|
HTML rendering created 2024-06-26 by Michael Kerrisk, author of The Linux Programming Interface. For details of in-depth Linux/UNIX system programming training courses that I teach, look here. Hosting by jambit GmbH. |
|
TOP(1) User Commands TOP(1)
top - display Linux processes
top [option ...]
The top program provides a dynamic real-time view of a running
system. It can display system summary information as well as a
list of processes or threads currently being managed by the Linux
kernel. The types of system summary information shown and the
types, order and size of information displayed for processes are
all user configurable and that configuration can be made
persistent across restarts.
The program provides a limited interactive interface for process
manipulation as well as a much more extensive interface for
personal configuration -- encompassing every aspect of its
operation. And while top is referred to throughout this
document, you are free to name the program anything you wish.
That new name, possibly an alias, will then be reflected on top's
display and used when reading and writing a configuration file.
Documentation
The remaining Table of Contents
OVERVIEW
Operation
Linux Memory Types
1. COMMAND-LINE Options
2. SUMMARY Display
a. UPTIME and LOAD Averages
b. TASK and CPU States
c. MEMORY Usage
3. FIELDS / Columns Display
a. DESCRIPTIONS of Fields
b. MANAGING Fields
4. INTERACTIVE Commands
a. GLOBAL Commands
b. SUMMARY AREA Commands
c. TASK AREA Commands
1. Appearance
2. Content
3. Size
4. Sorting
d. COLOR Mapping
5. ALTERNATE-DISPLAY Provisions
a. WINDOWS Overview
b. COMMANDS for Windows
c. SCROLLING a Window
d. SEARCHING in a Window
e. FILTERING in a Window
6. FILES
a. PERSONAL Configuration File
b. ADDING INSPECT Entries
c. SYSTEM Configuration File
d. SYSTEM Restrictions File
7. ENVIRONMENT VARIABLE(S)
8. STUPID TRICKS Sampler
a. Kernel Magic
b. Bouncing Windows
c. The Big Bird Window
d. The Ol' Switcheroo
9. BUGS, 10. SEE Also
Operation
When operating top, the two most important keys are the help (h
or ?) key and quit (‘q’) key. Alternatively, you could simply
use the traditional interrupt key (^C) when you're done.
When started for the first time, you'll be presented with these
traditional elements on the main top screen: 1) Summary Area; 2)
Fields/Columns Header; 3) Task Area. Each of these will be
explored in the sections that follow. There is also an
Input/Message line between the Summary Area and Columns Header
which needs no further explanation.
The main top screen is generally quite adaptive to changes in
terminal dimensions under X-Windows. Other top screens may be
less so, especially those with static text. It ultimately
depends, however, on your particular window manager and terminal
emulator. There may be occasions when their view of terminal
size and current contents differs from top's view, which is
always based on operating system calls.
Following any re-size operation, if a top screen is corrupted,
appears incomplete or disordered, simply typing something
innocuous like a punctuation character or cursor motion key will
usually restore it. In extreme cases, the following sequence
almost certainly will:
key/cmd objective
^Z suspend top
fg resume top
<Left> force a screen redraw (if necessary)
But if the display is still corrupted, there is one more step you
could try. Insert this command after top has been suspended but
before resuming it.
key/cmd objective
reset restore your terminal settings
Note: the width of top's display will be limited to 512
positions. Displaying all fields requires approximately 250
characters. Remaining screen width is usually allocated to any
variable width columns currently visible. The variable width
columns, such as COMMAND, are noted in topic 3a. DESCRIPTIONS of
Fields. Actual output width may also be influenced by the -w
switch, which is discussed in topic 1. COMMAND-LINE Options.
Lastly, some of top's screens or functions require the use of
cursor motion keys like the standard arrow keys plus the Home,
End, PgUp and PgDn keys. If your terminal or emulator does not
provide those keys, the following combinations are accepted as
alternatives:
key equivalent-keys
Left alt + h
Down alt + j
Up alt + k
Right alt + l
Home alt + ctrl + h
PgDn alt + ctrl + j
PgUp alt + ctrl + k
End alt + ctrl + l
The Up and Down arrow keys have special significance when
prompted for line input terminated with the <Enter> key. Those
keys, or their aliases, can be used to retrieve previous input
lines which can then be edited and re-input. And there are four
additional keys available with line oriented input.
key special-significance
Up recall older strings for re-editing
Down recall newer strings or erase entire line
Insert toggle between insert and overtype modes
Delete character removed at cursor, moving others left
Home jump to beginning of input line
End jump to end of input line
Linux Memory Types
For our purposes there are three types of memory, and one is
optional. First is physical memory, a limited resource where
code and data must reside when executed or referenced. Next is
the optional swap file, where modified (dirty) memory can be
saved and later retrieved if too many demands are made on
physical memory. Lastly we have virtual memory, a nearly
unlimited resource serving the following goals:
1. abstraction, free from physical memory addresses/limits
2. isolation, every process in a separate address space
3. sharing, a single mapping can serve multiple needs
4. flexibility, assign a virtual address to a file
Regardless of which of these forms memory may take, all are
managed as pages (typically 4096 bytes) but expressed by default
in top as KiB (kibibyte). The memory discussed under topic ‘2c.
MEMORY Usage’ deals with physical memory and the swap file for
the system as a whole. The memory reviewed in topic ‘3. FIELDS /
Columns Display’ embraces all three memory types, but for
individual processes.
For each such process, every memory page is restricted to a
single quadrant from the table below. Both physical memory and
virtual memory can include any of the four, while the swap file
only includes #1 through #3. The memory in quadrant #4, when
modified, acts as its own dedicated swap file.
Private | Shared
1 | 2
Anonymous . stack |
. malloc() |
. brk()/sbrk() | . POSIX shm*
. mmap(PRIVATE, ANON) | . mmap(SHARED, ANON)
-----------------------+----------------------
. mmap(PRIVATE, fd) | . mmap(SHARED, fd)
File-backed . pgms/shared libs |
3 | 4
The following may help in interpreting process level memory
values displayed as scalable columns and discussed under topic
‘3a. DESCRIPTIONS of Fields’.
%MEM - simply RES divided by total physical memory
CODE - the ‘pgms’ portion of quadrant 3
DATA - the entire quadrant 1 portion of VIRT plus all
explicit mmap file-backed pages of quadrant 3
RES - anything occupying physical memory which, beginning with
Linux-4.5, is the sum of the following three fields:
RSan - quadrant 1 pages, which include any
former quadrant 3 pages if modified
RSfd - quadrant 3 and quadrant 4 pages
RSsh - quadrant 2 pages
RSlk - subset of RES which cannot be swapped out (any quadrant)
SHR - subset of RES (excludes 1, includes all 2 & 4, some 3)
SWAP - potentially any quadrant except 4
USED - simply the sum of RES and SWAP
VIRT - everything in-use and/or reserved (all quadrants)
Note: Even though program images and shared libraries are
considered private to a process, they will be accounted for as
shared (SHR) by the kernel.
Mandatory arguments to long options are mandatory for short
options too.
Although not required, the equals sign can be used with either
option form and whitespace before and/or after the ‘=’ is
permitted.
-b, --batch
Starts top in Batch mode, which could be useful for sending
output from top to other programs or to a file. In this mode,
top will not accept input and runs until the iterations limit
you've set with the ‘-n’ command-line option or until killed.
-c, --cmdline-toggle
Starts top with the last remembered ‘c’ state reversed. Thus,
if top was displaying command lines, now that field will show
program names, and vice versa. See the ‘c’ interactive
command for additional information.
-d, --delay = SECS [.TENTHS]
Specifies the delay between screen updates, and overrides the
corresponding value in one's personal configuration file or
the startup default. Later this can be changed with the ‘d’
or ‘s’ interactive commands.
Fractional seconds are honored, but a negative number is not
allowed. In all cases, however, such changes are prohibited
if top is running in Secure mode, except for root (unless the
‘s’ command-line option was used). For additional information
on Secure mode see topic 6d. SYSTEM Restrictions File.
-E, --scale-summary-mem = k | m | g | t | p | e
Instructs top to force summary area memory to be scaled as:
k - kibibytes
m - mebibytes
g - gibibytes
t - tebibytes
p - pebibytes
e - exbibytes
Later this can be changed with the ‘E’ command toggle.
-e, --scale-task-mem = k | m | g | t | p
Instructs top to force task area memory to be scaled as:
k - kibibytes
m - mebibytes
g - gibibytes
t - tebibytes
p - pebibytes
Later this can be changed with the ‘e’ command toggle.
-H, --threads-show
Instructs top to display individual threads. Without this
command-line option a summation of all threads in each process
is shown. Later this can be changed with the ‘H’ interactive
command.
-h, --help
Display usage help text, then quit.
-i, --idle-toggle
Starts top with the last remembered ‘i’ state reversed. When
this toggle is Off, tasks that have not used any CPU since the
last update will not be displayed. For additional information
regarding this toggle see topic 4c. TASK AREA Commands, SIZE.
-n, --iterations = NUMBER
Specifies the maximum number of iterations, or frames, top
should produce before ending.
-O, --list-fields
This option acts as a form of help for the -o option shown
below. It will cause top to print each of the available field
names on a separate line, then quit. Such names are subject
to NLS (National Language Support) translation.
-o, --sort-override = FIELDNAME
Specifies the name of the field on which tasks will be sorted,
independent of what is reflected in the configuration file.
You can prepend a ‘+’ or ‘-’ to the field name to also
override the sort direction. A leading ‘+’ will force sorting
high to low, whereas a ‘-’ will ensure a low to high ordering.
This option exists primarily to support automated/scripted
batch mode operation.
-p, --pid = PIDLIST (as: 1,2,3, ... or
-p1 -p2 -p3 ...) Monitor only processes with specified
process IDs. However, when combined with Threads mode (‘H’),
all processes in the thread group (see TGID) of each monitored
PID will also be shown.
This option can be given up to 20 times, or you can provide a
comma delimited list with up to 20 pids. Co-mingling both
approaches is permitted.
A pid value of zero will be treated as the process id of the
top program itself once it is running.
This is a command-line option only and should you wish to
return to normal operation, it is not necessary to quit and
restart top -- just issue any of these interactive commands:
‘=’, ‘u’ or ‘U’.
The ‘p’, ‘u’ and ‘U’ command-line options are mutually
exclusive.
-S, --accum-time-toggle
Starts top with the last remembered ‘S’ state reversed. When
Cumulative time mode is On, each process is listed with the
cpu time that it and its dead children have used. See the ‘S’
interactive command for additional information regarding this
mode.
-s, --secure-mode
Starts top with secure mode forced, even for root. This mode
is far better controlled through a system configuration file
(see topic 6. FILES).
