Commit Graph

973 Commits

Author SHA1 Message Date
Jim Warner
02508b3d76 top: use CPU_ZEROTICS to embrace the Nehalem logic
The recently added logic dealing with "missing" tics
is mutually exclusive with logic associated with a
define called CPU_ZEROTICS.

This commit expands the use of that define to exclude
such Nehalem logic as appropriate.

It also extends programmer notes in top.h to include
an attribution for initiating the topic of potential
Nehalem type % CPU distortions, acknowledging:
  Jaromir Capik, <jcapik@redhat.com>

References
commit ce1410a51a
commit 9e7dd43ab7
commit a9041a5526
2012-02-12 21:52:23 +11:00
James Cloos
95f2201730 top: postpone the switch from Kb to Mb in summary display
The mem and swap lines have enough room to show eight significant
digits, so switch to showing MB when >=100MB, not >=10MB.

The extra detail is valuable; it should not be elided.

Signed-off-by: James Cloos <cloos@jhcloos.com>
Signed-off-by: Jim Warner <james.warner@comcast.net>
2012-02-12 21:52:23 +11:00
Sami Kerola
79d163568b libprocps: remove linux 2.7.0 check
Checking for linux kernel version 2.7.0 is meaningless now when 3.0.0
is out.

Reported-By: Christian Hofstaedtler
Bug-Debian: http://bugs.debian.org/635553
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:44:29 +01:00
Sami Kerola
529fa2e615 watch: inform about optional argument for -d switch
The -d, --differences switch(s) can use optional argument that was
not documented earlier.

Reported-By: Marian Sigler <m@qjym.de>
Bug-Debian: http://bugs.debian.org/597021
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:43:28 +01:00
Sami Kerola
1af18c260a pkill: remove -v match inversion option
The option -v does not make much sense in pkill context.

Reported-By: "Jason A. Spiro" <jasonspiro4@gmail.com>
Bug-Debian: http://bugs.debian.org/558044
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:43:13 +01:00
Sami Kerola
a1f8166642 pgrep: allow signal definition to be anywhere in command line
I as user often try to pkill something, notice that program did not
die and pkill again with signal -9.  Before this commit previous one
could not add signal at the end of command line, as the signal
definition had to be first argument, which was annoying.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:42:42 +01:00
Sami Kerola
e756a3db40 pgrep: use c style comments
The c++ comments e.g.  // should not be used in c code.  Additionally
white spaces are converted to tabs as a small clean up.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:42:40 +01:00
Sami Kerola
4ffc7c4f20 strutils: reindent the file
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:42:39 +01:00
Sami Kerola
ef4a6bef5f skill: fix argument parsing regression
Command skill accepted earlier command name as argument.  This did
not work since commit 991b8a94f0.

Reported-By: "Anne C. Hanna" <orion@ugcs.caltech.edu>
Bug-Debian: http://bugs.debian.org/659038
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:42:35 +01:00
Sami Kerola
4d8f9522a7 tests: fix dejagnu pwdx and vmstat checks
The pwdx check expected error message, which where in use for only
short period of time.  The old message was put in place in commit
9d47cb0c38

The vmstat issue was triggered by 'sr0' cdrom device, which gives
following unexpected output.  Fix simply ignores partitions which has
zero activity.  Besides such partitions probably would not apply as
good candidate of -p option anyway.

$ vmstat -p sr0
Partition was not found

Reported-By: Moritz Muehlenhoff <jmm@debian.org>
Bug-Debian: http://bugs.debian.org/656508
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-11 21:42:21 +01:00
Jim Warner
ce1410a51a top: fixup logic for Nehalem cpu 'idle' threshold
Due to a poorly constructed temporary fprintf
used during development, an earlier commit went
a little too far in its computations.  The net
result was the code looked nice but actually
accomplished nothing.

It is the /proc/stat line 1 (summary line)
whose tics must be used in establishing the
threshold boundary.  And that calculation
need be performed just once per frame.

This commit ensures one threshold calculation
per delay interval no matter how many cpus
are ultimately displayed.

It also corrects scalability by factoring in
the total number of online processors.

