I still have the problem of broken d2s from time to time, but I'm flying with the dirty workaround I described a few mails ago. It works ok for the moment as it takes a look at the current value of d2 and replaces it with the previous one, if it seems to be broken. In case of a replacement, a counter increases by one so I'm able to track the events. It occures randomly but I have often flights with at least one event and in one flight of about 25 minutes I saw 12 events in the log.
As you see, I don't have a solution which is pushable and I didn't have taken the time to search the cause of the problem so far. I also never tried it on master branch.
As no one else seems to have this problem, it could be a hard- and software specific problem of my setup. Anyway, I'll give an update when I have found the source of the problem.
Gesendet: Mittwoch, 17. April 2013 um 11:02 Uhr
Von: "Felix Ruess" <address@hidden>
An: "Paparazzi devel list" <address@hidden>
Betreff: Re: [Paparazzi-devel] Critical behaviour of baro_ms5611_i2c
what is the status on this?
If this needs to be fixed in v4.2 please let us know, or even better make a pull request with the fix on github ;-)
Did you also run long tests for the code in master?
It's really important that problems like this get properly fixed in the main Paparazzi repository.
But for that to become a reality we need the help from all of you to get the fixes back (not enough if you have them locally fixed for you).
There are so many drivers and combinations supported that not everything can be tested by the few core committers we have.