Commit Graph

76 Commits

Author SHA1 Message Date
Alfred Perlstein
5d320d4b95 This patch merges the sort fields for both pages, so you can (for
example) view io stats while sorting by process size.  Also adds
voluntary and involuntary context-switch stats to the io page because
there was lots of room.

Submitted by: Dan Nelson dnelson at allantgroup.com
2004-08-16 07:51:22 +00:00
Alfred Perlstein
dcd6f4bd48 Comment some of the 'io' functions. 2004-08-01 09:19:41 +00:00
Dag-Erling Smørgrav
9b30d69710 Adjust the show_self code (the test got inadvertantly reversed a couple of
revisions ago)

Submitted by:	Alex Vasylenko <lxv@omut.org>
2004-07-12 08:22:32 +00:00
Alfred Perlstein
98c151d6d9 Cache a pointer to the old proc (as well as negative cache) to make
computing the io statistics over and over not as expensive.
This is a bit of a cop out, as I should just allocate a struct with
the computed values, but this will do for now.
2004-07-12 04:55:07 +00:00
Giorgos Keramidas
26b2243ae7 Build upon the nice work of Alfred and add sorting capabilities to
the -m "io" mode of top.

Approved by:	alfred
2004-07-08 16:45:55 +00:00
Dag-Erling Smørgrav
e38035a512 Additional preemptive unsigned -> signed casts. 2004-07-05 14:55:58 +00:00
Dag-Erling Smørgrav
f92a11552e ki_pctcpu is unsigned, so we cast to long before subtracting. 2004-07-05 14:52:06 +00:00
Dag-Erling Smørgrav
4946a00b0b Unhealthy amount of manual code cleanup. Some long lines still remain. 2004-07-05 14:45:57 +00:00
Dag-Erling Smørgrav
ea9e70bf68 Apply consistent indentation. Long lines will be fixed in a separate
commit.
2004-07-05 13:12:16 +00:00
Dag-Erling Smørgrav
f6a10fea63 Protoize. 2004-07-05 13:03:35 +00:00
Dag-Erling Smørgrav
cd23263cdd Sort out the #include mess. In particular, do not #include "os.h", since
all it does is provide broken prototypes for standard library functions.
2004-07-05 12:58:47 +00:00
Dag-Erling Smørgrav
1ce57ca079 Mechanical whitespace cleanup. 2004-07-05 12:51:35 +00:00
Dag-Erling Smørgrav
d62a717dae My previous commit fixed uid filtering, but broke io mode. Unravel the
process filtering logic to prevent this from happening again.
2004-07-05 12:48:17 +00:00
Dag-Erling Smørgrav
a578eed9af Fix selecting processes by uid, which was broken in the previous commit. 2004-07-05 12:22:50 +00:00
Alfred Perlstein
7fe9a86a5a Tell what percentage of the total IO a process is doing.
TODO: Show system totals.
2004-07-01 15:49:28 +00:00
Alfred Perlstein
db6bb7fc01 New feature, provide a display that shows the amount of IO processes
are doing.  Toggle this mode by hitting "m" or passing the command line
option "-m io" to top(1).  This allows one to identify disk bandwidth
hogs much easier.
2004-07-01 09:12:38 +00:00
Alfred Perlstein
970636e94f style, remove register. 2004-06-30 04:19:23 +00:00
Brian Feldman
cac4b14c12 Do not use KERN_PROC_PROC with kvm_getproc(3); instead, if only process
(and not thread) scope is to be displayed, use KERN_PROC_ALL and
accrue CPU% ourselves, as the kernel makes no attempt to do so.

Of course, this doesn't make most stats any less bogus when displaying
threaded processes, but at least the CPU time is added up and not just
always 0.00%.  There are still issues with SCHED_ULE in top(1) that
cause other processes to display 0.00% CPU when they in fact have used
more.
2004-06-06 19:59:06 +00:00
Julian Elischer
e27d79bb28 Changes to allow top to decide whether or not to show multiple threads per
process. Option -H enables it and it is toggled at the interactive
screen by 'H'.

Submitted by:	Jung-uk Kim <jkim@niksun.com>
2003-07-17 23:56:40 +00:00
Poul-Henning Kamp
f341ca9891 Remove #include <sys/dkstat.h> 2003-02-16 14:13:23 +00:00
John Baldwin
0d6326492b Catch up to SMTX -> SLOCK changes. 2002-10-02 20:33:52 +00:00
Philippe Charnier
93b0017f88 Replace various spelling with FALLTHROUGH which is lint()able 2002-08-25 13:23:09 +00:00
Warner Losh
3f330d7d1a remove __P 2002-03-22 01:42:45 +00:00
Mark Murray
e35f9517d6 Remove NO_WERRORs and WARNS=n's. To be revisited after GCC3. 2002-02-08 23:07:37 +00:00
David Malone
24bd57cbbd Add new option HAVE_STRERROR to CFLAGS. 2002-01-24 17:59:39 +00:00
Mark Murray
19964b3233 Protect this against the coming WARNS=2 default. 2001-12-12 00:05:16 +00:00
Bruce Evans
6eabd84580 Compensate for "Compensate for header dethreading" by backing it out. 2001-10-10 17:48:44 +00:00
Ruslan Ermilov
7d20bafcad Generate top.local.h and manpage dynamically, with sed(1) script.
The generated manpage will now describe the actual behavior of
top(1) WRT how many processes it displays.  This also eliminates
all troff(1) warnings.
2001-07-12 15:07:30 +00:00
Thomas Moestl
e0038fe587 Cast size_t variables before printing them to cope with the case of
sizeof(size_t) != sizeof(int).
2001-05-31 22:36:51 +00:00
Poul-Henning Kamp
719bfad4aa We don't need to include <sys/conf.h> 2001-05-23 23:01:37 +00:00
Mark Murray
6e547eabb8 Compensate for header dethreading. 2001-05-01 08:46:02 +00:00
John Baldwin
2522796e9d Check the new kern.smp.active sysctl rather than the non-existent
smp.smp_active sysctl to determine if we are running on an SMP machine.
2001-04-27 19:33:50 +00:00
Thomas Moestl
6243f7aca4 Correct the top memory stats display: convert page counts to kB.
Submitted by:	Andrea Campi <andrea@webcom.it>
Approved by:	rwatson
Pointy hat to:	myself
2001-03-20 16:02:16 +00:00
Will Andrews
9e13809564 Fix top(1) display for SMP systems where the username is longer than 14
characters.  This should avoid unattractive wrapping for people who are
stuck in an 80x24 screen.  :-)

