Make default EEPROM readback values 0xff for STK500 v1 programmer

[stefanrueger]

Fixes #1713

That bug probably affects all classic parts. Rather than changing the config readback values to 0xff 0xff, this PR

• Keeps the readback values as specified in the .atdf (0x00 0x00)
• Insert a statement in stk500.c to change 0x00 0x00 values to 0xff 0xff

Please check a few other parts than m328p (basically any classic part) whether they exhibit the EEPROM writing problem without this PR and whether this PR would introduce a regression.

[mcuee]

I will carry out some tests with this PR. The plan is to test with ATmega328P, ATmega8A and ATmega256.

[mcuee]

Firstly this PR is good for ATmega328P. Tested under macOS 14.4.1 (Mac Mini M1 2020).

mcuee@mcuees-Mac-mini avrdude_bin % avrdude -c usbasp -p m328p -T "erase eeprom"
avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e950f (probably m328p)

avrdude: processing -T erase eeprom
Caching | ################################################## | 100% 0.33 s 
avrdude: synching cache to device ... 
Writing | ################################################## | 100% 10.85 s 

avrdude done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % ./avrdude_pr1753 -C ./avrdude_pr1753.conf -c stk500v1 -P /dev/tty.usbserial-1110 -p m328p -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude_pr1753: AVR device initialized and ready to accept instructions
avrdude_pr1753: device signature = 0x1e950f (probably m328p)

avrdude_pr1753: processing -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude_pr1753: reading input file holes_pack_my_box_1024B.hex for eeprom
                with 252 bytes in 2 sections within [0x82, 0x37d]
                using 64 pages and 4 pad bytes
avrdude_pr1753: writing 252 bytes eeprom ...
Writing | ################################################## | 100% 1.14 s 
avrdude_pr1753: 252 bytes of eeprom written
avrdude_pr1753: verifying eeprom memory against holes_pack_my_box_1024B.hex
Reading | ################################################## | 100% 0.17 s 
avrdude_pr1753: 252 bytes of eeprom verified

avrdude_pr1753 done.  Thank you.

[mcuee]

git main has issues with ATmega2560 as well. This PR is good to fix the issue again.
(Note: I erase the EEPROM using usbasp before the tesing. If not it may appear that the result is okay for git main).

mcuee@mcuees-Mac-mini avrdude_bin % avrdude -c usbasp -p m2560 -T "erase eeprom"
avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e9801 (probably m2560)

avrdude: processing -T erase eeprom
Caching | ################################################## | 100% 0.82 s 
avrdude: synching cache to device ... 
Writing | ################################################## | 100% 6.55 s 

avrdude done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % ./avrdude -c stk500v1 -P /dev/tty.usbserial-1110 -p m2560 -U eeprom:w:holes_pack_my_box_1024B.hex:i        
avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e9801 (probably m2560)

avrdude: processing -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude: reading input file holes_pack_my_box_1024B.hex for eeprom
         with 252 bytes in 2 sections within [0x82, 0x37d]
         using 32 pages and 4 pad bytes
avrdude: writing 252 bytes eeprom ...
Writing | ################################################## | 100% 0.53 s 
avrdude: 252 bytes of eeprom written
avrdude: verifying eeprom memory against holes_pack_my_box_1024B.hex
Reading | ################################################## | 100% 0.06 s 
avrdude warning: verification mismatch
        device 0xff != input 0x50 at addr 0x0082 (error)
avrdude error: verification mismatch

avrdude done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % ./avrdude_pr1753 -C ./avrdude_pr1753.conf -c stk500v1 -P /dev/tty.usbserial-1110 -p m2560 -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude_pr1753: AVR device initialized and ready to accept instructions
avrdude_pr1753: device signature = 0x1e9801 (probably m2560)

avrdude_pr1753: processing -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude_pr1753: reading input file holes_pack_my_box_1024B.hex for eeprom
                with 252 bytes in 2 sections within [0x82, 0x37d]
                using 32 pages and 4 pad bytes
avrdude_pr1753: writing 252 bytes eeprom ...
Writing | ################################################## | 100% 1.08 s 
avrdude_pr1753: 252 bytes of eeprom written
avrdude_pr1753: verifying eeprom memory against holes_pack_my_box_1024B.hex
Reading | ################################################## | 100% 0.12 s 
avrdude_pr1753: 252 bytes of eeprom verified

avrdude_pr1753 done.  Thank you.

