROM and bootrow
Caching | ################################################## | 100% 1.75 s 
Caching | ################################################## | 100% 0.00 s 
Caching | ################################################## | 100% 0.07 s 
Caching | ################################################## | 100% 0.18 s 
Synching cache to device ... done
after the next reset the part will have F_CPU = 4.000 MHz

Processing -U ALL:w:bak-pickit4_updi-avr32sd28.hex:I
Reading 33972 bytes for multiple memories from input file bak-pickit4_updi-avr32sd28.hex
Writing 256 bytes to eeprom
Writing | ################################################## | 100% 7.65 s 
Reading | ################################################## | 100% 1.75 s 
256 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 10.29 s 
Reading | ################################################## | 100% 5.93 s 
32768 bytes of flash verified
Writing 16 bytes to fuses, 16 bytes written, 16 verified
Writing 1 byte (0x00) to fuse0/wdtcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse1/bodcfg, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse5/syscfg0, 1 byte written, 1 verified
Writing 1 byte (0x48) to fuse6/syscfg1, 1 byte written, 1 verified
Writing 1 byte (0x30) to fuse7/codesize, 1 byte written, 1 verified
Writing 1 byte (0x07) to fuse8/bootsize, 1 byte written, 1 verified
Writing 2 bytes (0x03 00) to fusea/pdicfg, 2 bytes written, 2 verified
Writing 4 bytes (0x5C C5 C5 5C) to lock, 4 bytes written, 4 verified
Writing 256 bytes to bootrow
Writing | ################################################## | 100% 0.15 s 
Reading | ################################################## | 100% 0.07 s 
256 bytes of bootrow verified
Writing 512 bytes to userrow/usersig
Writing | ################################################## | 100% 0.23 s 
Reading | ################################################## | 100% 0.18 s 
512 bytes of userrow/usersig verified

Processing -U ALL:v:bak-pickit4_updi-avr32sd28.hex:I
Verifying 33972 bytes of multiple memories against input file bak-pickit4_updi-avr32sd28.hex
Reading | ################################################## | 100% 1.75 s 
256 bytes of eeprom verified
Reading | ################################################## | 100% 5.93 s 
32768 bytes of flash verified
16 bytes of fuses verified
1 byte of fuse0/wdtcfg verified
1 byte of fuse1/bodcfg verified
1 byte of fuse5/syscfg0 verified
1 byte of fuse6/syscfg1 verified
1 byte of fuse7/codesize verified
1 byte of fuse8/bootsize verified
2 bytes of fusea/pdicfg verified
4 bytes of lock verified
Reading | ################################################## | 100% 0.03 s 
128 bytes of prodsig/sigrow verified
3 bytes of signature verified
4 bytes of tempsense verified
16 bytes of sernum verified
Reading | ################################################## | 100% 0.07 s 
256 bytes of bootrow verified
Reading | ################################################## | 100% 0.18 s 
512 bytes of userrow/usersig verified
32 bytes of sib verified

AVR32SD28 + SerialUPDI

$ avrdude -cserialupdi -P ch340 -pavr32sd28
Warning: unsupported NVM type: 6, please update software
Error: decode SIB_INFO failed
Error: initialization failed  (rc = -1)
 - double check the connections and try again
 - use -b to set lower baud rate, e.g. -b 57600
 - use -F to override this check

Avrdude done.  Thank you.

[MCUdude]

There is something wrong with the way Avrdude reads from (or writes to?) the AVR-SD bootrow memory. I get the same behaviour with the PICkit4 in AVR mode, and the SerialUPDI when forcing NVMv5 mode:

$ avrdude -cpickit_basic_updi -pavr32sd28 -t
avrdude> read bootrow
Reading | ################################################## | 100% 0.04 s 
0000  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0010  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0020  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0030  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0040  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0050  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0060  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0070  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0080  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0090  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00a0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00b0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00c0  6f 77 6e 20 66 6f 78 20  6a 75 6d 70 73 20 6f 76  |own fox jumps ov|
00d0  65 72 20 74 68 65 20 6c  61 7a 79 20 64 6f 67 20  |er the lazy dog |
00e0  00 54 68 65 20 71 75 69  63 6b 20 62 72 6f 77 6e  |.The quick brown|
00f0  20 66 6f 78 20 6a 75 6d  70 73 20 6f 76 65 72 20  | fox jumps over |
avrdude> write bootrow 0 -1 "Lorem ipsum dolor sit amet" ...
Caching | ################################################## | 100% 0.00 s 
avrdude> flush
Synching cache to device ... 
Writing | ################################################## | 100% 0.20 s 
avrdude> read bootrow
Reading | ################################################## | 100% 0.00 s 
0000  4c 6f 72 65 6d 20 69 70  73 75 6d 20 64 6f 6c 6f  |Lorem ipsum dolo|
0010  72 20 73 69 74 20 61 6d  65 74 00 4c 6f 72 65 6d  |r sit amet.Lorem|
0020  20 69 70 73 75 6d 20 64  6f 6c 6f 72 20 73 69 74  | ipsum dolor sit|
0030  20 61 6d 65 74 00 4c 6f  72 65 6d 20 69 70 73 75  | amet.Lorem ipsu|
0040  6d 20 64 6f 6c 6f 72 20  73 69 74 20 61 6d 65 74  |m dolor sit amet|
0050  00 4c 6f 72 65 6d 20 69  70 73 75 6d 20 64 6f 6c  |.Lorem ipsum dol|
0060  6f 72 20 73 69 74 20 61  6d 65 74 00 4c 6f 72 65  |or sit amet.Lore|
0070  6d 20 69 70 73 75 6d 20  64 6f 6c 6f 72 20 73 69  |m ipsum dolor si|
0080  74 20 61 6d 65 74 00 4c  6f 72 65 6d 20 69 70 73  |t amet.Lorem ips|
0090  75 6d 20 64 6f 6c 6f 72  20 73 69 74 20 61 6d 65  |um dolor sit ame|
00a0  74 00 4c 6f 72 65 6d 20  69 70 73 75 6d 20 64 6f  |t.Lorem ipsum do|
00b0  6c 6f 72 20 73 69 74 20  61 6d 65 74 00 4c 6f 72  |lor sit amet.Lor|
00c0  65 6d 20 69 70 73 75 6d  20 64 6f 6c 6f 72 20 73  |em ipsum dolor s|
00d0  69 74 20 61 6d 65 74 00  4c 6f 72 65 6d 20 69 70  |it amet.Lorem ip|
00e0  73 75 6d 20 64 6f 6c 6f  72 20 73 69 74 20 61 6d  |sum dolor sit am|
00f0  65 74 00 4c 6f 72 65 6d  20 69 70 73 75 6d 20 64  |et.Lorem ipsum d|
avrdude> q

