> Watching the recent discussions on daylight saving time etc got me
> wondering again about the point of having the timezone hardwired
> into the kernel.
> 
> I can't see why. Is there a major problem with assuming the RTC in
> a machine is set to the local (default) time ?

A problem comes in with Daylight Savings Time.  Unix systems tend to
instantly and accurately adjust for it, and then keep all internally
stored values in UTC.  Many programs require that time is a
monotonically increasing function.

If we assume that the RTC is set to local time, what happens when
Daylight Savings kicks in?  This would be a problem, if you think
about it.  (Unless perhaps the kernel automatically wrote the new
value to the RTC)

In the current scheme, if someone wants to adjust their RTC when DST
comes around, they can recompile their kernel, shutdown, set their RTC
from the BIOS, and reboot.  This procedure keeps time monotonically
increasing.

This is kind of a pain though, which is why some people like to keep
their machines set to UTC instead.

	Mark
--------------------------------------------------------------------
Email: Mark_Weaver@brown.edu           | Brown University
PGP Key: finger mhw@cs.brown.edu       | Dept of Computer Science

------------------------------------------------------------------------------