-U, --filter-any-user = USER (as: number or name)
Display only processes with a user id or user name matching
that given. This option matches on any user (real, effective,
saved, or filesystem).
Prepending an exclamation point (‘!’) to the user id or name
instructs top to display only processes with users not
matching the one provided.
The ‘p’, ‘U’ and ‘u’ command-line options are mutually
exclusive.
-u, --filter-only-euser = USER (as: number or name)
Display only processes with a user id or user name matching
that given. This option matches on the effective user id
only.
Prepending an exclamation point (‘!’) to the user id or name
instructs top to display only processes with users not
matching the one provided.
The ‘p’, ‘U’ and ‘u’ command-line options are mutually
exclusive.
-V, --version
Display version information, then quit.
-w, --width [=COLUMNS]
In Batch mode, when used without an argument top will format
output using the COLUMNS= and LINES= environment variables, if
set. Otherwise, width will be fixed at the maximum 512
columns. With an argument, output width can be decreased or
increased (up to 512) but the number of rows is considered
unlimited.
In normal display mode, when used without an argument top will
attempt to format output using the COLUMNS= and LINES=
environment variables, if set. With an argument, output width
can only be decreased, not increased. Whether using
environment variables or an argument with -w, when not in
Batch mode actual terminal dimensions can never be exceeded.
Note: Without the use of this command-line option, output
width is always based on the terminal at which top was invoked
whether or not in Batch mode.
-1, --single-cpu-toggle
Starts top with the last remembered Cpu States portion of the
summary area reversed. Either all cpu information will be
displayed in a single line or each cpu will be displayed
separately, depending on the state of the NUMA Node command
toggle (‘2’).
See the ‘1’ and ‘2’ interactive commands for additional
information.
Each of the following three areas are individually controlled
through one or more interactive commands. See topic 4b. SUMMARY
AREA Commands for additional information regarding these
provisions.
2a. UPTIME and LOAD Averages
This portion consists of a single line containing:
program or window name, depending on display mode
current time and length of time since last boot
total number of users
system load avg over the last 1, 5 and 15 minutes
2b. TASK and CPU States
This portion consists of a minimum of two lines. In an SMP
environment, additional lines can reflect individual CPU state
percentages.
Line 1 shows total tasks or threads, depending on the state of
the Threads-mode toggle. That total is further classified
according to task state as follows:
displayed process status (‘S’)
--------- --------------------
running R
sleep S + any remaining
d-sleep D
stopped T + t
zombie Z
Line 2 shows CPU state percentages based on the interval since
the last refresh.
As a default, percentages for these individual categories are
displayed. Depending on your kernel version, the st field may
not be shown.
us : time running un-niced user processes
sy : time running kernel processes
ni : time running niced user processes
id : time spent in the kernel idle handler
wa : time waiting for I/O completion
hi : time spent servicing hardware interrupts
si : time spent servicing software interrupts
st : time stolen from this vm by the hypervisor
The ‘sy’ value above also reflects the time running a virtual cpu
for guest operating systems, including those that have been
niced.
Beyond the first tasks/threads line, there are alternate CPU
display modes available via the 4-way ‘t’ command toggle. They
show an abbreviated summary consisting of these elements:
a b c d
%Cpu(s): 75.0/25.0 100[ ... ]
Where: a) is the ‘user’ (us + ni) percentage; b) is the ‘system’
(sy + hi + si + guests) percentage; c) is the total percentage;
and d) is one of two visual graphs of those representations.
Such graphs also reflect separate ‘user’ and ‘system’ portions.
If the ‘4’ command toggle is used to yield more than two cpus per
line, results will be further abridged eliminating the a) and b)
elements. However, that information is still reflected in the
graph itself assuming color is active or, if not, bars vs. blocks
are being shown.
See topic 4b. SUMMARY AREA Commands for additional information on
the ‘t’ and ‘4’ command toggles.
2c. MEMORY Usage
This portion consists of two lines which may express values in
kibibytes (KiB) through exbibytes (EiB) depending on the scaling
factor enforced with the ‘E’ interactive command. The
/proc/meminfo source fields are shown in parenthesis.
Line 1 reflects physical memory, classified as:
total ( MemTotal )
free ( MemFree )
used ( MemTotal - MemAvailable )
buff/cache ( Buffers + Cached + SReclaimable )
Line 2 reflects mostly virtual memory, classified as:
total ( SwapTotal )
free ( SwapFree )
used ( SwapTotal - SwapFree )
avail ( MemAvailable, which is physical memory )
The avail number on line 2 is an estimation of physical memory
available for starting new applications, without swapping.
Unlike the free field, it attempts to account for readily
reclaimable page cache and memory slabs. It is available on
kernels 3.14, emulated on kernels 2.6.27+, otherwise the same as
free.
In the alternate memory display modes, two abbreviated summary
lines are shown consisting of these elements:
a b c
GiB Mem : 18.7/15.738 [ ... ]
GiB Swap: 0.0/7.999 [ ... ]
Where: a) is the percentage used; b) is the total available; and
c) is one of two visual graphs of those representations.
In the case of physical memory, the percentage represents the
total minus the estimated avail noted above. The ‘Mem’ graph
itself is divided between the non-cached portion of used and any
remaining memory not otherwise accounted for by avail. See topic
4b. SUMMARY AREA Commands and the ‘m’ command for additional
information on that special 4-way toggle.
This table may help in interpreting the scaled values displayed:
KiB = kibibyte = 1024 bytes
MiB = mebibyte = 1024 KiB = 1,048,576 bytes
GiB = gibibyte = 1024 MiB = 1,073,741,824 bytes
TiB = tebibyte = 1024 GiB = 1,099,511,627,776 bytes
PiB = pebibyte = 1024 TiB = 1,125,899,906,842,624 bytes
EiB = exbibyte = 1024 PiB = 1,152,921,504,606,846,976 bytes
3a. DESCRIPTIONS of Fields
Listed below are top's available process fields (columns). They
are shown in strict ascii alphabetical order. You may customize
their position and whether or not they are displayable with the
‘f’ (Fields Management) interactive command.
Any field is selectable as the sort field, and you control
whether they are sorted high-to-low or low-to-high. For
additional information on sort provisions see topic 4c. TASK AREA
Commands, SORTING.
The fields related to physical memory or virtual memory reference
‘(KiB)’ which is the unsuffixed display mode. Such fields may,
however, be scaled from KiB through PiB. That scaling is
influenced via the ‘e’ interactive command or established for
startup through a build option.
%CPU -- CPU Usage
The task's share of the elapsed CPU time since the last
screen update, expressed as a percentage of total CPU time.
In a true SMP environment, if a process is multi-threaded and
top is not operating in Threads mode, amounts greater than
100% may be reported. You toggle Threads mode with the ‘H’
interactive command.
Also for multi-processor environments, if Irix mode is Off,
top will operate in Solaris mode where a task's cpu usage
will be divided by the total number of CPUs. You toggle
Irix/Solaris modes with the ‘I’ interactive command.
Note: When running in forest view mode (‘V’) with children
collapsed (‘v’), this field will also include the CPU time of
those unseen children. See topic 4c. TASK AREA Commands,
CONTENT for more information regarding the ‘V’ and ‘v’
toggles.
%CUC -- CPU Utilization
This field is identical to %CUU below, except the percentage
also reflects reaped child processes.
%CUU -- CPU Utilization
A task's total CPU usage divided by its elapsed running time,
expressed as a percentage.
If a process currently displays high CPU usage, this field
can help determine if such behavior is normal. Conversely,
if a process has low CPU usage currently, %CUU may reflect
historically higher demands over its lifetime.
%MEM -- Memory Usage (RES)
A task's currently resident share of available physical
memory.
See ‘OVERVIEW, Linux Memory Types’ for additional details.
AGID -- Autogroup Identifier
The autogroup identifier associated with a process. This
feature operates in conjunction with the CFS scheduler to
improve interactive desktop performance.
When /proc/sys/kernel/sched_autogroup_enabled is set, a new
autogroup is created with each new session (see SID). All
subsequently forked processes in that session inherit
membership in this autogroup. The kernel then attempts to
equalize distribution of CPU cycles across such groups.
Thus, an autogroup with many CPU intensive processes (e.g
make -j) will not dominate an autogroup with only one or two
processes.
When -1 is displayed it means this information is not
available.
AGNI -- Autogroup Nice Value
The autogroup nice value which affects scheduling of all
processes in that group. A negative nice value means higher
priority, whereas a positive nice value means lower priority.
CGNAME -- Control Group Name
The name of the control group to which a process belongs, or
‘-’ if not applicable for that process.
Available with cgroup v1 only, this will typically be the
last entry in the full list of control groups as displayed
under the CGROUPS heading. And as is true there, this field
is also variable width.
CGROUPS -- Control Groups
The names of the control group(s) to which a process belongs,
or ‘-’ if not applicable for that process.
Control Groups provide for allocating resources (cpu, memory,
network bandwidth, etc.) among installation-defined groups of
processes. They enable fine-grained control over allocating,
denying, prioritizing, managing and monitoring those
resources.
Many different hierarchies of cgroups can exist
simultaneously on a system and each hierarchy is attached to
one or more subsystems. A subsystem represents a single
resource.
Note: The CGROUPS field, unlike most columns, is not fixed-
width. When displayed, it plus any other variable width
columns will be allocated all remaining screen width (up to
the maximum 512 characters). Even so, such variable width
fields could still suffer truncation. See topic 5c.
SCROLLING a Window for additional information on accessing
any truncated data.
CLS -- Scheduling Class
The task's current scheduling policy which can be one of:
- = not reported
TS = SCHED_OTHER
FF = SCHED_FIFO
RR = SCHED_RR
B = SCHED_BATCH
ISO = SCHED_ISO
IDL = SCHED_IDLE
DLN = SCHED_DEADLINE
? = unknown value
CODE -- Code Size (KiB)
The amount of physical memory currently devoted to executable
code, also known as the Text Resident Set size or TRS.
See ‘OVERVIEW, Linux Memory Types’ for additional details.
COMMAND -- Command Name or Command Line
Display the command line used to start a task or the name of
the associated program. You toggle between command line and
name with ‘c’, which is both a command-line option and an
interactive command.
When you've chosen to display command lines, processes
without a command line (like kernel threads) will be shown
with only the program name in brackets, as in this example:
[kthreadd]
This field may also be impacted by the forest view display
mode. See the ‘V’ interactive command for additional
information regarding that mode.