Avrdude done.  Thank you.

$ avrdude -cpickit_basic_updi -pavr32sd28 -t
avrdude> read bootrow
Reading | ################################################## | 100% 0.03 s 
0000  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0010  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0020  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0030  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0040  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0050  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0060  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0070  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0080  74 20 61 6d 65 74 00 4c  6f 72 65 6d 20 69 70 73  |t amet.Lorem ips|
0090  75 6d 20 64 6f 6c 6f 72  20 73 69 74 20 61 6d 65  |um dolor sit ame|
00a0  74 00 4c 6f 72 65 6d 20  69 70 73 75 6d 20 64 6f  |t.Lorem ipsum do|
00b0  6c 6f 72 20 73 69 74 20  61 6d 65 74 00 4c 6f 72  |lor sit amet.Lor|
00c0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00d0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00e0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00f0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
avrdude> read bootrow
Reading | ################################################## | 100% 0.00 s 
0000  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0010  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0020  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0030  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0040  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0050  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0060  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0070  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0080  74 20 61 6d 65 74 00 4c  6f 72 65 6d 20 69 70 73  |t amet.Lorem ips|
0090  75 6d 20 64 6f 6c 6f 72  20 73 69 74 20 61 6d 65  |um dolor sit ame|
00a0  74 00 4c 6f 72 65 6d 20  69 70 73 75 6d 20 64 6f  |t.Lorem ipsum do|
00b0  6c 6f 72 20 73 69 74 20  61 6d 65 74 00 4c 6f 72  |lor sit amet.Lor|
00c0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00d0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00e0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00f0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
avrdude> read bootrow
Reading | ################################################## | 100% 0.00 s 
0000  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0010  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0020  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0030  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0040  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0050  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0060  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0070  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0080  74 20 61 6d 65 74 00 4c  6f 72 65 6d 20 69 70 73  |t amet.Lorem ips|
0090  75 6d 20 64 6f 6c 6f 72  20 73 69 74 20 61 6d 65  |um dolor sit ame|
00a0  74 00 4c 6f 72 65 6d 20  69 70 73 75 6d 20 64 6f  |t.Lorem ipsum do|
00b0  6c 6f 72 20 73 69 74 20  61 6d 65 74 00 4c 6f 72  |lor sit amet.Lor|
00c0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00d0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00e0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00f0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
avrdude> q

Avrdude done.  Thank you.

[stefanrueger]

Suggesting to release v8.1 on the weekend of 14/15 June. I will be away from any laptops in July and most of August, so that gives me two weeks to update a release should we have produced a lemon.

[MCUdude]

Sounds good to me!

I have way less time to spend on testing, compared to when we released 8.0. I think my primary focus should be to test the PICkit4 and SNAP in MPLAB mode since I stumbled over some strange behaviour in #1999. Of course, I'll also play with the urboot: option.

[mcuee]

@MCUdude

Nice, #1999 has been fixed by your PR #2000.

• SNAP in "MPLAB mode" has an incorrect USB VID #1999
• Fix incorrect MPLAB SNAP VID #2000

Quite some PRs related to PICKit 4/5 and SNAP have been merged, mainly done by @MX682X but you have done lots of tests.

• PK5 ISP, dW, TPI, JTAG and theoretical PDI support #1935 (you have done extensive tests here)
• Suggest compatible command when SNAP/PICKit4 in PIC mode is detected #1947
• Correct PICkit5 programmer type entries #1949
• updated scripts.xml and added PK Basic in non cimsis-dap mode #1959
• PICkit Basic Documentation #1966
• adding possibility to keep supplying power with PK4/5 after exit #1976
• Fix incorrect MPLAB SNAP VID #2000

PICKit 5 support is now done.

• Microchip PICKit 5 support #1368

[mcuee]

Today I did some simple tests for PICKit 4 in AVR and PIC mode.

I can confirm that Windows MSVC build does not work with the PIC mode.

• avrdude 8.1 release discussion #1956 (reply in thread)

[mcuee]

@stefanrueger

Just wondering if we need to improve the documentation for test8 script or not.

I suggest to change to something like the following.

Syntax: test8 [<opts>] <programmer> <part>
Function: test AVRDUDE (8.0 release or later version) with -c programmer -p part;
  also leaves a file bak-<programmer>-<part.hex for inspection
Options:
  -i <n>     use avrdude -c programmer -xinit=<n> to initialise part and exit
  -r <n>     use avrdude -c programmer -xrandom=<n> to initialise part and exit
  -e <exe>   path of the avrdude executable
  -a <opt>   pass avrdude option <opt> through to avrdude test commands
  -m <mlist> use memory list <mlist> instead of ALL

My test run log.