Reference:
commit 9e7dd43ab7
2012-02-10 07:01:10 +11:00
Jim Warner
9349100dd3 top: reduce use of 'clear_screen' to absolute minimum
Along the way to width override support (-w switch),
this top began clearing the screen far more often
than his predecessor.  In fact, it happend with each
user keystroke.

This commit dramatically reduces those occurances.
The screen will now be cleared only when an actual
SIGWINCH is received.

Thanks for identifying this flaw belongs to:
James Cloos, cloos@jhcloos.com

References:
http://www.freelists.org/post/procps/Merge-request

Blame:   c2dcbef482
Author:  Jim Warner <james.warner@comcast.net>
Date:    Thu May 26 11:33:32 2011 +0200
subject: added output width/height override support to top, + misc
2012-02-08 21:00:18 +11:00
Jim Warner
9e7dd43ab7 top: tweak the new 'reduce % CPU distortions' algorithm
The original approach to potential % CPU distortion due
to Nehalem type cores being turned off completely when
idle worked ok until the user typed something.

At that point, elapsed tics would no longer equal the
calculated value producing an undesirable 100% idle
condition until the next update or <Enter/Space> key.

This commit employs actual elapsed tics in determining
whether a cpu should be considered idle and thus makes
top's individual cpu display immune to user keystrokes.
2012-02-08 21:00:00 +11:00
Jim Warner
a9041a5526 top: reduce % CPU distortions when a core was turned off
This patch provides for cpu cores which can be turned
off completely when idle (Nehalem, etc.) thus registering
very few or no tics since the last update cycle.

When CPU_ZEROTICS is not defined (the default), any
displayed cpu with less than a certain amount of total
tics will show as 100% idle.  That amount is tempered
by the delay interval and total number of cpus.

This commit also satisfies the Debian 'top_nohz' patch
(11/24/09) in a slightly more efficient manner.  That
patch concerned kernels built with CONFIG_NO_HZ.

Reference:
http://www.freelists.org/post/procps/CStates-handling-new-switch,4
2012-02-08 20:57:58 +11:00
Jim Warner
f348575edc top: tolerate loss of smp cpus, improve CPU_t management
Prior to this patch, top was able to handle any hotplugged
cpus *added* to the system in two distinct ways.

 1) Newly added cpus would be detected by sysinfo_refresh
    calling the library's cpuinfo function, which occurs
    at most every 5 minutes.

 2) The user could force a refresh using either the
    <Enter> or <Space> keys.

Unfortunately, the *loss* of a cpu would produce an early
exit due to a /proc/stat read failure.  Such a failure
can be produced in the following way:
  sudo echo 0 > /sys/devices/system/cpu/cpu??/online

This commit allows top to tolerate the loss of cpus.
It also provides for more efficient CPU_t management,
especially for massively parallel cpu environments.

Note: Changes to the cpu compliment can produce a single
cycle distortion of cpu percentages.  Such distortion is
most visible when each cpu is being displayed.  It can
be eliminated with a forced refresh via <Enter>/<Space>.
2012-02-08 20:57:51 +11:00
Jim Warner
d7c986cd30 top: treat additional keys like the find_string() keys
With the introduction of the 'locate string' provisions,
the precedent for calling tertiary helper functions from
secondary do_key helper functions was established.

This commit simply migrates some additional keys out of
the do_key function itself and into the more generalized
key table.
2012-02-08 20:57:44 +11:00
Jim Warner
0c6aa6af41 top: improve USE_X_COLHDR effect when colors toggled off
Normally, when the chosen sort column is displayed via the
'x' command toggle the entire column is highlighted.  And
while this version of top substantially reduced the cost
of such highlighting, a small pathlength increase remained.

The USE_X_COLHDR define was an experimental alternative which
eliminated all recurring runtime costs for such emphasis by
highlighting the column header, not the entire column.

The previous implementation required colors to be turned on
(the 'z' toggle) for such highlighting to be visible.  This
commit extends column header emphasis to include monochrome
displays as well.

