- Copy stable/10 (r259064) to releng/10.0 as part of the
10.0-RELEASE cycle.
- Update __FreeBSD_version [1]
- Set branch name to -RC1

[1] 10.0-CURRENT __FreeBSD_version value ended at '55', so
start releng/10.0 at '100' so the branch is started with
a value ending in zero.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

Copy head (r256279) to stable/10 as part of the 10.0-RELEASE cycle.

Approved by: re (implicit)
Sponsored by: The FreeBSD Foundation

Revert r253748,253749

This WIP should not have been committed yet.

Pointyhat to: avg

remove needless inclusion of machine/cpu.h in userland

MFC after: 21 days

Rework all non-contributed files that use `struct timezone'.

This structure is not part of POSIX. According to POSIX, gettimeofday()
has the following prototype:

int gettimeofday(struct timeval *restrict tp, void *restrict tzp);

Also, POSIX states that gettimeofday() shall return 0 (as long as tzp is
not used). Remove dead error handling code. Also use NULL for a
nul-pointer instead of integer 0.

While there, change all pieces of code that only use tv_sec to use
time(3), as this provides less overhead.

If not compiled for debugging, redirect standard input/output/error
to /dev/null before becoming a daemon.

Correct various mistakes in the last commit.

Use sysctlbyname

Invoke tzset(3) within the main loop to catch changes to /etc/localtime
when running as a daemon in the hope that it will fix situations
where the CMOS clock was apparently set using the stale TZ offset.

Use __FBSDID() to quiet GCC 3.3 warnings.

o Use real prototypes
o Remove __P.
o main(int, char *[])

Fix a prototype and set WARNS=2.

Submitted by: Mike Barcroft <mike@q9media.com>

$Id$ -> $FreeBSD$

Correct use of .Nm. Add missing .El. Add rcsid. Remove unused #include.

Implement Helbig idea of offset calculation reducing mktime() calls
Immediately exit if /etc/wall_cmos_clock not present

Add workaround suggestion for 'nonexistent time' diagnostic

Minimize clock drift between getting and setting time
Submitted by: bde

Add Id, add usage function.

compare return value from getopt against -1 rather than EOF, per the final
posix standard on the topic.

Detect if /etc/wall_cmos_clock gone while adjkerntz paused.

Dynamically sense /etc/wall_cmos_clock presence/absense

Fix error with multiply 'adjkerntz -a' copies running.
Manage adjkerntz kernel variable even for UTC clocks.
Code cleanup.

Remove trailing whitespace.

Use more AI to predict initial/final isdst,
latest isdst preffered.

Use timelocal() instead of mktime() hack

Fix resetting adjkerntz to 0 (pointed by Bruce),
forget to initialize mib variable.

Optimize setting of DISRTCSET

Clean internal logic, helps for some unusual cases

-i case goes to background pause and acts like -a case on receiving
SIGTERM. This helps to keep CMOS clock updated before reboot.
Idea from J.Wunsch.

Restore back -i, -a

Remove options, run adjkerntz as daemon now

Now:
adjkerntz -i writes kernel clock only, don't touch RTC clock
adjkerntz -a writes RTC clock only, don't touch kernel clock

Modify adjkerntz to not write RTC clock per initial call,
suggested by Bruce

rewritted to use sysctl, no more /var/run/.adjkerntz now

Cosmetique patch from bde

Fix sign of offset diff.
Use Bruce initial/final hour scheme to avoid critical hours.

This commit was generated by cvs2svn to compensate for changes in r867,
which included commits to RCS files with non-trunk default branches.

Attempt to import adjkerntz...