MINGW64 /c/work/avr/avrdude/tools
$ ./test8
test8: missing arguments
Syntax: test8 [<opts>] <programmer> <part>
Function: test AVRDUDE for v 8.0 release with -c programmer -p part;
  also leaves a file bak-<programmer>-<part.hex for inspection
Options:
  -i <n>     use avrdude -c dryrun -xinit=<n> to initialise part and exit
  -r <n>     use avrdude -c dryrun -xrandom=<n> to initialise part and exit
  -e <exe>   path of the avrdude executable
  -a <opt>   pass avrdude option <opt> through to avrdude test commands
  -m <mlist> use memory list <mlist> instead of ALL

Examples:
  - Prepare a part with non-trivial random content
    $ test8 -r1 usbasp t13a

  - Then test ...
    $ test8 avrisp2 t13a
    $ test8 snap_isp t13a

$ ./test8 -e ../../avrdude_bin/avrdude.exe -r1 usbasp m328p
Reading 33824 bytes for multiple memories from input file bak-usbasp-m328p.hex
Writing 1024 bytes to eeprom
Writing | ################################################## | 100% 10.91 s
Reading | ################################################## | 100% 0.62 s
1024 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 18.90 s
Reading | ################################################## | 100% 9.94 s
32768 bytes of flash verified
Writing 1 byte (0x62) to lfuse, 1 byte written, 1 verified
Writing 1 byte (0xDB) to hfuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to efuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to lock, 1 byte written, 1 verified

Avrdude done.  Thank you.
test8: m328p initialised

$ ./test8 -e ../../avrdude_bin/avrdude.exe usbasp m328p

Processing -U ALL:r:bak-usbasp-m328p.hex:I
Reading multiple memories ...
 - eeprom         | ################################################## | 100% 0.67 s
 - flash          | ################################################## | 100% 10.02 s
 - lfuse          | ################################################## | 100% 0.00 s
 - hfuse          | ################################################## | 100% 0.00 s
 - efuse          | ################################################## | 100% 0.00 s
 - lock           | ################################################## | 100% 0.00 s
 - prodsig/sigrow | ################################################## | 100% 0.05 s
 - sernum         | ################################################## | 100% 0.01 s
 - signature      | ################################################## | 100% 0.00 s
 - calibration    | ################################################## | 100% 0.00 s
Writing 33834 bytes to output file bak-usbasp-m328p.hex

Processing -T fact reset
performing chip erase; discarded pending writes to flash and EEPROM
Caching | ################################################## | 100% 0.63 s
Caching | ################################################## | 100% 0.02 s
after the next reset the part will have F_CPU = 1.000 MHz

Processing -U ALL:w:bak-usbasp-m328p.hex:I
Reading 33824 bytes for multiple memories from input file bak-usbasp-m328p.hex
Writing 1024 bytes to eeprom
Writing | ################################################## | 100% 10.89 s
Reading | ################################################## | 100% 0.63 s
1024 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 18.25 s
Reading | ################################################## | 100% 10.07 s
32768 bytes of flash verified
Writing 1 byte (0x62) to lfuse, 1 byte written, 1 verified
Writing 1 byte (0xDB) to hfuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to efuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to lock, 1 byte written, 1 verified

Processing -U ALL:v:bak-usbasp-m328p.hex:I
Verifying 33824 bytes of multiple memories against input file bak-usbasp-m328p.hex
Reading | ################################################## | 100% 0.63 s
1024 bytes of eeprom verified
Reading | ################################################## | 100% 10.07 s
32768 bytes of flash verified
1 byte of lfuse verified
1 byte of hfuse verified
1 byte of efuse verified
1 byte of lock verified
24 bytes of prodsig/sigrow verified
10 bytes of sernum verified
3 bytes of signature verified
1 byte of calibration verified

Avrdude done.  Thank you.

[mcuee]

@stefanrueger @MCUdude and @dl8dtl

I managed to get an AVR JTAGICE MKII programmer only clone (no debug function) from a Chinese vendor in Taobao. And it seems to work fine.

[C:\work\avr\avrdude_bin] $ .\avrdude.exe -c jtag2isp -p m328p -U m328p_readfull.hex -v
Avrdude version 8.0-20250529 (98784e02)
Copyright see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

System wide configuration file is C:\work\avr\avrdude_bin\avrdude.conf

Using port            : usb
Using programmer      : jtag2isp
Programmer baud rate  : 115200
Usbdev_open(): found JTAGICE MKII, serno: 098FE6230124
AVR part              : ATmega328P
Programming modes     : SPM, ISP, HVPP, debugWIRE
Programmer type       : JTAGMKII_ISP
Description           : Atmel JTAG ICE mkII in ISP mode
Vtarget               : 5.0 V
SCK period            : 4.0 us

AVR device initialized and ready to accept instructions
Device signature = 1E 95 0F (ATmega328P, ATA6614Q, LGT8F328P)
Auto-erasing chip as flash memory needs programming (-U flash:w:...)
specify the -D option to disable this feature
Erased chip
Reading 32768 bytes for flash from input file m328p_readfull.hex
in 1 section [0, 0x7fff]: 256 pages and 0 pad bytes
Writing 32768 bytes to flash
Writing | ################################################## | 100% 2.45 s
Reading | ################################################## | 100% 5.97 s
32768 bytes of flash verified

Avrdude done.  Thank you.

MINGW64 /c/work/avr/avrdude/tools
$ ./test8 -e ../../avrdude_bin/avrdude.exe -r1 jtag2isp m328p
Reading 33824 bytes for multiple memories from input file bak-jtag2isp-m328p.hex
Writing 1024 bytes to eeprom
Writing | ################################################## | 100% 1.72 s
Reading | ################################################## | 100% 0.79 s
1024 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 2.48 s
Reading | ################################################## | 100% 5.96 s
32768 bytes of flash verified
Writing 1 byte (0x62) to lfuse, 1 byte written, 1 verified
Writing 1 byte (0xDB) to hfuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to efuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to lock, 1 byte written, 1 verified