Reference:
http://www.freelists.org/post/procps/post-nls-merge,6
2012-02-08 20:57:38 +11:00
Jim Warner
9573a6fe5b top: properly handle an improper $HOME environment variable
Since its inception, this top has improperly handled an
empty HOME environment variable.  Under those conditions
a path to the root directory would have been constructed.

That caused no real harm upon startup since the display
defaults would have been employed.  However, except for
root, it would have been impossible to save the rc file.

This commit keeps the promise made in the documentation.
2012-02-08 20:57:30 +11:00
Sami Kerola
bc6b99e8ff libprocps: open file should be closed, not freed
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-06 17:37:11 +01:00
Sami Kerola
30b3c4958c vmstat: open file should be closed, not freed
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-05 23:15:54 +01:00
Sami Kerola
cbb9e54587 pmap: ensure arg1 is not null, and set qualifers to match with usage [smatch scan]
pmap.c:436 main(59) error: we previously assumed 'arg1' could be null (see line 427)
pmap.c:136:3: warning: return discards 'const' qualifier from pointer target type [enabled by default]
pmap.c:142:4: warning: return discards 'const' qualifier from pointer target type [enabled by default]
pmap.c:150:2: warning: return discards 'const' qualifier from pointer target type [enabled by default]
pmap.c:300:8: warning: assignment discards 'const' qualifier from pointer target type [enabled by default]

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-05 21:43:36 +01:00
Sami Kerola
b2ebca4a60 pgrep: fix potential null derefences [smatch scan]
pgrep.c:137 split_list(19) error: potential null derefence 'list'.
pgrep.c:540 select_procs(106) error: potential null derefence 'list'.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-05 21:13:36 +01:00
Sami Kerola
1689e769e6 sysctl: fix potential null derefence [smatch scan]
sysctl.c:560 PreloadSystem(29) error: potential null derefence 'cfgs'.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-05 21:02:34 +01:00
Sami Kerola
8cd53e16b5 skill: ensure optarg is not null [smatch scan]
skill.c:549 skillsnice_parse(87) error: we previously assumed 'optarg' could be null (see line 539)

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-05 20:54:58 +01:00
Sami Kerola
ffe5e0b08e watch: remove unreachable code
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-05 20:48:33 +01:00
Sami Kerola
c686e4c4dc pwdx: free memory before exit
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-02-05 20:44:39 +01:00
Mike Frysinger
96e86ef1a0 use pkg-config for ncurses by default
Newer ncurses install pkg-config files, so search those first.  If they
aren't found, fall back to existing detection logic.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-26 21:17:00 +01:00
Mike Frysinger
061de22bad fix AC_ARG_WITH(ncurses) handling
The third arg is for "the user has specified some flag", not "the user
has disabled things", so use $withval.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-26 21:16:59 +01:00
Samuli Suominen
62c0cf67f6 fix basic ncurses check
The first check for ncurses is for the non-wide variant, so drop the "w".
The wide version gets checked later on based on watch8bit.

Signed-off-by: Samuli Suominen <ssuominen@gentoo.org>
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-26 21:16:57 +01:00
Mike Frysinger
0da6537c5f fix printf format warnings
pgrep.c: In function 'main':
pgrep.c:793:4: warning: format '%d' expects argument of type 'int', but argument 4 has type 'long int' [-Wformat]

vmstat.c: In function 'diskpartition_format':
vmstat.c:382:9: warning: format '%u' expects argument of type 'unsigned int', but argument 5 has type 'long long unsigned int' [-Wformat]
vmstat.c:408:10: warning: format '%u' expects argument of type 'unsigned int', but argument 5 has type 'long long unsigned int' [-Wformat]

w.c: In function 'main':
w.c:394:4: warning: format '%d' expects argument of type 'int', but argument 4 has type 'long unsigned int' [-Wformat]

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-26 21:15:42 +01:00
Mike Frysinger
fc3bf171ff pmap: fix const warning
pmap.c:300:8: warning: assignment discards 'const' qualifier
	from pointer target type [enabled by default]

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-26 21:12:39 +01:00
Kent R. Spillner
81f64657ba watch: exit when command output changes.
Add new flags to watch (-g/--chgexit) so that it exits when the
output changes.  This is useful in builds and shell scripts, for
example when deploying webapps to block the remainder of the
deployment steps until after the webapp starts.