[mcuee]

For ATmega128A, both git main and this PR are okay.

mcuee@mcuees-Mac-mini avrdude_bin % avrdude -c usbasp -p m128 -T "erase eeprom"
avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e9702 (probably m128)

avrdude: processing -T erase eeprom
Caching | ################################################## | 100% 0.82 s 
avrdude: synching cache to device ... 
Writing | ################################################## | 100% 42.36 s 

avrdude done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % ./avrdude -c stk500v1 -P /dev/tty.usbserial-1110 -p m128 -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e9702 (probably m128)

avrdude: processing -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude: reading input file holes_pack_my_box_1024B.hex for eeprom
         with 252 bytes in 2 sections within [0x82, 0x37d]
         using 32 pages and 4 pad bytes
avrdude: writing 252 bytes eeprom ...
Writing | ################################################## | 100% 1.22 s 
avrdude: 252 bytes of eeprom written
avrdude: verifying eeprom memory against holes_pack_my_box_1024B.hex
Reading | ################################################## | 100% 0.06 s 
avrdude: 252 bytes of eeprom verified

avrdude done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % ./avrdude -c stk500v1 -P /dev/tty.usbserial-1110 -p m128 -U eeprom:w:the_quick_brown_fox_1024B.hex:i 
avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e9702 (probably m128)

avrdude: processing -U eeprom:w:the_quick_brown_fox_1024B.hex:i
avrdude: reading input file the_quick_brown_fox_1024B.hex for eeprom
         with 1024 bytes in 1 section within [0, 0x3ff]
         using 128 pages and 0 pad bytes
avrdude: writing 1024 bytes eeprom ...
Writing | ################################################## | 100% 9.60 s 
avrdude: 1024 bytes of eeprom written
avrdude: verifying eeprom memory against the_quick_brown_fox_1024B.hex
Reading | ################################################## | 100% 0.48 s 
avrdude: 1024 bytes of eeprom verified

avrdude done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % avrdude -c usbasp -p m128 -T "erase eeprom"
avrdude: AVR device initialized and ready to accept instructions
avrdude: device signature = 0x1e9702 (probably m128)

avrdude: processing -T erase eeprom
Caching | ################################################## | 100% 0.82 s 
avrdude: synching cache to device ... 
Writing | ################################################## | 100% 10.54 s 

avrdude done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % ./avrdude_pr1753 -C ./avrdude_pr1753.conf -c stk500v1 -P /dev/tty.usbserial-1110 -p m128 -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude_pr1753: AVR device initialized and ready to accept instructions
avrdude_pr1753: device signature = 0x1e9702 (probably m128)

avrdude_pr1753: processing -U eeprom:w:holes_pack_my_box_1024B.hex:i
avrdude_pr1753: reading input file holes_pack_my_box_1024B.hex for eeprom
                with 252 bytes in 2 sections within [0x82, 0x37d]
                using 32 pages and 4 pad bytes
avrdude_pr1753: writing 252 bytes eeprom ...
Writing | ################################################## | 100% 2.44 s 
avrdude_pr1753: 252 bytes of eeprom written
avrdude_pr1753: verifying eeprom memory against holes_pack_my_box_1024B.hex
Reading | ################################################## | 100% 0.12 s 
avrdude_pr1753: 252 bytes of eeprom verified

avrdude_pr1753 done.  Thank you.

