Make is worthwhile haveing #ifdef SMALL by making the Makefile set it when
SMALLPROG is defined (eg by src/distrib/commonMakefile.crunch).
Shrink SMALL version even more.

WARNS=2 is the default defined in sbin/Makefile.inc. (thanks wiz)

Remove main() prototype; WARNS=2

Oops, forgot to commit this file - actually remove the kmem setgid bit.

branches: 1.8.14;
convert to new RCS Id conventions; reduce my headache

specify man pages the new way.

clean up import

needs -lkvm, not -lutil

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Correct the 6 second offset issue between the time reported by
dmesg -T and the actual time a message was produced, noted on
current-users by Geoff Wing (Oct 27, 2018).

The size of the offset would depend upon architecture, and processor,
but was the delay from starting the clocks to initialising the time
of day (after mounting root, in case that is needed).

Change the kernel to set boottime to be the time at which the
clocks were started, rather than the time at which it is init'd
(by subtracting the interval between).

Correct dmesg to properly compute the ToD based upon the
boottime (which is a timespec, not a timeval, and has been
since Jan 2009) and the time logged in the message.

Note that this can (rarely) be 1 second earlier than date reports.
This occurs when the time when the message was logged was actually
in the next second, but the timecounters have not yet processed
the tick, and so the time of the last tick, near the end of the
previous second, is reported instead. Since times are always
truncated, rather than rounded, it is occasionally possible to
observe that disparity (if you try hard enough).

IOW: sys/kern/subr_prf.c:addtstamp() uses getnanouptime() rather
than nanouptime().

Note in dmesg(8) that -T conversions are gibberish other than
when the message comes from current the running kernel. (It
could be fixed when -M is used, for messages generated by the
kernel whose corpse is being observed. But hasn't been...)

s/or/of/ (a common typo of mine) and s/or// (too many of them.)

Remove a comma and some trailing whitespace.

changes from kre@
- pass intmax to fmtydhmsf instead of time_t to avoid extra conversions.
- make -TTT mean "always keep 3 decimal digits of duration precision" (ie:
always print ms) (including trailing 0's, even .000 if it happens)
- make -T (all forms) be subject to the locale (obey the radix character)
- don't print ymd, since that would require calendar calculations to get
right.

add -TT printing ISO 8601 duration

branches: 1.21.2;

Minor wording change (no date bump needed.)

Sort flags in SYNOPSIS.

match the linux dmesg flags.

Handle new timestamp sequences.

branches: 1.17.12;
dmesg first appeared in 3BSD
http://minnie.tuhs.org/cgi-bin/utree.pl?file=3BSD/usr/src/cmd/dmesg.c
Bump date.

Bump date for previous; add comma in SEE ALSO; use Dv instead of Em.

Talk a little about the system message buffer and cross-reference
MSGBUFSIZE in options(4).

branches: 1.14.2;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

branches: 1.13.2;
Drop arguments of .Os.

branches: 1.12.2;
The default core file is /dev/mem, not /dev/kmem.

branches: 1.11.2;
Mention /var/run/dmesg.boot

branches: 1.10.4;
- cleanup for WARNS=1
- cleanup .Nm usage

convert to new RCS Id conventions; reduce my headache

vmunix -> netbsd (PR #375)

branches: 1.7.2;
clean up import

Add RCS indentifiers.

Add FILES section.

Update file names.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision

Assert postts only if we really obtained a timestamp.

Do not declare unused tstamp for SMALL. NFC.

Get back to the statically-allocated tbuf.

If a string is longer than tbuf, it cannot be a timestamp.
Therefore, no need for buffering anymore in this case.

Oops, fix thinko in the previous.

Do not discard \[.*\] which is not identified as a timestamp.

KNF. Do not unnecessarily define ADDC() for SMALL. No binary changes.

Make recent change build for SMALL builds as well as normal ones.

Fix dmesg(8) to preserve leading whitespaces of kernel messages.

Closes PR/54729.

branches: 1.43.2;

Delete a redundant (useless) assignment. NFCI.

PR toolchain/54269

Make a SMALL dmesg even smaller.

When compiled -DSMALL, dmesg does nothing at all with the kernel
timestamps (it processes no options to be told what would be the
user's desire) so make it truly do (almost) nothing, rather than
some botched partial processing of them.

The "almost" is that a SMALL dmesg will now ignore spaces at the
beginning of each new message line ... those are (vanishingly)
unlikely to occur, as the kernel puts the timestamp (which starts
with '[' there) - fixing this would have meant even more #ifdef's
as the code that ignores that leading space is the only remaining
thing that (in a SMALL) dmesg looks at the value of the "tstamp"
variable, and if we don't keep some use of it, gcc complains...