Note: The COMMAND field, unlike most columns, is not fixed-
width. When displayed, it plus any other variable width
columns will be allocated all remaining screen width (up to
the maximum 512 characters). Even so, such variable width
fields could still suffer truncation. This is especially
true for this field when command lines are being displayed
(the ‘c’ interactive command.) See topic 5c. SCROLLING a
Window for additional information on accessing any truncated
data.
DATA -- Data + Stack Size (KiB)
The amount of private memory reserved by a process. It is
also known as the Data Resident Set or DRS. Such memory may
not yet be mapped to physical memory (RES) but will always be
included in the virtual memory (VIRT) amount.
See ‘OVERVIEW, Linux Memory Types’ for additional details.
Docker -- Docker Container ID
The identity (abbreviated hash) of the docker container
within which a task is running. If a process is not running
inside a container, a dash (‘-’) will be shown.
ELAPSED -- Elapsed Running Time
The length of time since a process was started. Thus, the
most recently started task will display the smallest time
interval.
The value will be expressed as ‘HH,MM’ (hours,minutes) but is
subject to additional scaling if the interval becomes too
great to fit column width. At that point it will be scaled
to ‘DD+HH’ (days+hours) and possibly beyond.
ENVIRON -- Environment variables
Display all of the environment variables, if any, as seen by
the respective processes. These variables will be displayed
in their raw native order, not the sorted order you are
accustomed to seeing with an unqualified ‘set’.
Note: The ENVIRON field, unlike most columns, is not fixed-
width. When displayed, it plus any other variable width
columns will be allocated all remaining screen width (up to
the maximum 512 characters). Even so, such variable width
fields could still suffer truncation. This is especially
true for this field. See topic 5c. SCROLLING a Window for
additional information on accessing any truncated data.
EXE -- Executable Path
Where available, this is the full path to the executable,
including the program name.
Note: The EXE field, unlike most columns, is not fixed-width.
When displayed, it plus any other variable width columns will
be allocated all remaining screen width (up to the maximum
512 characters).
Flags -- Task Flags
This column represents the task's current scheduling flags
which are expressed in hexadecimal notation and with zeros
suppressed. These flags are officially documented in
<linux/sched.h>.
GID -- Group Id
The effective group ID.
GROUP -- Group Name
The effective group name.
LOGID -- Login User Id
The user ID used at login. When -1 is displayed it means
this information is not available.
LXC -- Lxc Container Name
The name of the lxc container within which a task is running.
If a process is not running inside a container, a dash (‘-’)
will be shown.
NI -- Nice Value
The nice value of the task. A negative nice value means
higher priority, whereas a positive nice value means lower
priority. Zero in this field simply means priority will not
be adjusted in determining a task's dispatch-ability.
Note: This value only affects scheduling priority relative to
other processes in the same autogroup. See the ‘AGID’ and
‘AGNI’ fields for additional information on autogroups.
NU -- Last known NUMA node
A number representing the NUMA node associated with the last
used processor (‘P’). When -1 is displayed it means that
NUMA information is not available.
See the ‘2’ and ‘3’ interactive commands for additional NUMA
provisions affecting the summary area.
OOMa -- Out of Memory Adjustment Factor
The value, ranging from -1000 to +1000, added to the current
out of memory score (OOMs) which is then used to determine
which task to kill when memory is exhausted.
OOMs -- Out of Memory Score
The value, ranging from 0 to +1000, used to select task(s) to
kill when memory is exhausted. Zero translates to ‘never
kill’ whereas 1000 means ‘always kill’.
P -- Last used CPU (SMP)
A number representing the last used processor. In a true SMP
environment this will likely change frequently since the
kernel intentionally uses weak affinity. Also, the very act
of running top may break this weak affinity and cause more
processes to change CPUs more often (because of the extra
demand for cpu time).
PGRP -- Process Group Id
Every process is member of a unique process group which is
used for distribution of signals and by terminals to
arbitrate requests for their input and output. When a
process is created (forked), it becomes a member of the
process group of its parent. By convention, this value
equals the process ID (see PID) of the first member of a
process group, called the process group leader.
PID -- Process Id
The task's unique process ID, which periodically wraps,
though never restarting at zero. In kernel terms, it is a
dispatchable entity defined by a task_struct.
This value may also be used as: a process group ID (see
PGRP); a session ID for the session leader (see SID); a
thread group ID for the thread group leader (see TGID); and a
TTY process group ID for the process group leader (see
TPGID).
PPID -- Parent Process Id
The process ID (pid) of a task's parent.
PR -- Priority
The scheduling priority of the task. If you see ‘rt’ in this
field, it means the task is running under real time
scheduling priority.
Under linux, real time priority is somewhat misleading since
traditionally the operating itself was not preemptible. And
while the 2.6 kernel can be made mostly preemptible, it is
not always so.
PSS -- Proportional Resident Memory, smaps (KiB)
The proportion of this task's share of ‘RSS’ where each page
is divided by the number of processes sharing it. It is also
the sum of the ‘PSan’, ‘PSfd’ and ‘PSsh’ fields.
For example, if a process has 1000 resident pages alone and
1000 resident pages shared with another process, its ‘PSS’
would be 1500 (times page size).
Accessing smaps values is 10x more costly than other memory
statistics and data for other users requires root privileges.
PSan -- Proportional Anonymous Memory, smaps (KiB)
PSfd -- Proportional File Memory, smaps (KiB)
PSsh -- Proportional Shmem Memory, smaps (KiB)
As was true for ‘PSS’ above (total proportional resident
memory), these fields represent the proportion of this task's
share of each type of memory divided by the number of
processes sharing it.
Accessing smaps values is 10x more costly than other memory
statistics and data for other users requires root privileges.
RES -- Resident Memory Size (KiB)
A subset of the virtual address space (VIRT) representing the
non-swapped physical memory a task is currently using. It is
also the sum of the ‘RSan’, ‘RSfd’ and ‘RSsh’ fields.
It can include private anonymous pages, private pages mapped
to files (including program images and shared libraries) plus
shared anonymous pages. All such memory is backed by the
swap file represented separately under SWAP.
Lastly, this field may also include shared file-backed pages
which, when modified, act as a dedicated swap file and thus
will never impact SWAP.
See ‘OVERVIEW, Linux Memory Types’ for additional details.
RSS -- Resident Memory, smaps (KiB)
Another, more precise view of process non-swapped physical
memory. It is obtained from the ‘smaps_rollup’ file and is
generally slightly larger than that shown for ‘RES’.
Accessing smaps values is 10x more costly than other memory
statistics and data for other users requires root privileges.
RSan -- Resident Anonymous Memory Size (KiB)
A subset of resident memory (RES) representing private pages
not mapped to a file.
RSfd -- Resident File-Backed Memory Size (KiB)
A subset of resident memory (RES) representing the implicitly
shared pages supporting program images and shared libraries.
It also includes explicit file mappings, both private and
shared.
RSlk -- Resident Locked Memory Size (KiB)
A subset of resident memory (RES) which cannot be swapped
out.
RSsh -- Resident Shared Memory Size (KiB)
A subset of resident memory (RES) representing the explicitly
shared anonymous shm*/mmap pages.
RUID -- Real User Id
The real user ID.
RUSER -- Real User Name
The real user name.
S -- Process Status
The status of the task which can be one of:
D = uninterruptible sleep
I = idle
R = running
S = sleeping
T = stopped by job control signal
t = stopped by debugger during trace
Z = zombie
Tasks shown as running should be more properly thought of as
ready to run -- their task_struct is simply represented on
the Linux run-queue. Even without a true SMP machine, you
may see numerous tasks in this state depending on top's delay
interval and nice value.
SHR -- Shared Memory Size (KiB)
A subset of resident memory (RES) that may be used by other
processes. It will include shared anonymous pages and shared
file-backed pages. It also includes private pages mapped to
files representing program images and shared libraries.
See ‘OVERVIEW, Linux Memory Types’ for additional details.
SID -- Session Id
A session is a collection of process groups (see PGRP),
usually established by the login shell. A newly forked
process joins the session of its creator. By convention,
this value equals the process ID (see PID) of the first
member of the session, called the session leader, which is
usually the login shell.
STARTED -- Start Time Interval
The length of time since system boot when a process started.
Thus, the most recently started task will display the largest
time interval.
The value will be expressed as ‘MM:SS’ (minutes:seconds).
But if the interval is too great to fit column width it will
be scaled as ‘HH,MM’ (hours,minutes) and possibly beyond.
SUID -- Saved User Id
The saved user ID.
SUPGIDS -- Supplementary Group IDs
The IDs of any supplementary group(s) established at login or
inherited from a task's parent. They are displayed in a
comma delimited list.
Note: The SUPGIDS field, unlike most columns, is not fixed-
width. When displayed, it plus any other variable width
columns will be allocated all remaining screen width (up to
the maximum 512 characters).
SUPGRPS -- Supplementary Group Names
The names of any supplementary group(s) established at login
or inherited from a task's parent. They are displayed in a
comma delimited list.
Note: The SUPGRPS field, unlike most columns, is not fixed-
width. When displayed, it plus any other variable width
columns will be allocated all remaining screen width (up to
the maximum 512 characters).
SUSER -- Saved User Name
The saved user name.
SWAP -- Swapped Size (KiB)
The formerly resident portion of a task's address space
written to the swap file when physical memory becomes over
committed.
See ‘OVERVIEW, Linux Memory Types’ for additional details.
TGID -- Thread Group Id
The ID of the thread group to which a task belongs. It is
the PID of the thread group leader. In kernel terms, it
represents those tasks that share an mm_struct.
TIME -- CPU Time
Total CPU time the task has used since it started. When
Cumulative mode is On, each process is listed with the cpu
time that it and its dead children have used. You toggle
Cumulative mode with ‘S’, which is both a command-line option
and an interactive command. See the ‘S’ interactive command
for additional information regarding this mode.
TIME+ -- CPU Time, hundredths
The same as TIME, but reflecting more granularity through
hundredths of a second.
TPGID -- Tty Process Group Id
The process group ID of the foreground process for the
connected tty, or -1 if a process is not connected to a
terminal. By convention, this value equals the process ID
(see PID) of the process group leader (see PGRP).
TTY -- Controlling Tty
The name of the controlling terminal. This is usually the
device (serial port, pty, etc.) from which the process was
started, and which it uses for input or output. However, a
task need not be associated with a terminal, in which case
you'll see ‘?’ displayed.
UID -- User Id
The effective user ID of the task's owner.
USED -- Memory in Use (KiB)
This field represents the non-swapped physical memory a task
is using (RES) plus the swapped out portion of its address
space (SWAP).
See ‘OVERVIEW, Linux Memory Types’ for additional details.
USER -- User Name
The effective user name of the task's owner.
