I used the time to fix a few remaining bugs in the nanokernel. Maybe
the only remaining things are "features".
Specifically, I added binary compatibility to adjtime() [read my
other message about adjtime() in general], and I fixed some scaling
problems where old applications still expect microseconds. Most
notably, maybe, time doesn't run backwards occasionally any longer ;-)
Old applications will still compile, unless they are very Linux
specific, because the "#define NANO" magic is implemented now [I
don't like it, but that's the obvious standard]. Thus kernel parts
#including <linux/timex.h> need to #define NANO before. Ugly, but
true.
I made tests with ntpd 4.0.91f and xntpd3-5.93, and results were
equally good/bad. It's your turn now! [The next snapshot will most
likely have clock_getres() from POSIX.4, because the current
resolution of 1ns on my Pentium-100 is simply a lie...]
Regards,
Ulrich Windl
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/