Hi Joep, I tried to test this, it works fine with my PIC's slave echoing chars. I tried to turn off master, repeatedly and randomly, and it works. But I don't know if this is the correct test.
Anway, on normal use, it just works perfectly. Cheers, Seb 2009/1/2 Sebastien Lelong <[email protected]> > "ten" probably means "until" :) > > So how do I do 2. ? Randomly switch off the master ? Maybe I could test > without your modifications (observe the problem) then with (observe the fix) > several times, so hopefully there's one time which will be interrupt the > com. inside a packet ? > > Most importantly: this modifications is aimed to be included in the 0.1 > release. Are you confident enough to include it at the last minute ? Is it > worth to take the risk ? This is a real question, I don't mean I don't want > to include it, it's just up to you. If you decide to include it, you could > update the CHANGELOG to talk about this improvement (which IIRC did not > exist before, so using jallib really improves i2c com.) > > > Seb > > 2009/1/2 Joep Suijs <[email protected]> > >> >> Ten end of this week? When would that be? ;) >> >> I think it is important to test regression: does it still work like it >> should. >> 2 is quite dificult to repreduce and even within 2 there are multiple >> states to recover. We'll learn if the recovery is sufficient along the >> way. >> But having said that, I'm confident is does. I knew it would be >> possible, just did not look into this. And when I spoke to a friend, I >> told me about the atmel datasheet. >> And with my logic analyser, I could verify my waveform is equal to the >> one in the datasheet :) >> >> Joep >> >> 2009/1/2, Sebastien Lelong <[email protected]>: >> > I can test this, but probably not before the end of this WE... >> > What would be the test scenario ? >> > >> > 1. make 2 pics talking >> > 2. while master is sending a packet, reset it >> > 3. make sure after reset, pics can still talk >> > >> > >> > How would I do 2. ? How did you perform this test ? >> > >> > >> > Seb >> > >> > 2009/1/2 Joep Suijs <[email protected]> >> > >> > > >> > > Hi All, >> > > >> > > I added some code in i2c software master to clear the i2c bus, >> > > regardless the state it is in. This is usefull in cases where the >> > > master gets resets, while communicating with a slave. The slave may >> > > have been addressed and sending data and that needs recovery. >> > > >> > > I tested this with an i2c eeprom and seems to work okay in the way it >> > > does not cause any problems. Can someone else (Seb?) test this code >> > > also to make sure it does not cause any new issue's? >> > > >> > > I'll intend to add this code for hardware i2c too. >> > > >> > > Joep >> > > >> > > >> > > >> > >> > >> > >> > -- >> > Sébastien Lelong >> > http://www.sirloon.net >> > http://sirbot.org >> > >> > > >> > >> >> >> >> > > > -- > Sébastien Lelong > http://www.sirloon.net > http://sirbot.org > -- Sébastien Lelong http://www.sirloon.net http://sirbot.org --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "jallib" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/jallib?hl=en -~----------~----~----~----~------~----~------~--~---