USS -- Unique Set Size
The non-swapped portion of physical memory (‘RSS’) not shared
with any other process. It is derived from the
‘smaps_rollup’ file.
Accessing smaps values is 10x more costly than other memory
statistics and data for other users requires root privileges.
VIRT -- Virtual Memory Size (KiB)
The total amount of virtual memory used by the task. It
includes all code, data and shared libraries plus pages that
have been swapped out and pages that have been mapped but not
used.
See ‘OVERVIEW, Linux Memory Types’ for additional details.
WCHAN -- Sleeping in Function
This field will show the name of the kernel function in which
the task is currently sleeping. Running tasks will display a
dash (‘-’) in this column.
ioR -- I/O Bytes Read
The number of bytes a process caused to be fetched from the
storage layer.
Root privileges are required to display ‘io’ data for other
users.
ioRop -- I/O Read Operations
The number of read I/O operations (syscalls) for a process.
Such calls might not result in actual physical disk I/O.
ioW -- I/O Bytes Written
The number of bytes a process caused to be sent to the
storage layer.
ioWop -- I/O Write Operations
The number of write I/O operations (syscalls) for a process.
Such calls might not result in actual physical disk I/O.
nDRT -- Dirty Pages Count
The number of pages that have been modified since they were
last written to auxiliary storage. Dirty pages must be
written to auxiliary storage before the corresponding
physical memory location can be used for some other virtual
page.
This field was deprecated with linux 2.6 and is always zero.
nFD -- Number of File Descriptors
The total number of open files associated with a process.
nMaj -- Major Page Fault Count
The number of major page faults that have occurred for a
task. A page fault occurs when a process attempts to read
from or write to a virtual page that is not currently present
in its address space. A major page fault is when auxiliary
storage access is involved in making that page available.
nMin -- Minor Page Fault count
The number of minor page faults that have occurred for a
task. A page fault occurs when a process attempts to read
from or write to a virtual page that is not currently present
in its address space. A minor page fault does not involve
auxiliary storage access in making that page available.
nTH -- Number of Threads
The number of threads associated with a process.
nsCGROUP -- CGROUP namespace
The Inode of the namespace used to hide the identity of the
control group of which process is a member.
nsIPC -- IPC namespace
The Inode of the namespace used to isolate interprocess
communication (IPC) resources such as System V IPC objects
and POSIX message queues.
nsMNT -- MNT namespace
The Inode of the namespace used to isolate filesystem mount
points thus offering different views of the filesystem
hierarchy.
nsNET -- NET namespace
The Inode of the namespace used to isolate resources such as
network devices, IP addresses, IP routing, port numbers, etc.
nsPID -- PID namespace
The Inode of the namespace used to isolate process ID numbers
meaning they need not remain unique. Thus, each such
namespace could have its own ‘init/systemd’ (PID #1) to
manage various initialization tasks and reap orphaned child
processes.
nsTIME -- TIME namespace
The Inode of the namespace which allows processes to see
different system times in a way similar to the UTS namespace.
nsUSER -- USER namespace
The Inode of the namespace used to isolate the user and group
ID numbers. Thus, a process could have a normal unprivileged
user ID outside a user namespace while having a user ID of 0,
with full root privileges, inside that namespace.
nsUTS -- UTS namespace
The Inode of the namespace used to isolate hostname and NIS
domain name. UTS simply means "Unix Time-sharing System".
vMj -- Major Page Fault Count Delta
The number of major page faults that have occurred since the
last update (see nMaj).
vMn -- Minor Page Fault Count Delta
The number of minor page faults that have occurred since the
last update (see nMin).
3b. MANAGING Fields
After pressing the interactive command ‘f’ (Fields Management)
you will be presented with a screen showing: 1) the ‘current’
window name; 2) the designated sort field; 3) all fields in their
current order along with descriptions. Entries marked with an
asterisk are the currently displayed fields, screen width
permitting.
• As the on screen instructions indicate, you navigate among
the fields with the Up and Down arrow keys. The PgUp,
PgDn, Home and End keys can also be used to quickly reach
the first or last available field.
• The Right arrow key selects a field for repositioning and
the Left arrow key or the <Enter> key commits that field's
placement.
• The ‘d’ key or the <Space> bar toggles a field's display
status, and thus the presence or absence of the asterisk.
• The ‘s’ key designates a field as the sort field. See
topic 4c. TASK AREA Commands, SORTING for additional
information regarding your selection of a sort field.
• The ‘a’ and ‘w’ keys can be used to cycle through all
available windows and the ‘q’ or <Esc> keys exit Fields
Management.
The Fields Management screen can also be used to change the
‘current’ window/field group in either full-screen mode or
alternate-display mode. Whatever was targeted when ‘q’ or <Esc>
was pressed will be made current as you return to the top
display. See topic 5. ALTERNATE-DISPLAY Provisions and the ‘g’
interactive command for insight into ‘current’ windows and field
groups.
Note: Any window that has been scrolled horizontally will be
reset if any field changes are made via the Fields Management
screen. Any vertical scrolled position, however, will not be
affected. See topic 5c. SCROLLING a Window for additional
information regarding vertical and horizontal scrolling.
Listed below is a brief index of commands within categories.
Some commands appear more than once -- their meaning or scope
may vary depending on the context in which they are issued.
4a. Global-Commands
<Ent/Sp> ?, =, 0,
A, B, d, E, e, g, H, h, I, k, q, r, s, W, X, Y, Z,
^G, ^K, ^N, ^P, ^U, ^L, ^R
4b. Summary-Area-Commands
C, l, t, m, 1, 2, 3, 4, 5, !
4c. Task-Area-Commands
Appearance: b, J, j, x, y, z
Content: c, F, f, O, o, S, U, u, V, v, ^E
Size: #, i, n
Sorting: <, >, f, R
4d. Color-Mapping
<Ret>, a, B, b, H, M, q, S, T, w, z, @, 0 - 7
5b. Commands-for-Windows
-, _, =, +, A, a, G, g, w
5c. Scrolling-a-Window
C, Up, Dn, Left, Right, PgUp, PgDn, Home, End
5d. Searching-in-a-Window
L, &
5e. Filtering-in-a-Window
O, o, ^O, =, +
4a. GLOBAL Commands
The global interactive commands are always available in both
full-screen mode and alternate-display mode. However, some of
these interactive commands are not available when running in
Secure mode.
If you wish to know in advance whether or not your top has been
secured, simply ask for help and view the system summary on the
second line.
<Enter> or <Space> :Refresh-Display
These commands awaken top and following receipt of any
input the entire display will be repainted. They also
force an update of any hotplugged cpu or physical memory
changes.
Use either of these keys if you have a large delay
interval and wish to see current status,
? | h :Help
There are two help levels available. The first will
provide a reminder of all the basic interactive commands.
If top is secured, that screen will be abbreviated.
Typing ‘h’ or ‘?’ on that help screen will take you to
help for those interactive commands applicable to
alternate-display mode.
= :Exit-Display-Limits
Removes restrictions on what is shown. This command will
reverse any ‘i’ (idle tasks), ‘n’ (max tasks), ‘v’ (hide
children) and ‘F’ focus commands that might be active. It
also provides for an exit from PID monitoring, User
filtering, Other filtering, Locate processing and Combine
Cpus mode.
Additionally, if the window has been scrolled it will be
reset with this command.
0 :Zero-Suppress toggle
This command determines whether zeros are shown or
suppressed for many of the fields in a task window.
Fields like UID, GID, NI, PR or P are not affected by this
toggle.
A :Alternate-Display-Mode toggle
This command will switch between full-screen mode and
alternate-display mode. See topic 5. ALTERNATE-DISPLAY
Provisions and the ‘g’ interactive command for insight
into ‘current’ windows and field groups.
B :Bold-Disable/Enable toggle
This command will influence use of the bold terminfo
capability and alters both the summary area and task area
for the ‘current’ window. While it is intended primarily
for use with dumb terminals, it can be applied anytime.
Note: When this toggle is On and top is operating in
monochrome mode, the entire display will appear as normal
text. Thus, unless the ‘x’ and/or ‘y’ toggles are using
reverse for emphasis, there will be no visual confirmation
that they are even on.
* d | s :Change-Delay-Time-interval
You will be prompted to enter the delay time, in seconds,
between display updates.
Fractional seconds are honored, but a negative number is
not allowed. Entering 0 causes (nearly) continuous
updates, with an unsatisfactory display as the system and
tty driver try to keep up with top's demands. The delay
value is inversely proportional to system loading, so set
it with care.
If at any time you wish to know the current delay time,
simply ask for help and view the system summary on the
second line.
E :Enforce-Summary-Memory-Scale in Summary Area
With this command you can cycle through the available
summary area memory scaling which ranges from KiB
(kibibytes or 1,024 bytes) through EiB (exbibytes or
1,152,921,504,606,846,976 bytes).
If you see a ‘+’ between a displayed number and the
following label, it means that top was forced to truncate
some portion of that number. By raising the scaling
factor, such truncation can be avoided.
e :Enforce-Task-Memory-Scale in Task Area
With this command you can cycle through the available task
area memory scaling which ranges from KiB (kibibytes or
1,024 bytes) through PiB (pebibytes or
1,125,899,906,842,624 bytes).
While top will try to honor the selected target range,
additional scaling might still be necessary in order to
accommodate current values. If you wish to see a more
homogeneous result in the memory columns, raising the
scaling range will usually accomplish that goal. Raising
it too high, however, is likely to produce an all zero
result which cannot be suppressed with the ‘0’ interactive
command.
g :Choose-Another-Window/Field-Group
You will be prompted to enter a number between 1 and 4
designating the field group which should be made the
‘current’ window. You will soon grow comfortable with
these 4 windows, especially after experimenting with
alternate-display mode.
H :Threads-mode toggle
When this toggle is On, individual threads will be
displayed for all processes in all visible task windows.
Otherwise, top displays a summation of all threads in each
process.
I :Irix/Solaris-Mode toggle
When operating in Solaris mode (‘I’ toggled Off), a task's
cpu usage will be divided by the total number of CPUs.
After issuing this command, you'll be told the new state
of this toggle.
* k :Kill-a-task
You will be prompted for a PID and then the signal to
send.
Entering no PID or a negative number will be interpreted
as the default shown in the prompt (the first task
displayed). A PID value of zero means the top program
itself.
The default signal, as reflected in the prompt, is
SIGTERM. However, you can send any signal, via number or
name.
If you wish to abort the kill process, do one of the
following depending on your progress:
1) at the pid prompt, type an invalid number
2) at the signal prompt, type 0 (or any invalid signal)
3) at any prompt, type <Esc>
q :Quit
* r :Renice-a-Task
You will be prompted for a PID and then the value to nice
it to.