Avrdude done.  Thank you.
test8: m328p initialised

$ ./test8 -e ../../avrdude_bin/avrdude.exe jtag2isp m328p

Processing -U ALL:r:bak-jtag2isp-m328p.hex:I
Reading multiple memories ...
 - eeprom         | ################################################## | 100% 0.73 s
 - flash          | ################################################## | 100% 5.99 s
 - lfuse          | ################################################## | 100% 0.00 s
 - hfuse          | ################################################## | 100% 0.00 s
 - efuse          | ################################################## | 100% 0.00 s
 - lock           | ################################################## | 100% 0.00 s
 - prodsig/sigrow | ################################################## | 100% 0.04 s
 - sernum         | ################################################## | 100% 0.02 s
 - signature      | ################################################## | 100% 0.01 s
 - calibration    | ################################################## | 100% 0.00 s
Writing 33834 bytes to output file bak-jtag2isp-m328p.hex

Processing -T fact reset
performing chip erase; discarded pending writes to flash and EEPROM
Caching | ################################################## | 100% 0.78 s
Caching | ################################################## | 100% 0.03 s
after the next reset the part will have F_CPU = 1.000 MHz

Processing -U ALL:w:bak-jtag2isp-m328p.hex:I
Reading 33824 bytes for multiple memories from input file bak-jtag2isp-m328p.hex
Writing 1024 bytes to eeprom
Writing | ################################################## | 100% 1.70 s
Reading | ################################################## | 100% 0.77 s
1024 bytes of eeprom verified
Writing 32768 bytes to flash
Writing | ################################################## | 100% 2.47 s
Reading | ################################################## | 100% 6.00 s
32768 bytes of flash verified
Writing 1 byte (0x62) to lfuse, 1 byte written, 1 verified
Writing 1 byte (0xDB) to hfuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to efuse, 1 byte written, 1 verified
Writing 1 byte (0xFF) to lock, 1 byte written, 1 verified

Processing -U ALL:v:bak-jtag2isp-m328p.hex:I
Verifying 33824 bytes of multiple memories against input file bak-jtag2isp-m328p.hex
Reading | ################################################## | 100% 0.77 s
1024 bytes of eeprom verified
Reading | ################################################## | 100% 5.99 s
32768 bytes of flash verified
1 byte of lfuse verified
1 byte of hfuse verified
1 byte of efuse verified
1 byte of lock verified
24 bytes of prodsig/sigrow verified
10 bytes of sernum verified
3 bytes of signature verified
1 byte of calibration verified

Avrdude done.  Thank you.

[mcuee]

One more test.

MINGW64 /c/work/avr/avrdude/tools
$ ./test-avrdude -e ../../avrdude_bin/avrdude.exe -t . -d 0 -v -p "-c jtag2isp -p m328p"
Testing ../../avrdude_bin/avrdude.exe 8.0-20250529 (98784e02)
Prepare "-c jtag2isp -p m328p" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.629 s: fuse access: clear, set and read eesave fuse bit
✅   0.589 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   0.669 s: chip erase
✅   2.990 s: flash -U write/verify holes_rjmp_loops_32768B.hex
✅   1.647 s: flash -T write/verify holes_rjmp_loops_32768B.hex
✅   0.617 s: eeprom check whether programmer can flip 0s to 1s
✅   1.250 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
✅   1.963 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex
✅   1.700 s: chip erase and spot check flash is actually erased
✅   0.806 s: spot check eeprom is erased, too

[mcuee]

@stefanrueger , @MCUdude and @dl8dtl

As far as I am concerned, I think the 8.1RC branch is good enough and we can release avrdude 8.1.

[MX682X]

I've also run the tests on UPDI/t1614 JTAG/32u4 and m2560 ISP/32u4 and m2560 TPI/t9 with the Pickit Basic.
On Windows 10 with mingw64 build

Would apreciate if someone would test the attiny13, I think? The weird one that needs -B32. And the PDI interface

[MX682X]

@MCUdude Thanks for the tests. I don't know if I will be able to fix it before the planned release though as it gets a bit busy on my end aswell.
But it was working with PICkit 4 before, right? Could you verify it still works with that as well? This could narrow down where to look for.

[MCUdude]

@MX682X I'm actually seeing the exact same results for both ATtiny14 and ATxmega256A3BU when using the PICkit4 in MPLAB mode. However, the difference now compared to when ISP/PDI/JTAG/TPI support was added was that I only tested without initializing the parts with random data (-r1..N). It only fails when I do this, but it passes the "regular" test8 test.

[MCUdude]

Regarding the ATtiny15, I'm able to read and write to fuse memory just fine, but it reads incurrectly when trying to verify:

$ avrdude -cpickit4_mplab_isp -pattiny15 -B32 -Ufuse:w:0x1c:m
Reading 1 byte for fuse from input file 0x1c
Writing 1 byte (0x1C) to fuse, 1 byte written
Warning: fuse verification mismatch
  device 0xff != input 0x1c at addr 0x0000 (error)
Error: fuse verification mismatch

Avrdude done.  Thank you.


$ avrdude -cpickit4_mplab_isp -pattiny15 -B32 -Ufuse:r:-:h
Reading fuse memory ...
Writing 1 byte to output file <stdout>
0x1c

Avrdude done.  Thank you.

[MCUdude]

Stange. I rolled all the way back to when PR #1935 were merged, and I'm still experiencing the same issue on the ATtiny15. Is this a PICkit firmware issue?

$ git branch
* (HEAD detached at f2f2ac40)
  8.1RC
  main
  pack-update
  pack-update2
  pickit5_id
  serialupdi-sd-fix
  urboot-builtin

$ avrdude -cpickit5_isp -pattiny15 -B32 -Ufuse:r:-:h
Reading fuse memory ...
Writing 1 byte to output file <stdout>
0x1c

Avrdude done.  Thank you.

$ avrdude -cpickit5_isp -pattiny15 -B32 -Ufuse:w:0x1c:m
Reading 1 byte for fuse from input file 0x1c
Writing 1 byte (0x1C) to fuse, 1 byte written
Warning: fuse verification mismatch
  device 0xff != input 0x1c at addr 0x0000 (error)
Error: fuse verification mismatch

Avrdude done.  Thank you.