mcuee@mcuees-Mac-mini avrdude_bin % ./avrdude_pr1753 -C ./avrdude_pr1753.conf -c stk500v1 -P /dev/tty.usbserial-1110 -p m128 -U eeprom:w:the_quick_brown_fox_1024B.hex:i 
avrdude_pr1753: AVR device initialized and ready to accept instructions
avrdude_pr1753: device signature = 0x1e9702 (probably m128)

avrdude_pr1753: processing -U eeprom:w:the_quick_brown_fox_1024B.hex:i
avrdude_pr1753: reading input file the_quick_brown_fox_1024B.hex for eeprom
                with 1024 bytes in 1 section within [0, 0x3ff]
                using 128 pages and 0 pad bytes
avrdude_pr1753: writing 1024 bytes eeprom ...
Writing | ################################################## | 100% 9.60 s 
avrdude_pr1753: 1024 bytes of eeprom written
avrdude_pr1753: verifying eeprom memory against the_quick_brown_fox_1024B.hex
Reading | ################################################## | 100% 0.48 s 
avrdude_pr1753: 1024 bytes of eeprom verified

avrdude_pr1753 done.  Thank you.

[mcuee]

@stefanrueger

I think this PR is now good to be merged.

[mcuee]

One more sanity check, it does not cause regression with "Arduino as ISP" as well.

mcuee@mcuees-Mac-mini tools % ./test-avrdude -e ../build_darwin/src/avrdude -p "-c arduino_as_isp -P /dev/tty.usbserial-224570 -p m168p -b 19200"
Testing ../build_darwin/src/avrdude version 7.3-20240417 (4b2c9947)
Prepare "-c arduino_as_isp -P /dev/tty.usbserial-224570 -p m168p -b 19200" and press 'enter' or 'space' to continue. Press any other key to skip
✅   1.355 s: fuse access: clear, set and read eesave fuse bit
✅   1.327 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   1.688 s: chip erase
✅   6.676 s: flash -U write/verify holes_rjmp_loops_16384B.hex
✅   3.100 s: flash -T write/verify holes_rjmp_loops_16384B.hex
✅   1.783 s: eeprom check whether programmer can flip 0s to 1s
✅   8.138 s: eeprom -U write/verify holes_pack_my_box_512B.hex
✅  15.363 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_512B.hex
✅   3.447 s: chip erase and spot check flash is actually erased
✅   1.809 s: spot check eeprom is erased, too

BTW, git main is okay with "Arduino as ISP".

mcuee@mcuees-Mac-mini tools % ./test-avrdude -p "-c arduino_as_isp -P /dev/tty.usbserial-224570 -p m168p -b 19200"
Testing avrdude version 7.3-20240417 (8fe729f0)
Prepare "-c arduino_as_isp -P /dev/tty.usbserial-224570 -p m168p -b 19200" and press 'enter' or 'space' to continue. Press any other key to skip
✅   1.359 s: fuse access: clear, set and read eesave fuse bit
✅   1.326 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   1.694 s: chip erase
✅   6.682 s: flash -U write/verify holes_rjmp_loops_16384B.hex
✅   3.077 s: flash -T write/verify holes_rjmp_loops_16384B.hex
✅   1.776 s: eeprom check whether programmer can flip 0s to 1s
✅   8.123 s: eeprom -U write/verify holes_pack_my_box_512B.hex
✅  15.347 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_512B.hex
✅   3.448 s: chip erase and spot check flash is actually erased
✅   1.814 s: spot check eeprom is erased, too

[mcuee]

For ATmega128A, both git main and this PR are okay.

Same for ATmega8A.