Entering no PID or a negative number will be interpreted
as the default shown in the prompt (the first task
displayed). A PID value of zero means the top program
itself.
A positive nice value will cause a process to lose
priority. Conversely, a negative nice value will cause a
process to be viewed more favorably by the kernel. As a
general rule, ordinary users can only increase the nice
value and are prevented from lowering it.
If you wish to abort the renice process, do one of the
following depending on your progress:
1) at the pid prompt, type an invalid number
2) at the nice prompt, type <Enter> with no input
3) at any prompt, type <Esc>
W :Write-the-Configuration-File
This will save all of your options and toggles plus the
current display mode and delay time. By issuing this
command just before quitting top, you will be able restart
later in exactly that same state.
X :Extra-Fixed-Width
Some fields are fixed width and not scalable. As such,
they are subject to truncation which would be indicated by
a ‘+’ in the last position.
This interactive command can be used to alter the widths
of the following fields:
field default field default field default
GID 5 Docker 8 WCHAN 10
LOGID 5 GROUP 8 nsCGROUP 10
RUID 5 LXC 8 nsIPC 10
SUID 5 RUSER 8 nsMNT 10
UID 5 SUSER 8 nsNET 10
TTY 8 nsPID 10
USER 8 nsTIME 10
nsUSER 10
nsUTS 10
You will be prompted for the amount to be added to the
default widths shown above. Entering zero forces a return
to those defaults.
If you enter a negative number, top will automatically
increase the column size as needed until there is no more
truncated data.
Note: Whether explicitly or automatically increased, the
widths for these fields are never decreased by top. To
narrow them you must specify a smaller number or restore
the defaults.
Y :Inspect-Other-Output
After issuing the ‘Y’ interactive command, you will be
prompted for a target PID. Typing a value or accepting
the default results in a separate screen. That screen can
be used to view a variety of files or piped command output
while the normal top iterative display is paused.
Note: This interactive command is only fully realized when
supporting entries have been manually added to the end of
the top configuration file. For details on creating those
entries, see topic 6b. ADDING INSPECT Entries.
Most of the keys used to navigate the Inspect feature are
reflected in its header prologue. There are, however,
additional keys available once you have selected a
particular file or command. They are familiar to anyone
who has used the pager ‘less’ and are summarized here for
future reference.
key function
= alternate status-line, file or pipeline
/ find, equivalent to ‘L’ locate
n find next, equivalent to ‘&’ locate next
<Space> scroll down, equivalent to <PgDn>
b scroll up, equivalent to <PgUp>
g first line, equivalent to <Home>
G last line, equivalent to <End>
Z :Change-Color-Mapping
This key will take you to a separate screen where you can
change the colors for the ‘current’ window, or for all
windows. For details regarding this interactive command
see topic 4d. COLOR Mapping.
^G :Display-Control-Groups (Ctrl key + ‘g’)
^K :Display-Cmdline (Ctrl key + ‘k’)
^N :Display-Environment (Ctrl key + ‘n’)
^P :Display-Namesspaces (Ctrl key + ‘p’)
^U :Display-Supplementary-Groups (Ctrl key + ‘u’)
Applied to the first process displayed, these commands
will show that task's full (potentially wrapped)
information. Such data will be displayed in a separate
window at the bottom of the screen while normal top
monitoring continues.
Keying the same ‘Ctrl’ command a second time removes that
separate window as does the ‘=’ command. Keying a
different ‘Ctrl’ combination, while one is already active,
immediately transitions to the new information.
Notable among these provisions is the Ctrl+N (environment)
command. Its output can be extensive and not easily read
when line wrapped. A more readable version can be
achieved with an ‘Inspect’ entry in the rcfile like the
following.
pipe ^I Environment ^I cat /proc/%d/environ | tr '\0' '\n'
See the ‘Y’ interactive command above and topic 6b. ADDING
INSPECT Entries for additional information.
As an alternative to ‘Inspect’, and available to all of
these ‘Ctrl’ commands, the tab key can be used to
highlight individual elements in the bottom window.
^L :Logged-Messages (Ctrl key + ‘l’)
The 10 most recent messages are displayed in a separate
window at the bottom of the screen while normal top
monitoring continues. Keying ‘^L’ a second time removes
that window as does the ‘=’ command. Use the tab key to
highlight individual messages.
* ^R :Renice-an-Autogroup (Ctrl key + ‘r’)
You will be prompted for a PID and then the value for its
autogroup AGNI.
Entering no PID will be interpreted as the default shown
in the prompt (the first task displayed).
A positive AGNI value will cause processes in that
autogroup to lose priority. Conversely, a negative value
causes them to be viewed more favorably by the kernel.
Ordinary users are not allowed to set negative AGNI
values.
If you wish to abort the renice process type <Esc>.
* The commands shown with an asterisk (‘*’) are not available in
Secure mode, nor will they be shown on the level-1 help
screen.
4b. SUMMARY AREA Commands
The summary area interactive commands are always available in
both full-screen mode and alternate-display mode. They affect
the beginning lines of your display and will determine the
position of messages and prompts.
These commands always impact just the ‘current’ window/field
group. See topic 5. ALTERNATE-DISPLAY Provisions and the ‘g’
interactive command for insight into ‘current’ windows and field
groups.
C :Show-scroll-coordinates toggle
Toggle an informational message which is displayed
whenever the message line is not otherwise being used.
For additional information see topic 5c. SCROLLING a
Window.
l :Load-Average/Uptime toggle
This is also the line containing the program name
(possibly an alias) when operating in full-screen mode or
the ‘current’ window name when operating in
alternate-display mode.
t :Task/Cpu-States toggle
This command affects from 2 to many summary area lines,
depending on the state of the ‘1’, ‘2’ or ‘3’ command
toggles and whether or not top is running under true SMP.
This portion of the summary area is also influenced by the
‘H’ interactive command toggle, as reflected in the total
label which shows either Tasks or Threads.
This command serves as a 4-way toggle, cycling through
these modes:
1. detailed percentages by category
2. abbreviated user/system and total % + bar graph
3. abbreviated user/system and total % + block graph
4. turn off task and cpu states display
When operating in either of the graphic modes, the display
becomes much more meaningful when individual CPUs or NUMA
nodes are also displayed. See the the ‘1’, ‘2’ and ‘3’
commands below for additional information.
m :Memory/Swap-Usage toggle
This command affects the two summary area lines dealing
with physical and virtual memory.
This command serves as a 4-way toggle, cycling through
these modes:
1. detailed percentages by memory type
2. abbreviated % used/total available + bar graph
3. abbreviated % used/total available + block graph
4. turn off memory display
1 :Single/Separate-Cpu-States toggle
This command affects how the ‘t’ command's Cpu States
portion is shown. Although this toggle exists primarily
to serve massively-parallel SMP machines, it is not
restricted to solely SMP environments.
When you see ‘%Cpu(s):’ in the summary area, the ‘1’
toggle is On and all cpu information is gathered in a
single line. Otherwise, each cpu is displayed separately
as: ‘%Cpu0, %Cpu1, ...’ up to available screen height.
2 :NUMA-Nodes/Cpu-Summary toggle
This command toggles between the ‘1’ command cpu summary
display (only) or a summary display plus the cpu usage
statistics for each NUMA Node. It is only available if a
system has the requisite NUMA support.
3 :Expand-NUMA-Node
You will be invited to enter a number representing a NUMA
Node. Thereafter, a node summary plus the statistics for
each cpu in that node will be shown until the ‘1’, ‘2’ or
‘4’ command toggle is pressed. This interactive command
is only available if a system has the requisite NUMA
support.
4 :Display-Multiple-Elements-Adjacent toggle
This command toggle turns the ‘1’ toggle Off and shows
multiple CPU and Memory results on each line. Each
successive ‘4’ key adds another CPU until again reverting
to separate lines for CPU and Memory results.
A maximum of 8 CPUs per line can be displayed in this
manner. However, data truncation may occur before
reaching the maximum. That is definitely true when
displaying detailed statistics via the ‘t’ command toggle
since such data cannot be scaled like the graphic
representations.
If one wished to quickly exit adjacent mode without
cycling all the way to 8, simply use the ‘1’ command
toggle.
5 :Display-P-Cores-and-E-Cores toggle
This command toggle is only active when the ‘t’ toggle is
On and the ‘1’, ‘2’, ‘3’ and ‘!’ toggles are Off, thus
showing individual CPU results. It assumes a platform has
multiple cores of two distinct types, either multi-
threaded (P-Core) or single-threaded (E-Core).
While normally each cpu is displayed as ‘%Cpu0, %Cpu1,
...’, this toggle can be used to identify and/or filter
those cpus by their core type, either P-Core (performance)
or E-Core (efficient).
The 1st time ‘5’ is struck, each CPU is displayed as
‘%CpP’ or ‘%CpE’ representing the two core types. The 2nd
time, only P-Cores (%CpP) will be shown. The 3rd time,
only E-Cores (%CpE) are displayed. When this command
toggle is struck for the 4th time, the CPU display returns
to the normal ‘%Cpu’ convention.
If separate performance and efficient categories are not
present, this command toggle will have no effect.
! :Combine-Cpus-Mode toggle
This command toggle is intended for massively parallel SMP
environments where, even with the ‘4’ command toggle, not
all processors can be displayed. With each press of ‘!’
the number of cpus combined is doubled thus reducing the
total number of cpu lines displayed.
For example, with the first press of ‘!’ two cpus will be
combined and displayed as ‘0-1, 2-3, ...’ instead of the
normal ‘%Cpu0, %Cpu1, %Cpu2, %Cpu3, ...’. With a second
‘!’ command toggle four cpus are combined and shown as
‘0-3, 4-7, ...’. Then the third ‘!’ press, combining
eight cpus, shows as ‘0-7, 8-15, ...’, etc.
Such progression continues until individual cpus are again
displayed and impacts both the ‘1’ and ‘4’ toggles (one or
multiple columns). Use the ‘=’ command to exit Combine
Cpus mode.
Note: If the entire summary area has been toggled Off for any
window, you would be left with just the message line. In that
way, you will have maximized available task rows but
(temporarily) sacrificed the program name in full-screen mode or
the ‘current’ window name when in alternate-display mode.
4c. TASK AREA Commands
The task area interactive commands are always available in
full-screen mode.
The task area interactive commands are never available in
alternate-display mode if the ‘current’ window's task display has
been toggled Off (see topic 5. ALTERNATE-DISPLAY Provisions).
APPEARANCE of task window
J :Justify-Numeric-Columns toggle
Alternates between right-justified (the default) and left-
justified numeric data. If the numeric data completely
fills the available column, this command toggle may impact
the column header only.
j :Justify-Character-Columns toggle
Alternates between left-justified (the default) and right-
justified character data. If the character data
completely fills the available column, this command toggle
may impact the column header only.