[stefanrueger]

what's the deal with the line 205 error?

Is the locale variable LC_NUMERIC set so that commas are decimal separators? Maybe try

$ LC_ALL=C ./test-avrdude ...

[stefanrueger]

Are the mismatched fuse bits active? This could be an error in the bitmask value of the fuse in avrdude.conf.

[MCUdude]

Not sure. But when I read and write the fuses using a USBasp programmer, The factory default fuse values is read as 0x50, while the PICkit5/PICkit Basic read is as 0x5c.

$ avrdude -cusbasp -pattiny15 -B32 -Ufuse:r:-:h
Set SCK frequency to 16000 Hz
Reading fuse memory ...
Writing 1 byte to output file <stdout>
0x50

Avrdude done.  Thank you.


$ avrdude -cusbasp -pattiny15 -B32 -Ufuse:w:0x50:m
Set SCK frequency to 16000 Hz
Reading 1 byte for fuse from input file 0x50
Writing 1 byte (0x50) to fuse, 1 byte written, 1 verified

Avrdude done.  Thank you.


$ avrdude -cusbasp -pattiny15 -B32 -Ufuse:w:0x50:m
Set SCK frequency to 16000 Hz
Reading 1 byte for fuse from input file 0x50
Writing 1 byte (0x50) to fuse, 1 byte written, 1 verified

Avrdude done.  Thank you.


$ avrdude -cusbasp -pattiny15 -B32 -Ufuse:w:0x10:m
Set SCK frequency to 16000 Hz
Reading 1 byte for fuse from input file 0x10
Writing 1 byte (0x10) to fuse, 1 byte written, 1 verified

Avrdude done.  Thank you.

[MX682X]

I just checked the docs - according to those, the default value for the fuses should be
0101 00xx (xx is lock, which defaults to 11) so it's 0x50 or 0x53, while the avrdude conf states
initval = 0x5c;
bitmask = 0xf3;
for the fuse, aka 0101 11xx, so avrdude might be changing the start-up time.
Besides the fact that the bitmask is probably wrong aswell - or I might misunderstand how it works, as this values correspond to the programming commands masks

I've rechecked the code - as I'm not masking the output of the pickit the 0x5C is the same as 0x50 - I think I assumed at the point of writing that avrdude would do the masking internally

[MX682X]

I might have an idea where the problem comes from, I've uploaded the commit to my repo. I've changed the fuse reads for isp.

[mcuee]

Just another test under one of the so-called "Immutable Linux Distributions", Fedora Silverblue 42 in this case.

Need to build under containers (toolbox under Fedora based, distrobox under others), in this case, Fedora 42 toolbox. But then you cannnot run it under the container.

⬢ [mcuee@toolbx avrdude_81_rc]$ sudo avrdude -c stk600 -p m328p -v
Avrdude version 8.0-20250610 (9d7a893d)
Copyright see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

System wide configuration file is /usr/local/etc/avrdude.conf
User configuration file /root/.avrduderc does not exist

Using port            : usb
Using programmer      : stk600
Warning: cannot open device: Permission denied
Error: did not find any USB device usb (03eb:2106)
Error: unable to open port usb for programmer stk600

Avrdude done.  Thank you.

You need to run in the underline immunitable distro.

mcuee@silverblue:~/build/avrdude_81_rc/tools$ sudo ./test-avrdude -e ../build_linux/src/avrdude -v -p "-c stk600 -p m328p"
Testing ../build_linux/src/avrdude 8.0-20250610 (9d7a893d)
Prepare "-c stk600 -p m328p" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.380 s: fuse access: clear, set and read eesave fuse bit
✅   0.374 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   0.556 s: chip erase
✅   2.681 s: flash -U write/verify holes_rjmp_loops_32768B.hex
✅   1.311 s: flash -T write/verify holes_rjmp_loops_32768B.hex
✅   0.399 s: eeprom check whether programmer can flip 0s to 1s
✅   0.772 s: eeprom -U write/verify holes_pack_my_box_1024B.hex
✅   1.240 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_1024B.hex
✅   1.501 s: chip erase and spot check flash is actually erased
✅   0.449 s: spot check eeprom is erased, too

[stefanrueger]

@MX682X Great News! Perhaps best to create PR against the 8.1RC branch of AVRDUDE. Hans can merge that into the branch, and all tests are done against the release candidate.

[stefanrueger]

avrdude conf states initval = 0x5c; bitmask = 0xf3;

Hmm, initval has bits set outside the bitmask. Probably an error in avrdude.conf, which should follow the data sheet

[stefanrueger]

Hmm, initval has bits set outside the bitmask

I just checked: That's very common, indeed. Mostly with initval being 0xff (but by all means not always) and bitmask of a fuse covering a subset of bits. So, not an error per se.

[stefanrueger]

Nearly there!

• @MCUdude did you want to update/provide a few more examples for the tex documentation? You had ideas earlier
• @MX682X I noticed that v8.1 seems to be missing snap_mplab_dw: is this intended?
• For some reason v8.0 didn't have snap_dw based on jtagice3 but v8.1 has snap_dw (good); has anyone tested this?
• Testing found a problem with pickit_basic_pdi and atxmega256a3bu: how about documenting that as issue for the time being; I don't think it's necessary to address this before the release

So, for now it's only #2017 that blocks the release (as far as I am concerned).

[mcuee]

@stefanrueger

It seems to me #2017 is not a real issue but I have no idea why it happens.