e.g. watch --chgexit curl http://foo/bar
2012-01-26 08:42:37 -06:00
Sami Kerola
8fe81caa42 sysctl: fix coding style
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-25 09:18:38 +01:00
Sami Kerola
1cde286c75 docs: add to pgrep.1 count return value information
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-25 09:17:49 +01:00
Sami Kerola
da20fcbe42 docs: remove tcp_default_win_scale from sysctl.conf sample
The variable net.ipv4.tcp_default_win_scale has not been part of
kernel since 2.6.9 (Oct 2004).

Reference: http://oss.sgi.com/archives/netdev/2004-08/msg01589.html
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-24 22:34:12 +01:00
Sami Kerola
ea221b4e42 libprocps: hide private symbols
Clean the global list of symbols contain only functions and variables which
are being called from library.  The removed symbols are internal for the
library, which means they can be changed without changes to programs using
the library.  This clean up should help incoming libprocps clean up work.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-24 21:54:01 +01:00
Craig Small
8955ec26fe Merge remote-tracking branch 'sami/vapier' 2012-01-24 22:36:28 +11:00
Mike Frysinger
9c32fdbec3 ps: include error.h
This code currently uses error_at_line() from error.h, so pull it in.
Long term, this might get moved to c.h as a local helper on err.h,
but I have no idea.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-24 08:55:05 +01:00
Mike Frysinger
2b449855c7 proc: pull in ctype.h where needed
Fix the build warnings:

sig.c:227:5: warning: implicit declaration of function 'toupper' [-Wimplicit-function-declaration]
sig.c:231:3: warning: implicit declaration of function 'isdigit' [-Wimplicit-function-declaration]

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-24 08:55:05 +01:00
Mike Frysinger
afaf698896 use helpers from c.h
A few files still use the C library helpers from err.h, so migrate
them to the local c.h helpers.

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-24 08:55:05 +01:00
Mike Frysinger
13d8cc0681 watch: include sys/wait.h for waitpid
Fixes build warning:
	watch.c:682:3: warning: implicit declaration of
		function 'waitpid' [-Wimplicit-function-declaration]

Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-24 08:55:05 +01:00
Mike Frysinger
af61809a18 include error.h for error() and friends
Signed-off-by: Mike Frysinger <vapier@gentoo.org>
2012-01-24 08:55:05 +01:00
Sami Kerola
7fc80ffa42 skill: exclude debugging strings from nls
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-22 01:23:59 +01:00
Sami Kerola
2b99362139 tload: remove unnecessary goto
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-21 23:29:02 +01:00
Sami Kerola
8f7eefc9c8 docs: aligh top.1 header with other procps-ng utils
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-21 16:49:47 +01:00
Sami Kerola
624773957f docs: add exit values to watch.1
Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-21 16:41:31 +01:00
Sami Kerola
ae08eb8af6 tload: check write() return value
Exit at rare events when writing to tty is unsuccessful.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-21 16:24:33 +01:00
Sami Kerola
bc974b3c33 pkill: enable displaying what is killed
Add new command line options -e, --echo to display what is
killed. Cost of this change is greater run time memory footprint,
because an union had to be changed to struct to allow name and
pid printing which what I as an user would expect to see in
verbose kill output.

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-21 13:44:00 +01:00
Sami Kerola
12dce9f83b pgrep: fix compiler warning
pgrep.c:786:4: warning: format '%ld' expects argument of type 'long int', but argument 3 has type 'int' [-Wformat]

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-21 12:15:41 +01:00
Sami Kerola
3489b9343e pkill: fix error printing
Print correct pid number of process that was where not killed
because of kill was not successful. Incorrect behaviour before
this fix is bello.

-- snip
$ ps -elf | grep dh[c]
1 S root      1409     1  0  80   0 -  1782 poll_s 02:01 ?        00:00:00 dhclient wlan0
$ pkill dhc
pkill: killing pid 1 failed: Operation not permitted
-- snip

Signed-off-by: Sami Kerola <kerolasa@iki.fi>
2012-01-21 12:15:40 +01:00