The following commands will also be influenced by the state of
the global ‘B’ (bold enable) toggle.
b :Bold/Reverse toggle
This command will impact how the ‘x’ and ‘y’ toggles are
displayed. It may also impact the summary area when a bar
graph has been selected for cpu states or memory usage via
the ‘t’ or ‘m’ toggles.
x :Column-Highlight toggle
Changes highlighting for the current sort field. If you
forget which field is being sorted this command can serve
as a quick visual reminder, providing the sort field is
being displayed. The sort field might not be visible
because:
1) there is insufficient Screen Width
2) the ‘f’ interactive command turned it Off
y :Row-Highlight toggle
Changes highlighting for "running" tasks. For additional
insight into this task state, see topic 3a. DESCRIPTIONS
of Fields, the ‘S’ field (Process Status).
Use of this provision provides important insight into your
system's health. The only costs will be a few additional
tty escape sequences.
z :Color/Monochrome toggle
Switches the ‘current’ window between your last used color
scheme and the older form of black-on-white or white-on-
black. This command will alter both the summary area and
task area but does not affect the state of the ‘x’, ‘y’ or
‘b’ toggles.
CONTENT of task window
c :Command-Line/Program-Name toggle
This command will be honored whether or not the COMMAND
column is currently visible. Later, should that field
come into view, the change you applied will be seen.
F :Maintain-Parent-Focus toggle
When in forest view mode, this key serves as a toggle to
retain focus on a target task, presumably one with forked
children. If forest view mode is Off this key has no
effect.
The toggle is applied to the first (topmost) process in
the ‘current’ window. Once established, that task is
always displayed as the first (topmost) process along with
its forked children. All other processes will be
suppressed.
Note: keys like ‘i’ (idle tasks), ‘n’ (max tasks), ‘v’
(hide children) and User/Other filtering remain accessible
and can impact what is displayed.
f :Fields-Management
This key displays a separate screen where you can change
which fields are displayed, their order and also designate
the sort field. For additional information on this
interactive command see topic 3b. MANAGING Fields.
O | o :Other-Filtering
You will be prompted for the selection criteria which then
determines which tasks will be shown in the ‘current’
window. Your criteria can be made case sensitive or case
can be ignored. And you determine if top should include
or exclude matching tasks.
See topic 5e. FILTERING in a window for details on these
and additional related interactive commands.
S :Cumulative-Time-Mode toggle
When Cumulative mode is On, each process is listed with
the cpu time that it and its dead children have used.
When Off, programs that fork into many separate tasks will
appear less demanding. For programs like ‘init’ or a
shell this is appropriate but for others, like compilers,
perhaps not. Experiment with two task windows sharing the
same sort field but with different ‘S’ states and see
which representation you prefer.
After issuing this command, you'll be informed of the new
state of this toggle. If you wish to know in advance
whether or not Cumulative mode is in effect, simply ask
for help and view the window summary on the second line.
U | u :Show-Specific-User-Only
You will be prompted for the uid or name of the user to
display. The -u option matches on effective user whereas
the -U option matches on any user (real, effective, saved,
or filesystem).
Thereafter, in that task window only matching users will
be shown, or possibly no processes will be shown.
Prepending an exclamation point (‘!’) to the user id or
name instructs top to display only processes with users
not matching the one provided.
Different task windows can be used to filter different
users. Later, if you wish to monitor all users again in
the ‘current’ window, re-issue this command but just press
<Enter> at the prompt.
V :Forest-View-Mode toggle
In this mode, processes are reordered according to their
parents and the layout of the COMMAND column resembles
that of a tree. In forest view mode it is still possible
to toggle between program name and command line (see the
‘c’ interactive command) or between processes and threads
(see the ‘H’ interactive command).
Note: Typing any key affecting the sort order will exit
forest view mode in the ‘current’ window. See topic 4c.
TASK AREA Commands, SORTING for information on those keys.
v :Hide/Show-Children toggle
When in forest view mode, this key serves as a toggle to
collapse or expand the children of a parent.
The toggle is applied against the first (topmost) process
in the ‘current’ window. See topic 5c. SCROLLING a Window
for additional information regarding vertical scrolling.
If the target process has not forked any children, this
key has no effect. It also has no effect when not in
forest view mode.
^E :Scale-CPU-Time-fields (Ctrl key + ‘e’)
The ‘time’ fields are normally displayed with the greatest
precision their widths permit. This toggle reduces that
precision until it wraps. It also illustrates the scaling
those fields might experience automatically, which usually
depends on how long the system runs.
For example, if ‘MMM:SS.hh’ is shown, each ^E keystroke
would change it to: ‘MM:SS’, ‘Hours,MM’, ‘Days+Hours’ and
finally ‘Weeks+Days’.
Not all time fields are subject to the full range of such
scaling.
SIZE of task window
i :Idle-Process toggle
Displays all tasks or just active tasks. When this toggle
is Off, tasks that have not used any CPU since the last
update will not be displayed. However, due to the
granularity of the %CPU and TIME+ fields, some processes
may still be displayed that appear to have used no CPU.
If this command is applied to the last task display when
in alternate-display mode, then it will not affect the
window's size, as all prior task displays will have
already been painted.
n | # :Set-Maximum-Tasks
You will be prompted to enter the number of tasks to
display. The lessor of your number and available screen
rows will be used.
When used in alternate-display mode, this is the command
that gives you precise control over the size of each
currently visible task display, except for the very last.
It will not affect the last window's size, as all prior
task displays will have already been painted.
Note: If you wish to increase the size of the last visible
task display when in alternate-display mode, simply
decrease the size of the task display(s) above it.
SORTING of task window
For compatibility, this top supports most of the former top
sort keys. Since this is primarily a service to former top
users, these commands do not appear on any help screen.
command sorted-field supported
A start time (non-display) No
M %MEM Yes
N PID Yes
P %CPU Yes
T TIME+ Yes
Before using any of the following sort provisions, top
suggests that you temporarily turn on column highlighting
using the ‘x’ interactive command. That will help ensure that
the actual sort environment matches your intent.
The following interactive commands will only be honored when
the current sort field is visible. The sort field might not
be visible because:
1) there is insufficient Screen Width
2) the ‘f’ interactive command turned it Off
< :Move-Sort-Field-Left
Moves the sort column to the left unless the current
sort field is the first field being displayed.
> :Move-Sort-Field-Right
Moves the sort column to the right unless the current
sort field is the last field being displayed.
The following interactive commands will always be honored
whether or not the current sort field is visible.
f :Fields-Management
This key displays a separate screen where you can
change which field is used as the sort column, among
other functions. This can be a convenient way to
simply verify the current sort field, when running top
with column highlighting turned Off.
R :Reverse/Normal-Sort-Field toggle
Using this interactive command you can alternate
between high-to-low and low-to-high sorts.
4d. COLOR Mapping
When you issue the ‘Z’ interactive command, you will be presented
with a separate screen. That screen can be used to change the
colors in just the ‘current’ window or in all four windows before
returning to the top display.
The following interactive commands are available.
5 upper case letters to select a target
8 numbers to select a color (@ selects no color)
normal toggles available
B :bold disable/enable
b :running tasks "bold"/reverse
z :color/mono
other commands available
a/w :apply, then go to next/prior
<Enter> :apply and exit
q :abandon current changes and exit
If you use ‘a’ or ‘w’ to cycle the targeted window, you will have
applied the color scheme that was displayed when you left that
window. You can, of course, easily return to any window and
reapply different colors or turn colors Off completely with the
‘z’ toggle.
The Color Mapping screen can also be used to change the ‘current’
window/field group in either full-screen mode or
alternate-display mode. Whatever was targeted when ‘q’ or
<Enter> was pressed will be made current as you return to the top
display.
5a. WINDOWS Overview
Field Groups/Windows:
In full-screen mode there is a single window represented by
the entire screen. That single window can still be changed to
display 1 of 4 different field groups (see the ‘g’ interactive
command, repeated below). Each of the 4 field groups has a
unique separately configurable summary area and its own
configurable task area.
In alternate-display mode, those 4 underlying field groups can
now be made visible simultaneously, or can be turned Off
individually at your command.
The summary area will always exist, even if it's only the
message line. At any given time only one summary area can be
displayed. However, depending on your commands, there could
be from zero to four separate task displays currently showing
on the screen.
Current Window:
The ‘current’ window is the window associated with the summary
area and the window to which task related commands are always
directed. Since in alternate-display mode you can toggle the
task display Off, some commands might be restricted for the
‘current’ window.
A further complication arises when you have toggled the first
summary area line Off. With the loss of the window name (the
‘l’ toggled line), you'll not easily know what window is the
‘current’ window.
5b. COMMANDS for Windows
- | _ :Show/Hide-Window(s) toggles
The ‘-’ key turns the ‘current’ window's task display On
and Off. When On, that task area will show a minimum of
the columns header you've established with the ‘f’
interactive command. It will also reflect any other task
area options/toggles you've applied yielding zero or more
tasks.
The ‘_’ key does the same for all task displays. In other
words, it switches between the currently visible task
display(s) and any task display(s) you had toggled Off.
If all 4 task displays are currently visible, this
interactive command will leave the summary area as the
only display element.
* = | + :Equalize/Reset-Window(s)
The ‘=’ key forces the ‘current’ window's task display to
be visible. It also reverses any active ‘i’ (idle tasks),
‘n’ (max tasks), ‘u/U’ (user filter), ‘o/O’ (other
filter), ‘v’ (hide children), ‘F’ focused, ‘L’ (locate)
and ‘!’ (combine cpus) commands. Also, if the window had
been scrolled, it will be reset with this command. See
topic 5c. SCROLLING a Window for additional information
regarding vertical and horizontal scrolling.
The ‘+’ key does the same for all windows. The four task
displays will reappear, evenly balanced, while retaining
any customizations previously applied beyond those noted
for the ‘=’ command toggle.
* A :Alternate-Display-Mode toggle
This command will switch between full-screen mode and
alternate-display mode.
The first time you issue this command, all four task
displays will be shown. Thereafter when you switch modes,
you will see only the task display(s) you've chosen to
make visible.
* a | w :Next-Window-Forward/Backward
This will change the ‘current’ window, which in turn
changes the window to which commands are directed. These
keys act in a circular fashion so you can reach any
desired window using either key.
Assuming the window name is visible (you have not toggled
‘l’ Off), whenever the ‘current’ window name loses its
emphasis/color, that's a reminder the task display is Off
and many commands will be restricted.