No issues on my fork -- just sync my fork with avrdude git main and there is no issue at all.
https://github.com/mcuee/avrdude/actions/runs/15721502141/job/44303172484

[mcuee]

@stefanrueger

You can try on your fork to see if you see issue #2017 or not.
https://github.com/stefanrueger/avrdude

[MCUdude]

@stefanrueger I can confirm that snap_dw works. However, we@re missing pickit4_dw. I'll create a PR for this ASAP

$ avrdude -csnap_dw -pattiny13 -v
Avrdude version 8.0-20250614 (696decb5)
Copyright see https://github.com/avrdudes/avrdude/blob/main/AUTHORS

System wide configuration file is /usr/local/etc/avrdude.conf
User configuration file /Users/hans/.avrduderc does not exist

Using port            : usb
Using programmer      : snap_dw
AVR part              : ATtiny13
Programming modes     : SPM, ISP, HVSP, debugWIRE
Programmer type       : JTAGICE3_DW
Description           : MPLAB(R) SNAP in debugWire mode
ICE HW version        : 1
ICE FW version        : 1.11 (rel. 20)
Serial number         : BUR183076609
Vtarget               : 5.19 V

JTAG ID returned: 0x07 0x90 0x00 0x00

AVR device initialized and ready to accept instructions
Device signature = 1E 90 07 (ATtiny13, ATtiny13A)

Avrdude done.  Thank you.

[MCUdude]

@MCUdude did you want to update/provide a few more examples for the tex documentation? You had ideas earlier

We left for vacation two days ago. I brought my computer, but my time is limited to late evenings.

One idea was to show a terminal regfile example where one can control actual GPIO operations through Avrdude.
All existing examples in the docs use "short options" line -c and -p. Maybe we should add/update a few examples where we also show the use of "long options"?

[stefanrueger]

Good ideas.

also show the use of "long options"?

At the very least for --version as this doesn't have a short option

[stefanrueger]

TLDR; Looks like moving the release back to 23 June.

There are still some tweaks that are currently explored (PR #2020 and PR #2021). The main niggle is that we have had to refrain from building the python GUI in the Windows CI #2017. It's not great to publish a release without @dl8dtl's gui being built. As I am not aware of us having changed something related to that, I am hoping it's a github bug or similar that will iron itself out without us.

So, ... I suggest waiting a few days before tagging the 8.1 release. I've reserved Mon 23 June for that. @mcuee, could you check on 22 June, please, whether PR #2018 can be undone before the release? This gives the team a bit of time to check/update documentation.

[mcuee]

So, ... I suggest waiting a few days before tagging the 8.1 release. I've reserved Mon 23 June for that. @mcuee, could you check on 22 June, please, whether PR #2018 can be undone before the release? This gives the team a bit of time to check/update documentation.

@stefanrueger

Good idea to postpone the release by a few days. I will check out whether the Python Binding building issue can go away by itself.

[mcuee]

@stefanrueger

BTW, I see your fork repo also have no issues with github CI. So the issue is very strange, somehow only specific to the official avrdude repo. It is as if github allocates a "bad" runner for official avrdude repo's CI build.

Your repo's CI is fine.
https://github.com/stefanrueger/avrdude/actions/runs/15734672121

My repo's CI is also fine.
https://github.com/mcuee/avrdude/actions/runs/15747585532

[mcuee]

@stefanrueger

Because of the above, I tend to think we can still proceed with avrdude 8.1 release without the CI build for Windows Python binding.

In any case, we have not sorted out Issue #1946 and therefore the CI build does not archive the Windows Python Binding artifacts anyway.

[mcuee]

@stefanrueger

#2017 has been fixed -- it is not really an avrdude issue after all, but a Windows CPython bug.

[MCUdude]

@MX682X, now that we have figured out the ATtiny15 issue, the last known issue we should look more into is the flash write errors we see on the Xmega series. I've only tested the ATxmega256A3BU with the PICkit4 in MPLAB mode recently, but this was the error I got when I was "priming" the device with random data before test:

$ ./test8 -r8 pickit4_mplab_pdi atxmega256a3bu
Target successfully erased
Reading 275006 bytes for multiple memories from input file bak-pickit4_mplab_pdi-atxmega256a3bu.hex
Writing 4096 bytes to eeprom
Writing | ################################################## | 100% 1.03 s 
Reading | ################################################## | 100% 0.51 s 
4096 bytes of eeprom verified
Writing 270336 bytes to flash
Writing | ################################################## | 100% 13.06 s 
Reading | ################################################## | 100% 7.46 s 
Warning: flash verification mismatch
  device 0xff != input 0xc0 at addr 0x40001 (error)
Error: flash verification mismatch
Writing 262144 bytes to application
Writing | ################################################## | 100% 12.80 s 
Reading | ################################################## | 100% 7.24 s 
Warning: application verification mismatch
  device 0x00 != input 0xff at addr 0x0000 (error)
Error: application verification mismatch
Writing 8192 bytes to apptable
Writing | ################################################## | 100% 0.40 s 
Reading | ################################################## | 100% 0.22 s 
8192 bytes of apptable verified
Writing 8192 bytes to boot
Writing | ################################################## | 100% 0.26 s 
Reading | ################################################## | 100% 0.22 s 
Warning: boot verification mismatch
  device 0xff != input 0xc0 at addr 0x0001 (error)
Error: boot verification mismatch
Writing 6 bytes to fuses, 6 bytes written, 6 verified
Writing 1 byte (0xFF) to fuse0, 1 byte written, 1 verified
Writing 1 byte (0x00) to fuse1, 1 byte written, 1 verified
Writing 1 byte (0xBF) to fuse2, 1 byte written, 1 verified
Writing 1 byte (0xFE) to fuse4, 1 byte written, 1 verified
Writing 1 byte (0xFF) to fuse5, 1 byte written, 1 verified
Writing 1 byte (0xFF) to lock, 1 byte written, 1 verified
Writing 512 bytes to usersig
Writing | ################################################## | 100% 0.03 s 
Reading | ################################################## | 100% 0.02 s 
512 bytes of usersig verified

Avrdude done.  Thank you.

I've created a dummy file that writes "The quick brown fox jumps over the lazy dog" To the entire 270336 byte flash memory, and I've attached both the original file and the flash dump. As you can see, the flash memory is empry after the 0x40000 mark.

