Bootloaders and Serial Based Programmers which works with very low baudrate <=1200bps

[mcuee]

• Fix r/w timeout for Windows ser_send() #1987
• Retrigger ser_posix.c ser_recv() timeout #1983

Interestingly we have bug reports for the low level serial communications after so many years.

From @stefanrueger

First there is no bus for 20 years and then there are two buses within a couple of days (unrelated edge cases in low-level serial I/O: one for Posix in PR 1983, the other one for Windows...).

Should be thoroughly tested with serial I/O, best with large parts with large page sizes and low baud rates.

Now the issue is how to carry out the test with low baud rates.

[mcuee]

From my testing of #1987, urboot does not seem to work at 1200 bps or below.

I have tried Optiboot for classic AVRs as well and it does not seem to work either.

[mcuee]

I have not tried the programmers -- but I will try.

Last time I was more chasing faster programmers and bootloaders.
https://github.com/avrdudes/avrdude/wiki/avrdude-benchmark-for-Flash-and-EEPROM-programming

• avrdude benchmark results discussion #1712
• Flash Read and Write Speed Benchmark #1022

[mcuee]

@dl8dtl and @MCUdude

Just wondering if you know of such programmers and bootloaders which work below 1200bps, or even 300 bps and below. Thanks.

[stefanrueger]

@mcuee It is pretty tough to test low-baud-rate programming owing to the sheer time it takes. Writing, then reading 256 kB at 300 baud takes some 5 hours, and that's without the comms overhead. Should you test with PR #1993 make sure you burn urboot bootloaders that reduce comms overhead as much as possible, ie, select the chip erase feature so that at least chip erasing won't send 256 kB 0xff to clear the flash. I recommend -U urboot:16mhz_300baud_ce_ee_best. And if you do, start the test, set the laptop aside and check the next day. 😄

[mcuee]

First test at 250bps works. Nice.

• Extend -c urclock to work with lower baud rates #1993 (comment)

[mcuee]

PR #1933 has been merged.