These changes affect only SMALL dmesg (as installed on boot
floppies, etc) and have no intended effect on the version that's
installed on a normal (full size) running system.

Correct the 6 second offset issue between the time reported by
dmesg -T and the actual time a message was produced, noted on
current-users by Geoff Wing (Oct 27, 2018).

The size of the offset would depend upon architecture, and processor,
but was the delay from starting the clocks to initialising the time
of day (after mounting root, in case that is needed).

Change the kernel to set boottime to be the time at which the
clocks were started, rather than the time at which it is init'd
(by subtracting the interval between).

Correct dmesg to properly compute the ToD based upon the
boottime (which is a timespec, not a timeval, and has been
since Jan 2009) and the time logged in the message.

Note that this can (rarely) be 1 second earlier than date reports.
This occurs when the time when the message was logged was actually
in the next second, but the timecounters have not yet processed
the tick, and so the time of the last tick, near the end of the
previous second, is reported instead. Since times are always
truncated, rather than rounded, it is occasionally possible to
observe that disparity (if you try hard enough).

IOW: sys/kern/subr_prf.c:addtstamp() uses getnanouptime() rather
than nanouptime().

Note in dmesg(8) that -T conversions are gibberish other than
when the message comes from current the running kernel. (It
could be fixed when -M is used, for messages generated by the
kernel whose corpse is being observed. But hasn't been...)

Whitespace. NFC.

Use the locale's numeric radix character ("decimal point") more generally
(in !SMALL versions of dmesg).

Remove #if 0'd code that somehow got mangled in the (somewhat messy)
commit process and ended up being a syntax error...

changes from kre@
- pass intmax to fmtydhmsf instead of time_t to avoid extra conversions.
- make -TTT mean "always keep 3 decimal digits of duration precision" (ie:
always print ms) (including trailing 0's, even .000 if it happens)
- make -T (all forms) be subject to the locale (obey the radix character)
- don't print ymd, since that would require calendar calculations to get
right.

Allow a third use of -T to force .NNN (always three digits) in -TT
(ISO8601 durations) format for timestamps - that is,not dropping trailing 0's.

In -TT mode (ISO8601 duration timestamps) print milliseconds as
decimal fractions of a second (as they should be) rather than integer
milliseconds (ie nnn.1means nnn seconds and 1/10 of a second, not
nnn seconds, and 1 millisecond). While here convert some inappropriate
time_t usage to intmax_t which works better (int, or long, would probably
work just as well).

add -TT printing ISO 8601 duration

branches: 1.33.2;

Make timestamp reading code adapt to whatever precision (up to
nanoseconds, 9 digits) the kernel happens to send in the timestamps
in log messages.

Output (numeric) timestamps (when produced) are unchanged and always
in microseconds (for now).

Update usage.

match the linux dmesg flags.

eat NUL's first so that the state machine is not altered by them.

handle log being before timestamp...

Handle new timestamp sequences.

branches: 1.27.40;
Use __dead

Remove the \n and tabs from the __COPYRIGHT() strings.
(Tweak some to use a consistent format.)

branches: 1.25.18; 1.25.22;
c99 initializer

Make is worthwhile haveing #ifdef SMALL by making the Makefile set it when
SMALLPROG is defined (eg by src/distrib/commonMakefile.crunch).
Shrink SMALL version even more.

Remove main() prototype; WARNS=2

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22308, verified by myself.

Make it possible to see \ in dmesg.

branches: 1.20.2;
If "SMALL" is defined don't include kvm support, primarily for install
media.

Don't check namelist when choosing to use the sysctl interface.

ANSIfy.

Use new KERN_MSGBUF sysctl if running on a live kernel, and remove
setgid kmem bit from executable.

branches: 1.16.8;
discard privs as soon as possible.

Cosmetic change;

- indent continuation line by four space.
- put an empty line at the beginning of function which has no local variable.

Fix revived bug, described in PR#3285. But use a bit simple way;
Use loop count instead of hackery pointer play.

Don't include <sys/cdefs.h> twice in same file.

Commit userland part of pr-1891.

- cleanup for WARNS=1
- cleanup .Nm usage

We need a hint of *what* address is bad on the kvm_read. Also,
there are two reads- one for the msgbuf pointer, and then that
for what it points to.

Don't get stuck in an infinite loop if the msgbuf index is 0 or 1, from
enami tsugutomo <enami@ba2.so-net.or.jp>, PR #3285.

branches: 1.8.6;
convert to new RCS Id conventions; reduce my headache

more cleanups from Jim Jegers, passed over by me.

clean up import

#include <unistd.h>, so getopt(), optarg, opterr, optind, and optopt
declarations can be removed from <stdlib.h>.

Add RCS identifiers.

changed "Id" to "Header" for rcsids

added rcs ids to all files

branches: 1.1.1;
Initial revision