$ avrdude -cpickit4_mplab_pdi -patxmega256a3bu -Uflash:w:the_quick_brown_fox_256kib.hex:i
Target successfully erased
Reading 270336 bytes for flash from input file the_quick_brown_fox_256kib.hex
Writing 270336 bytes to flash
Writing | ################################################## | 100% 13.06 s 
Reading | ################################################## | 100% 7.47 s 
Warning: flash verification mismatch
  device 0xff != input 0x61 at addr 0x40000 (error)
Error: flash verification mismatch

Avrdude done.  Thank you.

atxmega256a3bu_flash_dump.zip

The flash (256+4kiB) memory contains both the application (256kiB) and the boot (4kiB) section of the memory. I am able to write to the application part of the flash, but not the boot section.

$ avrdude -cpickit4_mplab_pdi -patxmega256a3bu -Uboot:w:test_files/lorem_ipsum_4096B.srec 
Target successfully erased
Reading 4096 bytes for boot from input file lorem_ipsum_4096B.srec
Writing 4096 bytes to boot
Writing | ################################################## | 100% 0.13 s 
Reading | ################################################## | 100% 0.11 s 
Warning: boot verification mismatch
  device 0xff != input 0x4c at addr 0x0000 (error)
Error: boot verification mismatch

Avrdude done.  Thank you.


$ avrdude -cpickit4_mplab_pdi -patxmega256a3bu -T 'read boot'
Reading | ################################################## | 100% 0.03 s 
0000  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0010  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0020  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0030  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0040  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0050  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0060  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0070  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0080  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
0090  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00a0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00b0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00c0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00d0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00e0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|
00f0  ff ff ff ff ff ff ff ff  ff ff ff ff ff ff ff ff  |................|

Avrdude done.  Thank you.

[MCUdude]

@MX682X BTW, is the PICkit4/5's Programmer-to-go functionality worth looking into? It sure can be useful to have a programmer who can flash a target by the click of a button, but if this is a proprietary mess, I'm not so sure.

Avrdude supports "multi-memory" hex files. Is this something that the programmer-to-go could support?

[MX682X]

The pseudo code for the programming script looks more or less like this:

uint32_t addr;
uint32_t length;
uint32_t new_len = length /= 0x200;

while (new_length --) {
uint32_t temp1 = 0x10001CA;    // "CMD" NVM Register
uint32_t tempA = *temp1;
uint32_t temp2 = 0x10001C4;  // "DATA0" NVM Register
uint32_t tempB = *temp2;

uint32_t temp3 = 0x10001CA;
uint32_t temp4 = 0x23;        // PDI Cmd: Load flash page buffer
*temp3 = temp4;

uint32_t temp5 = 0x200;
set_repeat(temp5);
// unknown instruction with paramter temp5
uint32_t temp6 = 0x10001CA;
uint32_t temp7 = 0x25;       // PDI Cmd: Erase and write application section page (0x2C for bootmem, which is not being erased beforehand. A bug in the scripts?)
*temp6 = temp7;
// Set pointer for indirect addressing to value in addr;
uint8_t temp8 = 0xFF;
uint32_t temp9 = 0x01;
// unknown instruction with paramter temp9
uint32_t temp9 = 0x10001CF
uint32_t temp;
do {
  temp = *temp9;
while((temp & 0x80) != 0x00)
temp1 = 0x10001CA;
*temp1 = tempA;
temp2 = 0x10001C4;
*temp2 = tempB;

addr += 0x200

}

My conclusion is, that the scripts don't allow for the possibility to write the whole flash in one go. it has to be separated in application and boot. The scripts currently only seem to support application section (Progmem) and boot loader section. Weirdly enough, the PDI does specify the "Erase and write flash page" command. Further more, the WriteBootmem script is using only the "Write boot loader section page" but not the "Erase and write boot loader section page" meaning that without a previous chip erase the programming of the boot loader would fail.

My conclusion is that for this fix I have to hack together a custom flash write script once again incorporating the necessary fixes.

And for the PTG functionality - I believe I could figure it out as well, but I do try to keep the supported functionality lower on purpose. Basically I try to avoid making a replacement for MPLAB (there is still the debugging, but still)

[MCUdude]

My conclusion is that for this fix, I have to hack together a custom flash write script once again, incorporating the necessary fixes.

So this is a limitation that MPLAB also has? I mean not being able to write to the boot section of the flash memory? It is not critical to have this ready for the 8.1 release, but I'm ready to test if you think you have a solution.

And for the PTG functionality - I believe I could figure it out as well, but I do try to keep the supported functionality lower on purpose.

Got it. And it would suck if we for some reason got into legal trouble. However, it could be useful to have a dedicated CLI tool that could read/write/manage PTG functionality. And if this would be a no-go for Microchip, it would only affect that particular project. And support for other architectures rather than just AVRs would also make more sense in a dedicated CLI application.

[MX682X]

So this is a limitation that MPLAB also has? I mean not being able to write to the boot section of the flash memory? It is not critical to have this ready for the 8.1 release, but I'm ready to test if you think you have a solution.

avrdude exposes two ways to address a flash address of this devices - through "flash" and "application"/"boot".
MPLAB is only using "application" and "boot". As MPLAB knows where what goes it can write those areas sequentially.
And as MPLAB is erasing the device before an upload, the boot memory is always empty, so there is no problem that it does only "Write boot loader section page"

But I won't tackle this issue now as I need to figure out the "unknown" instruction and this will probably take some time.

[MCUdude]

I'll create a dedicated issue where we can continue the discussion

[MCUdude]

@stefanrueger, a few new packs are available at https://packs.microchip.com, most noticeably packs for "classic" ATmegas and the AVR-SD series. Do these packages contain any changes we care about?

[stefanrueger]

The flash (256+4kiB) memory contains both the application (256kiB) and the boot (4kiB) section of the memory.