G :Change-Window/Field-Group-Name
You will be prompted for a new name to be applied to the
‘current’ window. It does not require that the window
name be visible (the ‘l’ toggle to be On).
* The interactive commands shown with an asterisk (‘*’) have use
beyond alternate-display mode.
=, A, g are always available
a, w act the same with color mapping
and fields management
* g :Choose-Another-Window/Field-Group
You will be prompted to enter a number between 1 and 4
designating the field group which should be made the
‘current’ window.
In full-screen mode, this command is necessary to alter
the ‘current’ window. In alternate-display mode, it is
simply a less convenient alternative to the ‘a’ and ‘w’
commands.
5c. SCROLLING a Window
Typically a task window is a partial view into a system's total
tasks/threads which shows only some of the available
fields/columns. With these scrolling keys, you can move that
view vertically or horizontally to reveal any desired task or
column.
Up,PgUp :Scroll-Tasks
Move the view up toward the first task row, until the first
task is displayed at the top of the ‘current’ window. The Up
arrow key moves a single line while PgUp scrolls the entire
window.
Down,PgDn :Scroll-Tasks
Move the view down toward the last task row, until the last
task is the only task displayed at the top of the ‘current’
window. The Down arrow key moves a single line while PgDn
scrolls the entire window.
Left,Right :Scroll-Columns
Move the view of displayable fields horizontally one column
at a time.
Note: As a reminder, some fields/columns are not fixed-width
but allocated all remaining screen width when visible. When
scrolling right or left, that feature may produce some
unexpected results initially.
Additionally, there are special provisions for any variable
width field when positioned as the last displayed field.
Once that field is reached via the right arrow key, and is
thus the only column shown, you can continue scrolling
horizontally within such a field. See the ‘C’ interactive
command below for additional information.
Home :Jump-to-Home-Position
Reposition the display to the un-scrolled coordinates.
End :Jump-to-End-Position
Reposition the display so that the rightmost column reflects
the last displayable field and the bottom task row represents
the last task.
Note: From this position it is still possible to scroll down
and right using the arrow keys. This is true until a single
column and a single task is left as the only display element.
C :Show-scroll-coordinates toggle
Toggle an informational message which is displayed whenever
the message line is not otherwise being used. That message
will take one of two forms depending on whether or not a
variable width column has also been scrolled.
scroll coordinates: y = n/n (tasks), x = n/n (fields)
scroll coordinates: y = n/n (tasks), x = n/n (fields) + nn
The coordinates shown as n/n are relative to the upper left
corner of the ‘current’ window. The additional ‘+ nn’
represents the displacement into a variable width column when
it has been scrolled horizontally. Such displacement occurs
in normal 8 character tab stop amounts via the right and left
arrow keys.
y = n/n (tasks)
The first n represents the topmost visible task and is
controlled by scrolling keys. The second n is updated
automatically to reflect total tasks.
x = n/n (fields)
The first n represents the leftmost displayed column and
is controlled by scrolling keys. The second n is the
total number of displayable fields and is established
with the ‘f’ interactive command.
The above interactive commands are always available in
full-screen mode but never available in alternate-display mode if
the ‘current’ window's task display has been toggled Off.
Note: When any form of filtering is active, you can expect some
slight aberrations when scrolling since not all tasks will be
visible. This is particularly apparent when using the Up/Down
arrow keys.
5d. SEARCHING in a Window
You can use these interactive commands to locate a task row
containing a particular value.
L :Locate-a-string
You will be prompted for the case-sensitive string to locate
starting from the current window coordinates. There are no
restrictions on search string content.
Searches are not limited to values from a single field or
column. All of the values displayed in a task row are
allowed in a search string. You may include spaces, numbers,
symbols and even forest view artwork.
Keying <Enter> with no input will effectively disable the ‘&’
key until a new search string is entered.
& :Locate-next
Assuming a search string has been established, top will
attempt to locate the next occurrence.
When a match is found, the current window is repositioned
vertically so the task row containing that string is first. The
scroll coordinates message can provide confirmation of such
vertical repositioning (see the ‘C’ interactive command).
Horizontal scrolling, however, is never altered via searching.
The availability of a matching string will be influenced by the
following factors.
a. Which fields are displayable from the total available,
see topic 3b. MANAGING Fields.
b. Scrolling a window vertically and/or horizontally,
see topic 5c. SCROLLING a Window.
c. The state of the command/command-line toggle,
see the ‘c’ interactive command.
d. The stability of the chosen sort column,
for example PID is good but %CPU bad.
If a search fails, restoring the ‘current’ window home
(unscrolled) position, scrolling horizontally, displaying
command-lines or choosing a more stable sort field could yet
produce a successful ‘&’ search.
The above interactive commands are always available in
full-screen mode but never available in alternate-display mode if
the ‘current’ window's task display has been toggled Off.
5e. FILTERING in a Window
You can use this ‘Other Filter’ feature to establish selection
criteria which will then determine which tasks are shown in the
‘current’ window. Such filters can be made persistent if
preserved in the rcfile via the ‘W’ interactive command.
Establishing a filter requires: 1) a field name; 2) an operator;
and 3) a selection value, as a minimum. This is the most complex
of top's user input requirements so, when you make a mistake,
command recall will be your friend. Remember the Up/Down arrow
keys or their aliases when prompted for input.
Filter Basics
1. field names are case sensitive and spelled as in the header
2. selection values need not comprise the full displayed field
3. a selection is either case insensitive or sensitive to case
4. the default is inclusion, prepending ‘!’ denotes exclusions
5. multiple selection criteria can be applied to a task window
6. inclusion and exclusion criteria can be used simultaneously
7. the 1 equality and 2 relational filters can be freely mixed
8. separate unique filters are maintained for each task window
If a field is not turned on or is not currently in view, then
your selection criteria will not affect the display. Later,
should a filtered field become visible, the selection criteria
will then be applied.
Keyboard Summary
O :Other-Filter (upper case)
You will be prompted to establish a case sensitive filter.
o :Other-Filter (lower case)
You will be prompted to establish a filter that ignores
case when matching.
^O :Show-Active-Filters (Ctrl key + ‘o’)
This can serve as a reminder of which filters are active in
the ‘current’ window. A summary will be shown on the
message line until you press the <Enter> key.
= :Reset-Filtering in current window
This clears all of your selection criteria in the ‘current’
window. It also has additional impact so please see topic
4a. GLOBAL Commands.
+ :Reset-Filtering in all windows
This clears the selection criteria in all windows, assuming
you are in alternate-display mode. As with the ‘=’
interactive command, it too has additional consequences so
you might wish to see topic 5b. COMMANDS for Windows.
Input Requirements
When prompted for selection criteria, the data you provide
must take one of two forms. There are 3 required pieces of
information, with a 4th as optional. These examples use
spaces for clarity but your input generally would not.
#1 #2 #3 ( required )
Field-Name ? include-if-value
! Field-Name ? exclude-if-value
#4 ( optional )
Items #1, #3 and #4 should be self-explanatory. Item #2
represents both a required delimiter and the operator which
must be one of either equality (‘=’) or relation (‘<’ or ‘>’).
The ‘=’ equality operator requires only a partial match and
that can reduce your ‘if-value’ input requirements. The ‘>’
or ‘<’ relational operators always employ string comparisons,
even with numeric fields. They are designed to work with a
field's default justification and with homogeneous data. When
some field's numeric amounts have been subjected to scaling
while others have not, that data is no longer homogeneous.
If you establish a relational filter and you have changed the
default Numeric or Character justification, that filter is
likely to fail. When a relational filter is applied to a
memory field and you have not changed the scaling, it may
produce misleading results. This happens, for example,
because ‘100.0m’ (MiB) would appear greater than ‘1.000g’
(GiB) when compared as strings.
If your filtered results appear suspect, simply altering
justification or scaling may yet achieve the desired
objective. See the ‘j’, ‘J’ and ‘e’ interactive commands for
additional information.
Potential Problems
These GROUP filters could produce the exact same results or
the second one might not display anything at all, just a blank
task window.
GROUP=root ( only the same results when )
GROUP=ROOT ( invoked via lower case ‘o’ )
Either of these RES filters might yield inconsistent and/or
misleading results, depending on the current memory scaling
factor. Or both filters could produce the exact same results.
RES>9999 ( only the same results when )
!RES<10000 ( memory scaling is at ‘KiB’ )
This nMin filter illustrates a problem unique to scalable
fields. This particular field can display a maximum of 4
digits, beyond which values are automatically scaled to KiB or
above. So while amounts greater than 9999 exist, they will
appear as 2.6m, 197k, etc.
nMin>9999 ( always a blank task window )
Potential Solutions
These examples illustrate how Other Filtering can be
creatively applied to achieve almost any desired result.
Single quotes are sometimes shown to delimit the spaces which
are part of a filter or to represent a request for status (^O)
accurately. But if you used them with if-values in real life,
no matches would be found.
Assuming field nTH is displayed, the first filter will result
in only multi-threaded processes being shown. It also reminds
us that a trailing space is part of every displayed field.
The second filter achieves the exact same results with less
typing.
!nTH=‘ 1 ’ ( ‘’ for clarity only )
nTH>1 ( same with less i/p )
With Forest View mode active and the COMMAND column in view,
this filter effectively collapses child processes so that just
3 levels are shown.
!COMMAND=‘ ‘- ’ ( ‘’ for clarity only )
The final two filters appear as in response to the status
request key (^O). In reality, each filter would have required
separate input. The PR example shows the two concurrent
filters necessary to display tasks with priorities of 20 or
more, since some might be negative. Then by exploiting
trailing spaces, the nMin series of filters could achieve the
failed ‘9999’ objective discussed above.
‘PR>20’ + ‘!PR=-’ ( 2 for right result )
‘!nMin=0 ’ + ‘!nMin=1 ’ + ‘!nMin=2 ’ + ‘!nMin=3 ’ ...
6a. PERSONAL Configuration File
This file is created or updated via the ‘W’ interactive command.
The legacy version is written as ‘$HOME/.your-name-4-top’ + ‘rc’
with a leading period.
A newly created configuration file is written as
procps/your-name-4-top' + ‘rc’ without a leading period. The
procps directory will be subordinate to either $XDG_CONFIG_HOME
when set as an absolute path or the $HOME/.config directory.
While not intended to be edited manually, here is the general
layout:
global # line 1: the program name/alias notation
" # line 2: id,altscr,irixps,delay,curwin
per ea # line a: winname,fieldscur
window # line b: winflags,sortindx,maxtasks,etc
" # line c: summclr,msgsclr,headclr,taskclr
global # line 15: additional miscellaneous settings
" # any remaining lines are devoted to optional
" # active ‘other filters’ discussed in section 5e above
" # plus ‘inspect’ entries discussed in section 6b below
If a valid absolute path to the rcfile cannot be established,
customizations made to a running top will be impossible to
preserve.
