Il 15/02/2013 03:49, Antoine Mathys ha scritto: > First, the ds1338 code was in a poor state and never handled the 12 hour > clock correctly. My first patch failed to fully fix the problem so I had > to write a second one, but at no point did Peter or I introduce a > regression, quite the opposite. > > Second, I don't know where you got the idea that I refuse to write test > cases. I just didn't have one ready or in the works at the time. > > Third, bug 1090558 in mc146818rtc is a good example of a bug which was > not due to insufficient testing, but to poorly structured code. > > There is no point worrying about unit testing if you accept code of such > low quality. This goes for the tests too. For instance > cmos_get_date_time() in tests/rtc-test.c doesn't work correctly in 12 > hour mode. > > Fourth, I am not interested in the PC architecture, I only wrote a fix > for bug 1090558 because Paolo asked me to. It is nice to see that fixing > your crappy code makes me "not a nice guy" who is making things worse. > But don't worry, I'll focus on ARM from now on.
Hey hey, no reason to get excited. Yes, some code is of pretty low quality. We're getting better, the main problem is that without a testing infrastructure there's only so much you can do for code quality. Hence Andreas's request. Thanks for your PC patch. Nobody said you're making things worse. Paolo