Commit Graph

110 Commits

Author SHA1 Message Date
Ruslan Ermilov
a89237ae95 Sync program's usage() with manpage's SYNOPSIS. 2005-02-09 17:37:39 +00:00
Christian S.J. Peron
d8654c644a Use statfs instead of getmntinfo(). This will make the procfs checks
play nicer in prisons. It also simplifies things.

Reviewed by:	rwatson
Bumped into by:	Jilles Tjoelker
2004-11-19 16:11:54 +00:00
Christian S.J. Peron
a951d1f83b Currently if the user specifies -e and procfs is not mounted on /proc,
printing of the process environment will fail silently.

-define a function which will check to see if procfs is mounted on /proc
-Implement this test if the user specified -e
-If procfs is not mounted on /proc and -e was specified, print a warning.
 informing the user that procfs(5) is required.

Reviewed by:	wes, rwatson
2004-11-14 19:30:02 +00:00
Ruslan Ermilov
832b0b55ac Removed bitrot. 2004-11-13 17:12:22 +00:00
Garance A Drosehn
f2fbfae6b0 Improve checking for `ps -t <dev>', and give better error messages when
an invalid <dev> is specified.  Aside:  It turns out that the S_ISCHR()
check is true for almost every device that we have (not just tty's).
2004-06-27 22:56:58 +00:00
Garance A Drosehn
bd6233fd5d Fix a bug I introduced by some last-minute changes in -r 1.102. I ended
up checking the wrong variable for NULL.

Submitted by:	bde
2004-06-24 03:15:18 +00:00
Garance A Drosehn
23b8efad7a Rework the logic for `-t <tty>', such that it accepts "ttyp0" and "console",
in addition to "/dev/ttyp0" or "p0"  and  "/dev/console" or "co".
2004-06-24 01:57:59 +00:00
Garance A Drosehn
bdf8ab4692 Change "struct varent" to use the standard queue(8) macros, instead of
using it's own version of the same basic algorithm.

Submitted by:	part by Cyrille Lefevre, part of it done by me
2004-06-23 23:48:09 +00:00
Garance A Drosehn
dd693acf27 Add a check for defunct processes in saveuser(), so the output for "args"
(aka "command") will display "<defunct>", as does the output from "comm"
for those processes.  Also do better checking for malloc() failures.

Submitted by:	Cyrille Lefevre
2004-06-23 21:17:25 +00:00
Garance A Drosehn
31172f6e1b Include the `-c' option in the usage() message.
Submitted by:	Cyrille Lefevre
2004-06-23 11:40:56 +00:00
Garance A Drosehn
b4b243242b In the sorting routine, sort by thread-id if two processes have the same PID.
Submitted by:	Cyrille Lefevre
2004-06-23 11:28:17 +00:00
Garance A Drosehn
4bac4483db Have the main() routine calculate %CPU and (if needed) memory information
when copying per-process info before starting to sort the list.  This way,
sort-by-CPU or sort-by-memory will only calculate values once-per-process,
instead of twice-per-comparison.  Also take advantage of this to simplify
the pscomp() routine.
2004-06-20 21:25:10 +00:00
Garance A Drosehn
ecbb06f865 Try to change the isdigitch() macro to something that Bruce won't roll
his eyes at quite so much...  (actually someone else pointed this out
to me a long time ago, but apparently I never fixed it)
2004-06-01 23:27:11 +00:00
Garance A Drosehn
c23b00b7ca A few more style-fixes from Bruce. The only non-cosmetic change
is to drop a call to setuid() which has not been needed for years.

Noticed by:	bde
2004-06-01 22:19:16 +00:00
Garance A Drosehn
a9626fb3b4 Since I'm not ready to add the non-standard ADD_PS_LISTRESET feature,
remove the #ifdef for it for now.  I might add the feature for real at
some later date, there isn't much reason for the #ifdef for now.
2004-06-01 19:00:42 +00:00
Garance A Drosehn
251130839b Make a few style-istic improvements to the previous commits.
Noticed by:	bde
2004-06-01 18:12:04 +00:00
Garance A Drosehn
de1702f459 Fix so `ps' catches and complains about null-values specified for a
process id, instead of using pid==0.  Ie, `ps -p 12,' and `ps -p ,12'
are now errors (instead of being treated like `ps -p 0 -p 12').

Noticed by:	Cyrille Lefevre on freebsd-arch
2004-06-01 03:01:51 +00:00
Garance A Drosehn
bf46a3bf10 Additional tiny adjustment to kludge-option processing so `ps t p0'
is treated like `ps -t p0', instead of changing it to `ps -T p0'.
Note that `ps t' is still changed to `ps -T', since that is one of
the main reasons for this kludge processing...