6b. ADDING INSPECT Entries
To exploit the ‘Y’ interactive command, you must add entries at
the end of the top personal configuration file. Such entries
simply reflect a file to be read or command/pipeline to be
executed whose results will then be displayed in a separate
scrollable, searchable window.
If you don't know the location or name of your top rcfile, use
the ‘W’ interactive command to rewrite it and note those details.
Inspect entries can be added with a redirected echo or by editing
the configuration file. Redirecting an echo risks overwriting
the rcfile should it replace (>) rather than append (>>) to that
file. Conversely, when using an editor care must be taken not to
corrupt existing lines, some of which could contain unprintable
data or unusual characters depending on the top version under
which that configuration file was saved.
Those Inspect entries beginning with a ‘#’ character are ignored,
regardless of content. Otherwise they consist of the following 3
elements, each of which must be separated by a tab character
(thus 2 ‘\t’ total):
.type: literal ‘file’ or ‘pipe’
.name: selection shown on the Inspect screen
.fmts: string representing a path or command
The two types of Inspect entries are not interchangeable. Those
designated ‘file’ will be accessed using fopen and must reference
a single file in the ‘.fmts’ element. Entries specifying ‘pipe’
will employ popen, their ‘.fmts’ element could contain many
pipelined commands and, none can be interactive.
If the file or pipeline represented in your ‘.fmts’ deals with
the specific PID input or accepted when prompted, then the format
string must also contain the ‘%d’ specifier, as these examples
illustrate.
.fmts= /proc/%d/numa_maps
.fmts= lsof -P -p %d
For ‘pipe’ type entries only, you may also wish to redirect
stderr to stdout for a more comprehensive result. Thus the
format string becomes:
.fmts= pmap -x %d 2>&1
Here are examples of both types of Inspect entries as they might
appear in the rcfile. The first entry will be ignored due to the
initial ‘#’ character. For clarity, the pseudo tab depictions
(^I) are surrounded by an extra space but the actual tabs would
not be.
# pipe ^I Sockets ^I lsof -n -P -i 2>&1
pipe ^I Open Files ^I lsof -P -p %d 2>&1
file ^I NUMA Info ^I /proc/%d/numa_maps
pipe ^I Log ^I tail -n100 /var/log/syslog | sort -Mr
Except for the commented entry above, these next examples show
what could be echoed to achieve similar results, assuming the
rcfile name was ‘.toprc’. However, due to the embedded tab
characters, each of these lines should be preceded by ‘/bin/echo
-e’, not just a simple an ‘echo’, to enable backslash
interpretation regardless of which shell you use.
"pipe\tOpen Files\tlsof -P -p %d 2>&1" >> ~/.toprc
"file\tNUMA Info\t/proc/%d/numa_maps" >> ~/.toprc
"pipe\tLog\ttail -n200 /var/log/syslog | sort -Mr" >> ~/.toprc
If any inspect entry you create produces output with unprintable
characters they will be displayed in either the ^C notation or
hexadecimal <FF> form, depending on their value. This applies to
tab characters as well, which will show as ‘^I’. If you want a
truer representation, any embedded tabs should be expanded. The
following example takes what could have been a ‘file’ entry but
employs a ‘pipe’ instead so as to expand the embedded tabs.
# next would have contained ‘\t’ ...
# file ^I <your_name> ^I /proc/%d/status
# but this will eliminate embedded ‘\t’ ...
pipe ^I <your_name> ^I cat /proc/%d/status | expand -
Note: Some programs might rely on SIGINT to end. Therefore, if a
‘pipe’ such as the following is established, one must use Ctrl-C
to terminate it in order to review the results. This is the
single occasion where a ‘^C’ will not also terminate top.
pipe ^I Trace ^I /usr/bin/strace -p %d 2>&1
Lastly, while ‘pipe’ type entries have been discussed in terms of
pipelines and commands, there is nothing to prevent you from
including shell scripts as well. Perhaps even newly created
scripts designed specifically for the ‘Y’ interactive command.
For example, as the number of your Inspect entries grows over
time, the ‘Options:’ row will be truncated when screen width is
exceeded. That does not affect operation other than to make some
selections invisible. However, if some choices are lost to
truncation but you want to see more options, there is an easy
solution hinted at below.
Inspection Pause at pid ...
Use: left/right then <Enter> ...
Options: help 1 2 3 4 5 6 7 8 9 10 11 ...
The entries in the top rcfile would have a number for the ‘.name’
element and the ‘help’ entry would identify a shell script you've
written explaining what those numbered selections actually mean.
In that way, many more choices can be made visible.
6c. SYSTEM Configuration File
This configuration file represents defaults for users who have
not saved their own configuration file. The format mirrors
exactly the personal configuration file and can also include
‘inspect’ entries as explained above.
Creating it is a simple process.
1. Configure top appropriately for your installation and preserve
that configuration with the ‘W’ interactive command.
2. Add and test any desired ‘inspect’ entries.
3. Copy that configuration file to the /etc/ directory as
‘topdefaultrc’.
6d. SYSTEM Restrictions File
The presence of this file will influence which version of the
help screen is shown to an ordinary user.
More importantly, it will limit what ordinary users are allowed
to do when top is running. They will not be able to issue the
following commands.
k Kill a task
r Renice a task
d or s Change delay/sleep interval
This configuration file is not created by top. Rather, it is
created manually and placed it in the /etc/ directory as ‘toprc’.
It should have exactly two lines, as shown in this example:
s # line 1: secure mode switch
5.0 # line 2: delay interval in seconds
The value set for the following is unimportant, just its
presence.
LIBPROC_HIDE_KERNEL
This will prevent display of any kernel threads and
exclude such processes from the summary area Tasks/Threads
counts.
Many of these tricks work best when you give top a scheduling
boost. So plan on starting him with a nice value of -10,
assuming you've got the authority.
7a. Kernel Magic
For these stupid tricks, top needs full-screen mode.
• The user interface, through prompts and help, intentionally
implies that the delay interval is limited to tenths of a
second. However, you're free to set any desired delay. If
you want to see Linux at his scheduling best, try a delay of
.09 seconds or less.
For this experiment, under x-windows open an xterm and
maximize it. Then do the following:
. provide a scheduling boost and tiny delay via:
nice -n -10 top -d.09
. keep sorted column highlighting Off so as to
minimize path length
. turn On reverse row highlighting for emphasis
. try various sort columns (TIME/MEM work well),
and normal or reverse sorts to bring the most
active processes into view
What you'll see is a very busy Linux doing what he's always
done for you, but there was no program available to illustrate
this.
• Under an xterm using ‘white-on-black’ colors, on top's Color
Mapping screen set the task color to black and be sure that
task highlighting is set to bold, not reverse. Then set the
delay interval to around .3 seconds.
After bringing the most active processes into view, what
you'll see are the ghostly images of just the currently
running tasks.
• Delete the existing rcfile, or create a new symlink. Start
this new version then type ‘T’ (a secret key, see topic 4c.
Task Area Commands, SORTING) followed by ‘W’ and ‘q’.
Finally, restart the program with -d0 (zero delay).
Your display will be refreshed at three times the rate of the
former top, a 300% speed advantage. As top climbs the TIME
ladder, be as patient as you can while speculating on whether
or not top will ever reach the top.
7b. Bouncing Windows
For these stupid tricks, top needs alternate-display mode.
• With 3 or 4 task displays visible, pick any window other than
the last and turn idle processes Off using the ‘i’ command
toggle. Depending on where you applied ‘i’, sometimes several
task displays are bouncing and sometimes it's like an
accordion, as top tries his best to allocate space.
• Set each window's summary lines differently: one with no
memory (‘m’); another with no states (‘t’); maybe one with
nothing at all, just the message line. Then hold down ‘a’ or
‘w’ and watch a variation on bouncing windows -- hopping
windows.
• Display all 4 windows and for each, in turn, set idle
processes to Off using the ‘i’ command toggle. You've just
entered the "extreme bounce" zone.
7c. The Big Bird Window
This stupid trick also requires alternate-display mode.
• Display all 4 windows and make sure that 1:Def is the
‘current’ window. Then, keep increasing window size with the
‘n’ interactive command until all the other task displays are
"pushed out of the nest".
When they've all been displaced, toggle between all
visible/invisible windows using the ‘_’ command toggle. Then
ponder this:
is top fibbing or telling honestly your imposed truth?
7d. The Ol' Switcheroo
This stupid trick works best without alternate-display mode,
since justification is active on a per window basis.
• Start top and make COMMAND the last (rightmost) column
displayed. If necessary, use the ‘c’ command toggle to
display command lines and ensure that forest view mode is
active with the ‘V’ command toggle.
Then use the up/down arrow keys to position the display so
that some truncated command lines are shown (‘+’ in last
position). You may have to resize your xterm to produce
truncation.
Lastly, use the ‘j’ command toggle to make the COMMAND column
right justified.
Now use the right arrow key to reach the COMMAND column.
Continuing with the right arrow key, watch closely the
direction of travel for the command lines being shown.
some lines travel left, while others travel right
eventually all lines will Switcheroo, and move right
Please send bug reports to ⟨procps@freelists.org⟩.
free(1), ps(1), uptime(1), atop(1), slabtop(1), vmstat(8), w(1)
This page is part of the procps-ng (/proc filesystem utilities)
project. Information about the project can be found at
⟨https://gitlab.com/procps-ng/procps⟩. If you have a bug report
for this manual page, see
⟨https://gitlab.com/procps-ng/procps/blob/master/Documentation/bugs.md⟩.
This page was obtained from the project's upstream Git repository
⟨https://gitlab.com/procps-ng/procps.git⟩ on 2024-06-14. (At
that time, the date of the most recent commit that was found in
the repository was 2024-06-04.) If you discover any rendering
problems in this HTML version of the page, or you believe there
is a better or more up-to-date source for the page, or you have
corrections or improvements to the information in this COLOPHON
(which is not part of the original manual page), send a mail to
man-pages@man7.org
procps-ng March 2024 TOP(1)
Pages that refer to this page: free(1), htop(1), irqtop(1), pidstat(1), pmie(1), ps(1), pstree(1), slabtop(1), systemd-cgtop(1), tload(1), uptime(1), w(1), proc_sys_vm(5), sched(7), iotop(8), vmstat(8)
|
HTML rendering created 2024-06-26 by Michael Kerrisk, author of The Linux Programming Interface. For details of in-depth Linux/UNIX system programming training courses that I teach, look here. Hosting by jambit GmbH. |
|