And there is apptable, which is located at the top end of, and within, application.

I am able to write to the application part of the flash, but not the boot section.

@MCUdude Have you checked where to the contents for the boot section is written instead, if it's not boot?

[MCUdude]

@MCUdude Have you checked where to the contents for the boot section is written instead, if it's not boot?

The content isn't written to a different location. A factory fresh chip strays blank even if I try to write to the boot memory using -c pickit4_mplab_pdi or -c pickit4_mplab_jtag. I've checked all memories.
Let's see what @MX682X comes up with. It seems like he got his head around the problem.

But shouldn't test-avrdude have been able to catch this? Right now, only test8 catches this.

[stefanrueger]

But shouldn't test-avrdude have been able to catch this? Right now, only test8 catches this.

No. They do different things. test-avrdude deploys a scripted set of checks on flash, eeprom, fuses and usersig with difficult files, at least for flash and eeprom, in particular holes. test8 was designed to test backup and restore of all memories, but won't test particularly difficult files. Our own testing tools have evolved considerably over the last few years and, for now, I am quite happy with the state of affairs. The real problem is that programmers are not perfect. We won't catch every programmer bug with our testing, no matter how hard we try, as the ingenuity of bugs, sadly, outstrips our own 😞

[MCUdude]

Got it. But even though we can't catch all programmer-related bugs, I'd argue that we're pretty close! The extensive testing we did before the 8.0 release really paid off, and we caught lots of edge case bugs that had been present for years.

[mcuee]

@umbynos

Just a headsup for the upcoming avrdude 8.1 release. It would be good to have the avrdude-packing release of avrdude 8.1 release, especially for macOS build with the digital signature.

I have created a new enhance request for arduino crossbuild project.

• Cross build for Apple macOS arm64 arduino/crossbuild#17

I have also updated two existing issues for arduino crossbuild project.

• Add libserialport for future avrdude release arduino/crossbuild#13
• Update libusb, libusb-compact, libftdi and hidapi library version arduino/crossbuild#14

I understand the issues may take time to be fixed.

[umbynos]

Hey @mcuee, unfortunately I'm no longer part of Arduino 😭 I cannot help you with this anymore, I'm sorry.

[mcuee]

Hey @mcuee, unfortunately I'm no longer part of Arduino 😭 I cannot help you with this anymore, I'm sorry.

I see. No problem. Hopefully someone else will take over.

[mcuee]

@chenrui333

Just a headsup for the upcoming avrdude 8.1 release.

As of now, homebrew avrdude formula is broken for avrdude git main.

mcuee@Macmini src/.libs  - (main)> brew install avrdude --head
==> Fetching avrdude
==> Downloading https://raw.githubusercontent.com/Homebrew/homebrew-core/7262e89eee5c01139208706c08b31b7e53535de4/Formula/a/avrdude.
############################################################################################################################# 100.0%
==> Cloning https://github.com/avrdudes/avrdude.git
Updating /Users/mcuee/Library/Caches/Homebrew/avrdude--git
From https://github.com/avrdudes/avrdude
   4e473c04..504267d9  main       -> origin/main
==> Checking out branch main
Already on 'main'
Your branch is behind 'origin/main' by 58 commits, and can be fast-forwarded.
  (use "git pull" to update your local branch)
HEAD is now at 504267d9 Update NEWS
==> cmake -S . -B build/shared -DCMAKE_INSTALL_SYSCONFDIR=/opt/homebrew/etc -DBUILD_SHARED_LIBS=ON -DCMAKE_INSTALL_RPATH=@loader_pat
==> cmake --build build/shared
==> cmake --install build/shared
Last 15 lines from /Users/mcuee/Library/Logs/Homebrew/avrdude/03.cmake:
-- Installing: /opt/homebrew/Cellar/avrdude/HEAD-504267d/lib/libavrdude.dylib
-- Installing: /opt/homebrew/Cellar/avrdude/HEAD-504267d/include/libavrdude.h
-- Installing: /opt/homebrew/Cellar/avrdude/HEAD-504267d/include/libavrdude-avrintel.h
-- Installing: /opt/homebrew/etc/avrdude.conf
-- Installing: /opt/homebrew/Cellar/avrdude/HEAD-504267d/share/man/man1/avrdude.1
-- Installing: /opt/homebrew/lib/python3.13/site-packages/_swig_avrdude.so
CMake Error at build/shared/src/cmake_install.cmake:119 (file):
  file INSTALL cannot copy file
  "/private/tmp/avrdude-20250621-21407-uqdn87/build/shared/src/_swig_avrdude.so"
  to "/opt/homebrew/lib/python3.13/site-packages/_swig_avrdude.so": Operation
  not permitted.
Call Stack (most recent call first):
  build/shared/cmake_install.cmake:42 (include)



READ THIS: https://docs.brew.sh/Troubleshooting

avrdude's formula was built from an unstable upstream --HEAD.
This build failure is expected behaviour.

[mcuee]

@chenrui333
Please help if you got the time. Thanks.

[stefanrueger]

OK, last merges done. I'll give this a "cooling off" period of a few hours before bumping the version to 8.1 this (European) afternoon.

[mcuee]

OK, last merges done. I'll give this a "cooling off" period of a few hours before bumping the version to 8.1 this (European) afternoon.

I just checked the latest CI builds and they seem to be working as intended.

[stefanrueger]

latest CI builds [...] seem to be working as intended

Thanks @mcuee. The CI doesn't work right now but I think it's a temporary unavailability of some package github-side.

[mcuee]

latest CI builds [...] seem to be working as intended

Thanks @mcuee. The CI doesn't work right now but I think it's a temporary unavailability of some package github-side.

Everything is fine now.

[mcuee]

Not for the two deployment github actions. Further discussions to be done in the below two issues.

• GitHub deploy action for avrdude failed to run #2038
• github avrdude_packing_release action failed #2039