Noticed by:	Jilles Tjoelker on freebsd-arch
2004-06-01 02:31:44 +00:00
Garance A Drosehn
c675340a5c Rewrite the kludge-option processing to improve how it handles a few
more special situations.  This is the code which process `ps blah',
when "blah" does not include a leading '-'.

This change also removes a long-undocumented BACKWARD_COMPATIBILITY
compile-time option, where:
     ps -options arg1 arg2
(with no '-' on "arg1" and "arg2") was treated as:
    ps -options -N arg1 -M arg2

This also changes `ps' to check for any additional arguments after
processing all the '-'-options, and attempt to use those arguments as
a pid or pidlist.  If an extra argument is not a valid pidlist, then
`ps' will print an error and exit.  This seems a more generally useful
extension of the kludge-option processing than the -N/-M behavior, and
has fewer confusing side-effects.

Reviewed by:	freebsd-arch
2004-06-01 02:03:21 +00:00
Garance A Drosehn
259fcfacc4 Add the 'sid' info to the output of `ps -j', to make up for the 'sess'
(session-pointer) info which was dropped from `ps' earlier in 5.x.

PR:		bin/59423
Submitted by:	Jilles Tjoelker
2004-05-23 21:35:35 +00:00
Garance A Drosehn
2631c7774f Fix the kludge-old-options processing so `ps tpt' will be treated the
same as `ps -tpt', instead of being changed into `ps -tpT'.

PR:		bin/52489
Submitted by:	Jilles Tjoelker
MFC after:	1 week
2004-05-23 21:21:07 +00:00
Garance A Drosehn
7bd421650b Change `ps' to use the KERN_PROC_RGID and KERN_PROC_SESSION options
(if trying to match only one real-group or one session-id), now that
those options are implemented in src/sys/kern/kern_proc.c (v1.203).

PR:		bin/65803  (a very tiny piece of the PR)
Submitted by:	Cyrille Lefevre
2004-05-22 23:13:58 +00:00
Mark Murray
6195fb4102 Remove clause 3 from the UCB licenses.
OK'ed by:	imp, core
2004-04-06 20:06:54 +00:00
Garance A Drosehn
b193355dc4 Drop the include of <stdint.h>, since r1.84 removed references of intmax_t. 2004-04-04 17:43:48 +00:00
Garance A Drosehn
9bf69950e1 Add back the `-e' option, which was mistakenly dropped when cleaning
up the PS_ARGS string in revision 1.69 (and which was apparently not
missed by anyone...).
2004-04-04 17:15:44 +00:00
Garance A Drosehn
d822163f3b Give a name of 'l' (list) to the union in struct listinfo. This is
because some compilers (such as gcc 2.95.4) do not support having an
unnamed union for a field in a struct.
2004-04-04 04:41:51 +00:00
Garance A Drosehn
0b42be7cb7 Switch to using strtoul() for parsing a potential UID or GID, which gets
this to correctly handle UID's and GID's larger than 2147483647.

Noticed by:	bde
MFC after:	1 week
2004-03-30 04:20:33 +00:00
Garance A Drosehn
f35e071573 Bruce would really like the prototype for fmt() to be split across lines
this way (although I still think it "looks weird"...).

Requested by:	bde
MFC after:	1 week
2004-03-30 02:02:40 +00:00
Garance A Drosehn
ba50b0e03a Minor style fixes, mostly adding indent-protection on some comment-blocks.
Noticed by:	bde
MFC after:	1 week
2004-03-30 01:59:22 +00:00
Garance A Drosehn
5bd7b1f3c0 Replace pscomp() with a cleaner version, mostly written by bde (*).
This corrects a problem of lost-precision for `-r' (sort-by-CPU).  Also,
for sort-by-CPU and sort-by-memory, any processes which have the same
value CPU or MEMORY are now sorted by TTY and then (if needed) by pid.

(* - I just added the NODEV checks, after doing some testing of my own)

Submitted by:	bde
MFC after:	1 week
2004-03-30 01:45:23 +00:00
Garance A Drosehn
c46bb4b3ff Oops. Remove some ';'s in #defines added by a previous update.
Noticed by:	bde
2004-03-29 03:03:28 +00:00
Garance A Drosehn
c7910c3ae3 Have this source explicitly include <sys/proc.h>, since it references
values such as P_CONTROLT and PS_INMEM.  But this still won't define
PID_MAX for us, since that is hidden inside of '#ifdef _KERNEL'.

Noticed by:	bde
2004-03-29 01:27:13 +00:00
Garance A Drosehn
87e47750f2 Since "kp" is a pointer, I should be comparing against NULL not 0.
Noticed by:	bde
2004-03-29 01:18:13 +00:00
Garance A Drosehn
352b652398 Various style improvements, mostly in comments and indentation.
Suggested by:	bde (well, for most of them)
2004-03-29 01:15:27 +00:00
Garance A Drosehn
ca62e195dd In the routines I've been working on, sort the variable declartions
so that non-pointers are listed after pointer-type variables.