This PR is good.

mcuee@mcuees-Mac-mini tools % ./test-avrdude -e ../build_darwin/src/avrdude  -p "-c stk500v1 -P /dev/tty.usbserial-1110 -p m8a"
Testing ../build_darwin/src/avrdude version 7.3-20240417 (4b2c9947)
Prepare "-c stk500v1 -P /dev/tty.usbserial-1110 -p m8a" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.999 s: fuse access: clear, set and read eesave fuse bit
✅   1.015 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   1.143 s: chip erase
✅   2.027 s: flash -U write/verify holes_rjmp_loops_8192B.hex
✅   1.417 s: flash -T write/verify holes_rjmp_loops_8192B.hex
✅   1.166 s: eeprom check whether programmer can flip 0s to 1s
✅   2.354 s: eeprom -U write/verify holes_pack_my_box_512B.hex
✅   3.744 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_512B.hex
✅   1.538 s: chip erase and spot check flash is actually erased
✅   1.097 s: spot check eeprom is erased, too

git main is also good.

mcuee@mcuees-Mac-mini tools % ./test-avrdude -p "-c stk500v1 -P /dev/tty.usbserial-1110 -p m8a"  
Testing avrdude version 7.3-20240417 (8fe729f0)
Prepare "-c stk500v1 -P /dev/tty.usbserial-1110 -p m8a" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.993 s: fuse access: clear, set and read eesave fuse bit
✅   1.025 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   1.151 s: chip erase
✅   2.026 s: flash -U write/verify holes_rjmp_loops_8192B.hex
✅   1.425 s: flash -T write/verify holes_rjmp_loops_8192B.hex
✅   1.166 s: eeprom check whether programmer can flip 0s to 1s
✅   2.346 s: eeprom -U write/verify holes_pack_my_box_512B.hex
✅   3.745 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_512B.hex
✅   1.536 s: chip erase and spot check flash is actually erased
✅   1.109 s: spot check eeprom is erased, too

[mcuee]

For ATmega168P, situation is the same as ATmega328P. git main is bad but this PR is good.

mcuee@mcuees-Mac-mini tools % ./test-avrdude -p "-c stk500v1 -P /dev/tty.usbserial-1110 -p m168p"
Testing avrdude version 7.3-20240417 (8fe729f0)
Prepare "-c stk500v1 -P /dev/tty.usbserial-1110 -p m168p" and press 'enter' or 'space' to continue. Press any other key to skip
✅   1.005 s: fuse access: clear, set and read eesave fuse bit
✅   1.025 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   1.211 s: chip erase
✅   3.448 s: flash -U write/verify holes_rjmp_loops_16384B.hex
✅   2.068 s: flash -T write/verify holes_rjmp_loops_16384B.hex
✅   1.046 s: eeprom check whether programmer can flip 0s to 1s
❌   1.727 s: eeprom -U write/verify holes_pack_my_box_512B.hex (failed command below)
$ avrdude -qq -c stk500v1 -P /dev/tty.usbserial-1110 -p m168p -Ueeprom:w:./test_files/holes_pack_my_box_512B.hex
✅   3.731 s: eeprom -U write/verify the_quick_brown_fox_512B.hex
✅   2.480 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_512B.hex
✅   2.257 s: chip erase and spot check flash is actually erased
✅   1.117 s: spot check eeprom is erased, too

mcuee@mcuees-Mac-mini tools % ./test-avrdude -e ../build_darwin/src/avrdude -p "-c stk500v1 -P /dev/tty.usbserial-1110 -p m168p"
Testing ../build_darwin/src/avrdude version 7.3-20240417 (4b2c9947)
Prepare "-c stk500v1 -P /dev/tty.usbserial-1110 -p m168p" and press 'enter' or 'space' to continue. Press any other key to skip
✅   0.990 s: fuse access: clear, set and read eesave fuse bit
✅   1.027 s: fuse access: set eesave fusebit to delete EEPROM on chip erase
✅   1.201 s: chip erase
✅   3.446 s: flash -U write/verify holes_rjmp_loops_16384B.hex
✅   2.067 s: flash -T write/verify holes_rjmp_loops_16384B.hex
✅   1.147 s: eeprom check whether programmer can flip 0s to 1s
✅   1.766 s: eeprom -U write/verify holes_pack_my_box_512B.hex
✅   2.577 s: eeprom -T write/verify holes_{the_five_boxing_wizards,pack_my_box}_512B.hex
✅   2.256 s: chip erase and spot check flash is actually erased
✅   1.118 s: spot check eeprom is erased, too