Commit Graph

24 Commits

Author SHA1 Message Date
peter 7644942c0e Minor KNF, replace some spaces with tabs. 2006-09-24 18:24:55 +00:00
tsutsui 292139732b Tweak some debug messages. 2006-09-20 09:37:28 +00:00
gdamore 06522e0f81 Pass the filesystem time to clock drivers in the todr_chip_handle, so that
vax, and pmax can use it.
2006-09-16 00:50:52 +00:00
gdamore 747bc86364 Comment nit (remove first person), from simon@. 2006-09-12 21:38:55 +00:00
gdamore ba5268f32b Centralize rtc_offset handling for ports that __HAVE_GENERIC_TODR and use
a simple second counter instead of the clock_ymdhms support.
2006-09-12 20:47:39 +00:00
gdamore 2a1a021930 Remove attempt to incorrectly support fictional double leap seconds.
While here, make the comparisions for hour, min, and sec use ">" consistently
like the date comparisions.
2006-09-12 15:25:05 +00:00
gdamore 73700aff66 Fix to compile on platforms without __HAVE_GENERIC_TODR. Sorry about
that.  Pointed out jdc@ and he@.
2006-09-12 14:27:17 +00:00
gdamore 58af66bd2c Add some optional debug (enabled with TODR_DEBUG) to see what your RTC
is giving you.  Also, while here, bump the threshold for "preposterous"
dates by 20 years.  Now any date earlier than late 1994 is considered
preposterous.
2006-09-12 05:47:47 +00:00
gdamore 9a6f8be8b3 Don't bother checking wday, that breaks some clock chips. Thank to Kurt Schreiner <ks at ub dot uni-mainz dot de>. 2006-09-11 20:48:55 +00:00
gdamore a012c5d7ea Allow for leap seconds. 2006-09-10 16:25:32 +00:00
gavan 843ed515f0 dsrtc_settime already takes a volatile struct timeval *, also qualify
with volatile for dsrtc_gettime.
2006-09-10 14:06:54 +00:00
tsutsui ca3e00268a - add more sanity checks (from playstation2/clock.c)
- fix a typo
2006-09-10 07:06:48 +00:00
gdamore 6347eaa294 Run tc_setclock at splclock(). Suggested by kardel@. 2006-09-07 15:49:49 +00:00
dogcow fe798dd421 Now compiles on systems without __HAVE_GENERIC_TODR 2006-09-07 07:26:07 +00:00
gdamore 8e82fb3308 Eliminate the 2038 year check, because clock_ymdhms_to_sec already performs
that check.  So instead, we test to make sure we have a non-negative second
after the conversion.

While here, we also add a check to round the second counter up when setting
time if we are more than half-way into the second.
2006-09-07 04:51:42 +00:00
gdamore 6b6be489ab Add a few consistency checks for ymdhms. Taken from the mc146818 code. 2006-09-07 04:24:26 +00:00
uwe d9aa7ac028 Make timeval argument to todr_settime volatile to keep non-timecounter
ports happy.
2006-09-07 01:50:49 +00:00
uwe 076748ce26 Mark sizeof(tvp->tv_sec) check CONSTCOND.
No space between the sizeof and the paren.
2006-09-07 01:43:52 +00:00
gdamore 942bfe3e30 Add support for new todr entry points for drivers- todr_gettime_ymdhms and
todr_settime_ymdhms.  This allows drivers to avoid repeated conversions, and
allows us to centralize check for Y2038 overflow.   As discussed on
tech-kern.
2006-09-07 00:10:49 +00:00
matt 7cdae31eb1 Before calling todr_gettime, fill in tv.tv_sec. At least one port (vax)
needs that functionality.
2006-09-05 19:32:17 +00:00
gdamore 7b0092122f Update the base time (used when no filesystem or rtc time exists) to 12pm,
Jan 1, 2006. This is somewhat arbitrary, but its a heck of a lot better than
mid 1991. :-)
2006-09-03 17:13:04 +00:00
gdamore 487ed45995 Don't warn about RTC losing/gaining days if the filesystem time is bogus. 2006-09-03 06:37:55 +00:00
gdamore a6abab1852 Incorporate changes from x86/i386 as follows:
1) don't set a clock when panicing during early boot
2) if the filesystem time is newer than the rtc time (by at least 2 days) then
revert to the filesystem time.
3) use x86 style messaging.

We still use a threshold of 2 days of gain or loss in time to warn though.
2006-09-03 05:25:05 +00:00
gdamore c583da0d13 Add MI implementation of inittodr, todr_attach, and resettodr.
This is triggered upon __HAVE_GENRIC_TODR in machine/types.h.  Conversion of
evbmips port forthcoming.
2006-09-02 20:18:00 +00:00