fenugrec
fenugrec
Follow-up to discussion in PR https://github.com/fenugrec/freediag/pull/71 , as well as issue #68 . There's already some use of `assert()` throughout the codebase, but there could be more - for instance...
https://www.matthewsvolvosite.com/forums/viewtopic.php?f=37&t=83711 forked around 2017 and build on D2 / 850 protocol work by @aaeegg . Will need a bit of work to rebase on current master
Summary of email conversations with Tomasz K. There are issues with linux/unix timeouts in diag_tty_read() and diag_tty_write(). - SIGUSR1 is used with timer_settime() to interrupt the read() and write() calls,...
needing to recompile to add keysets is pain. - .ini file with name-value pairs ? mediocre - any "noSQL" key-value store is not really a good fit. gdbm, tdb, berkeley...
- maybe doable with simple heuristics (IIRC accessing >512kB on 512kB ROMs wraps around). - maybe , included in https://github.com/fenugrec/nisprog/issues/10 , associate ECUID with proper parameters (setdev, etc)
(must be fixed via freediag). I've had enough of "it gave an error, I don't remember what it was, it broke everything" reports
practice mode doesn't erase, and doesn't write, to the ROM. But it still tries to do the post-write verify, which will obviously fail . This produces off-putting (but spurious) errors...
adding an extra warning for those cases to avoid reflashing a corrupt or completely wrong ROM.
cks_add8() is dumb and only used for ` ... ` frames. Using the crc16 code we already have anyway would eliminate some code, at the cost of 1 more byte...
(needs to be fixed via freediag). e.g. `nisprog: Type SCAN to start ODBII Scan ` etc.