Kernel:
Change statistics to use the *uptime() timescale (ie: relative to
boottime) rather than the UTC aligned timescale. This makes the
device statistics code oblivious to clock steps.
Change timestamps to bintime format, they are cheaper.
Remove the "busy_count", and replace it with two counter fields:
"start_count" and "end_count", which are updated in the down and
up paths respectively. This removes the locking constraint on
devstat.
Add a timestamp argument to devstat_start_transaction(), this will
normally be a timestamp set by the *_bio() function in bp->bio_t0.
Use this field to calculate duration of I/O operations.
Add two timestamp arguments to devstat_end_transaction(), one is
the current time, a NULL pointer means "take timestamp yourself",
the other is the timestamp of when this transaction started (see
above).
Change calculation of busy_time to operate on "the salami principle":
Only when we are idle, which we can determine by the start+end
counts being identical, do we update the "busy_from" field in the
down path. In the up path we accumulate the timeslice in busy_time
and update busy_from.
Change the byte_* and num_* fields into two arrays: bytes[] and
operations[].
Userland:
Change the misleading "busy_time" name to be called "snap_time" and
make the time long double since that is what most users need anyway,
fill it using clock_gettime(CLOCK_MONOTONIC) to put it on the same
timescale as the kernel fields.
Change devstat_compute_etime() to operate on struct bintime.
Remove the version 2 legacy interface: the change to bintime makes
compatibility far too expensive.
Fix a bug in systat's "vm" page where boot relative busy times would
be bogus.
Bump __FreeBSD_version to 500107
Review & Collaboration by: ken
one three times before we did the dump. Also, we printed 0x00 for the
tuple type rather than the actual tuple type. Now, we print the
actual tuple type. This appears to have no ill effects.
Should get rid of the
Code NN not found
and
code Unknown ignored
messages. The ignored messages are still generated for tuples tuples
who have a minimum length set and we find a tuple of that type that's
shorter than the minimum length.
how `crc' is actually defined.
- Remove an unnecessary `extern' variable declaration.
Data type corrections:
- Define a variable which contains a file byte offset value as type
off_t as required by the `crc' function.
- Change the type of a variable carrying a CRC checksum from `u_long'
to `uint32_t'.
- Substitute the wrong `extern' variable declaration of `crc_total'
by putting a correct one in the shared header extern.h.
`crc_total' is defined as an `uint32_t', thus fixing
incorrect mtree checksums on big-endian LP64 machines.
how `crc' is actually defined.
Data type corrections:
- Define variables which contain file byte offset values as type
off_t as required by the `crc' function.
- Change the type of a variable carrying a CRC checksum from `u_long'
to `uint32_t'.
- Parse the length of a file with sscanf as `intmax_t'
(as there is no conversion specifier for `off_t').
Style(9):
- Put an empty line between #include directives for system and user
header files.
and due to buffering they would sometimes come out after the actual
error message when mkheaders() failed due to an unknown device, so you'd
get an error messages followed by 20 or 30 lines of harmless warnings.
There are lots of other warning messages in config(8) that are printed
on stdout, but these were the most egregious (at least with LINT).
a filename pattern, and also wrt filenames given on the command line.
Now if a file is listed as a specific entry, it will not *also* be
processed by an entry specifying a pattern. And filename-patterns
will now only match existing files (ignoring directories, etc).
MFC after: 3 weeks
will contain the pid for a process group. This means the file must
contain a negative value (as would be needed in the 'kill' commmand).
I still need to write man-page update before MFC-ing.
This started by rewriting the get_pid() routine. Later I looked at
what OpenBSD has, and included a few ideas from their send_signal()
routine. So, parts of this change are from OpenBSD, even though
OpenBSD does not actually have a 'U' flag.
PR: bin/28435
Reviewed by: no objections on freebsd-arch
MFC after: 3 weeks
warning message if -s is specified and it rotates a file that expects
to be compressed. This warning message is not printed if -R is also
specified, because we assume a -sR request is coming from the process
which would have been signaled, and that it has already released the
logfile.
Indirectly noticed by: sheldonh
ifconfig equivalents. This is the first step in removing them from
the system. Users of wicontrol to configure the wireless card are
strongly encouraged to change their scripts, as sometime in the future
all configuration of the cards that isn't in ifconfig will be removed
with extreme prejustice.
and config-file entries which specify a filename-pattern (glob). It is
still not perfectly-right, but at least it isn't completely-wrong.
Reviewed by: no objections on freebsd-arch
MFC after: 3 weeks
MFC addendum: (or after the code-freeze of 4.x is lifted)
should rotate all files given on the command, even if they don't seem to
need to be rotated. This would be used by some other command that decides
the given log file(s) should be rotated, but wants the "how" of that rotation
to be determined by entries to newsyslog. Wes expects to change syslogd to
take advantage of this. Man page will be updated after we're sure this is
all working the way we want it to.
Reviewed by: no objections on freebsd-arch
MFC after: 3 weeks
MFC addendum: (or after the code-freeze of 4.x is lifted)
not send a signal to any processes. Also add a config-file flag of 'N' or
'n', which indicates that the given logfile has no process which needs a
signal when it is rotated. Both of these are based on changes NetBSD
has made, although the implementation is somewhat different.
PR: bin/36553 (2nd half)
Reviewed by: no objections on freebsd-arch
Obtained from: NetBSD (in spirit, at least)
MFC after: 3 weeks
read from CD from 2k to 16k, because in the modern world of meta-packages
(Gnome et al) the length of this list could easily owerflow limit causing
strange things to happen, ranging from installation failure due to list
truncation to complete stack trashing (there is very vague bounds checking).
For example, x11/gnome2-fifth-toe runtime dependencies list is 2,418 bytes
long.
Due to obvious reasons, this is an immediate MFC candidate.
Sponsored by: Porta Software Ltd
MFC after: 1 day
