> Technically you could still implement this and if firmware happens to > read the RTC (and doesn't correct things) then we won't be able to > correct things. ...but certainly if you read the old time and then > the new time and the old time was < 11/31 and the new time was >= > 11/31 you know you need to correct. > > I'd say that there's a good chance that other firmware (UBoot) doesn't > actually read the RTC anyway. Why would it? We only do it for even > log, right?
Hah! Sounds like you assume U-Boot was doing the things it does with comprehensible reasoning. From my experience, that's a dangerous wager (not that I'm biased or anything... ;) ). FWIW, they seem to be having a huge (175kloc, of only the finest copied kernel code from 10+ years ago, I presume) repository of RTC drivers, a separate command line command to read/write to it, and an SNTP client. Still, you're right that adding the GET_TIME check wouldn't hurt... at worst it just does nothing. I'll try it out. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