PR:		22270
Submitted by:	William Carrel <williamc@go2net.com>
2001-03-12 05:53:54 +00:00
Andrew Gallatin
98e53e5bfb - An array of 4 32-bit ints for avenrun doesn't work on the alpha because
fscale is a (64-bit) long.  So just use a struct loadavg.
This fixes the recent failure of top on alphas:
	top: sysctl(vm.loadavg...) failed: Cannot allocate memory

- use size_t for sizeof() so as to fix a few int/long warnings on alpha

Reviewed by: Thomas Moestl <tmoestl@gmx.net>
2001-02-27 17:11:19 +00:00
Robert Watson
0ce14ceb84 Remove directive to install top as group kmem. Remove directive to
install top with setgid bit.
2001-02-23 18:55:43 +00:00
Robert Watson
39d513ecdd Adapt the top utility to not use kmem_read to retrieve variables now
available via sysctl().  As a result, top should now be able to run without
setgid kmem.

Submitted by:	Thomas Moestl <tmoestl@gmx.net>
Reviewed by:	freebsd-audit
2001-02-23 18:52:37 +00:00
Mark Murray
a28d666c49 Adjust columns for wide nicenesses in 'top -S'. 2001-02-18 11:08:47 +00:00
Jake Burkholder
4c85452ba9 Catch up to new priority interface. 2001-02-12 00:21:38 +00:00
John Baldwin
0821b7ca3a Catch up to proc flag change. 2001-01-24 14:17:23 +00:00
Kirk McKusick
1f7d250182 Change the proc information returned from the kernel so that it
no longer contains kernel specific data structures, but rather
only scalar values and structures that are already part of the
kernel/user interface, specifically rusage and rtprio. It no
longer contains proc, session, pcred, ucred, procsig, vmspace,
pstats, mtx, sigiolst, klist, callout, pasleep, or mdproc. If
any of these changed in size, ps, w, fstat, gcore, systat, and
top would all stop working. The new structure has over 200 bytes
of unassigned space for future values to be added, yet is nearly
100 bytes smaller per entry than the structure that it replaced.
2000-12-12 07:25:57 +00:00
John Baldwin
7f18d5d343 Make use of the full screen width to display p_comm rather than assuming a
hardcoded screen width of 80 chars.
2000-11-29 23:03:02 +00:00
John Baldwin
089f9b7e2f Display the name of the mutex we are blocked on in the state field. To
differentiate mutex names from wait channel names, prefix mutex names with
an asterisk.

Submitted by:	Dan Nelson <dnelson@emsphone.com>
2000-11-29 20:22:34 +00:00
Jason Evans
0384fff8c5 Major update to the way synchronization is done in the kernel. Highlights
include:

* Mutual exclusion is used instead of spl*().  See mutex(9).  (Note: The
  alpha port is still in transition and currently uses both.)

* Per-CPU idle processes.

* Interrupts are run in their own separate kernel threads and can be
  preempted (i386 only).

Partially contributed by:	BSDi (BSD/OS)
Submissions by (at least):	cp, dfr, dillon, grog, jake, jhb, sheldonh
2000-09-07 01:33:02 +00:00
David Nugent
dc7c1750c2 Reduce max length of user names to 15 characters to reflect reality.
This also prevents the line-wrap and messed up display that occurs
when there happens to be one or more names with 15 chars.
1999-11-17 16:31:51 +00:00
Bruce Evans
a8224b82fd Fixed sorting on time. On i386's, time differences of more than 2147
seconds caused overflow.  Use a type-safe but slightly slower comparison.
Comparisons for other fields are still fragile.

Fixed rounding of cputime (don't do extra work to get it slightly wrong
by first converting without rounding to milliseconds).

Removed dead code for setting cputime.

Fixed comments about cputime.
1999-11-17 03:25:54 +00:00
Peter Wemm
c3aac50f28 $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00
Nick Hibma
33d56839e3 Man pages for top refer to only 2.x, but not to 3.x and later 1999-06-14 12:06:11 +00:00
Peter Wemm
bf0e1ee370 Tidy up references to <sys/rlist.h> and support for the old swap management
that went away in January.
1999-05-11 14:32:18 +00:00
John Hay
e7f807a217 Fix the display of the "nice" value of processes like ntpd that use the
posix sched_setscheduler() to set their priority.

Noticed by:	Mark Allwright <mallwri@orion.didata.co.za>
1999-04-22 14:34:53 +00:00