Noticed by:	bde
2004-03-29 00:25:09 +00:00
Garance A Drosehn
1d2324f442 Split two 'fmt' strings so they're easier to read on 80-char windows. 2004-03-29 00:16:19 +00:00
Garance A Drosehn
de800cd4f8 Sort the declarations of global variables. 2004-03-29 00:12:03 +00:00
Garance A Drosehn
4857f240e1 Sort the routine prototypes. 2004-03-28 23:44:29 +00:00
Garance A Drosehn
5dfd7724a3 If <x> is a process id that does not exist, then just print the header
(if any) and exit, thus matching the behavior on -stable and other OS's.
My earlier attempt to fix this (v1.65) only seemed to work because of a
lucky random value in nentries (which was not being initialized back
when I tested that earlier patch).
2004-03-28 03:00:56 +00:00
Garance A Drosehn
e3c4e1dd77 If a non-existent user is given as part of `-U userlist', treat it as
a fatal error instead of a minor warning.  It is possible that a few
users are used to the previous behavior, but I'm claiming it was a bug.
2004-03-27 22:14:42 +00:00
Garance A Drosehn
29d58b106b Explicitly wrap two long-ish linesi of code, to make them easier to read. 2004-03-27 21:59:54 +00:00
Garance A Drosehn
b34f38ae9e Change the #if-ish logic which is used to add the -f' option when ps'
is compiled with LAZY_PS, so that there is only one PS_ARGS string to
modify when changing the option-list.  Also get `-f' to show up in the
usage() statement when compiled with LAZY_PS.
2004-03-27 21:56:49 +00:00
Garance A Drosehn
4a355d176d Move the 'f' case so it shows up in the right place, alphabetically. 2004-03-27 21:40:04 +00:00
Garance A Drosehn
a4c8a745a8 Support more POSIX/SUSv3 options:
- Change `-p' to allow a list of process IDs, and `-t' to allow a list
  of terminal names, instead of only a single value for each.
- Add the `-A' option of SUSv3, which is exactly the same as `-ax'.
- Add the `-G gidlist' (group id).
- Allow any of these "selector options" to be specified multiple times,
  and have `ps' keep adding to a given list -- instead of replacing the
  previously-specified values.
- Fix interactions between selector-options, so that: "If any are
  specified, ... ps shall select the processes represented by the
  inclusive OR of all the selection-criteria options." (from SUSv3)
- Add a `-X' option, which is the reverse of the `-x' option.

- various minor improvements in parsing and error handling.

This does not get us to match POSIX/SUSv3, but it gets us closer.  The
`-g pgidlist', `-R ruserlist' and `-s sidlist' options mentioned in
freebsd-standards are still under debate, so they skipped for now.
It should be true that this introduces no user-visible incompatible
changes, except to support "new stuff" that was not supported before.
2004-03-27 18:22:17 +00:00
Garance A Drosehn
2e6c6ac4d7 Fix 'ps -p proclist' and 'ps -u userlist' so the command returns non-zero
if no processes were matched.  Also sorts the list of 'int's in main, as
long as I had to add another one...

Noticed by:	Nate Lawson
MFC after:	10 days
2004-03-18 01:28:23 +00:00
Garance A Drosehn
4e8b6a6fb0 Improvements to 'ps -p <x>'. If <x> is a process id that does not exist, then
just print the header (if any) and exit, thus matching the behavior on -stable
and other OS's.

Also adds support for <x> being a comma-separated list of processes, and does
a much better checking for invalid-values of <x>, such as 'ps -p someword'.

Reviewed by:	mentioned on freebsd-current
MFC after:	10 days
2004-03-17 22:46:58 +00:00
Daniel Eischen
d75c1d8381 Allow the -H option to show threads when selecting by uid, tty,
and pid.
2004-02-22 17:59:31 +00:00
Scott Long
48b8c0de5a Add the -H option to ps(1) to display all kernel visible threads in each
process.  The default behavior of showing only the process is retained as
the default.
2003-06-12 16:53:55 +00:00
Philippe Charnier
eaed5652bf Display elapsed time (-o etime) using [[dd-]hh:]mm:ss, which according to
Solaris man page is the POSIX way.

Reviewed by:	jmallett
2003-02-05 13:18:17 +00:00
Juli Mallett
78b1878a16 Per-variable\ entry headers, to allow the 'ps -otime -otime=FOO' or similar
case to do the right thing and affect exactly one column.  This is consistent
with GNU ps(1) in BSD mode, and POLA.
2003-01-19 00:31:16 +00:00