strace.c: support simultaneous usage of -r and -t options

[strace] / strace.1.in

.\" Copyright (c) 1991, 1992 Paul Kranenburg <pk@cs.few.eur.nl>
.\" Copyright (c) 1993 Branko Lankester <branko@hacktic.nl>
.\" Copyright (c) 1993, 1994, 1995, 1996 Rick Sladkey <jrs@world.std.com>
.\" Copyright (c) 1996-2017 The strace developers.
.\" All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\" 3. The name of the author may not be used to endorse or promote products
.\"    derived from this software without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
.\" IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
.\" OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
.\" IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
.\" INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
.\" NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
.\" DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
.\" THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
.\" (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
.\" THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
.de CW
.sp
.in +4n
.nf
.ft CW
..
.de CE
.ft R
.fi
.in
.sp
..
.\" Like .OP, but with ellipsis at the end in order to signify that option
.\" can be provided multiple times. Based on .OP definition in groff's
.\" an-ext.tmac.
.de OM
.  ie \\n(.$-1 \
.    RI "[\fB\\$1\fP" "\ \\$2" "]...\&"
.  el \
.    RB "[" "\\$1" "]...\&"
..
.\" Required option.
.de OR
.  ie \\n(.$-1 \
.    RI "\fB\\$1\fP" "\ \\$2"
.  el \
.    BR "\\$1"
..
.TH STRACE 1 "@MANPAGE_DATE@" "strace @VERSION@"
.SH NAME
strace \- trace system calls and signals
.SH SYNOPSIS
.SY strace
.OP \-CdffhikqrtttTvVxxy
.OP \-I n
.OP \-b execve
.OM \-e expr
.OP \-a column
.OP \-o file
.OP \-s strsize
.OM \-P path
.OM \-p pid
.BR "" {
.OR \-p pid
.BR "" |
.OP \-D
.OM \-E var\fR[=\fIval\fR]
.OP \-u username
.IR command " [" args ]
.BR "" }
.YS
.SY strace
.B \-c
.OP \-df
.OP \-I n
.OP \-b execve
.OM \-e expr
.OP \-O overhead
.OP \-S sortby
.OM \-P path
.OM \-p pid
.BR "" {
.OR \-p pid
.BR "" |
.OP \-D
.OM \-E var\fR[=\fIval\fR]
.OP -u username
.IR command " [" args ]
.BR "" }
.YS

.SH DESCRIPTION
.IX "strace command" "" "\fLstrace\fR command"
.LP
In the simplest case
.B strace
runs the specified
.I command
until it exits.
It intercepts and records the system calls which are called
by a process and the signals which are received by a process.
The name of each system call, its arguments and its return value
are printed on standard error or to the file specified with the
.B \-o
option.
.LP
.B strace
is a useful diagnostic, instructional, and debugging tool.
System administrators, diagnosticians and trouble-shooters will find
it invaluable for solving problems with
programs for which the source is not readily available since
they do not need to be recompiled in order to trace them.
Students, hackers and the overly-curious will find that
a great deal can be learned about a system and its system calls by
tracing even ordinary programs.  And programmers will find that
since system calls and signals are events that happen at the user/kernel
interface, a close examination of this boundary is very
useful for bug isolation, sanity checking and
attempting to capture race conditions.
.LP
Each line in the trace contains the system call name, followed
by its arguments in parentheses and its return value.
An example from stracing the command "cat /dev/null" is:
.CW
open("/dev/null", O_RDONLY) = 3
.CE
Errors (typically a return value of \-1) have the errno symbol
and error string appended.
.CW
open("/foo/bar", O_RDONLY) = \-1 ENOENT (No such file or directory)
.CE
Signals are printed as signal symbol and decoded siginfo structure.
An excerpt from stracing and interrupting the command "sleep 666" is:
.CW
sigsuspend([] <unfinished ...>
--- SIGINT {si_signo=SIGINT, si_code=SI_USER, si_pid=...} ---
+++ killed by SIGINT +++
.CE
If a system call is being executed and meanwhile another one is being called
from a different thread/process then
.B strace
will try to preserve the order of those events and mark the ongoing call as
being
.IR unfinished .
When the call returns it will be marked as
.IR resumed .
.CW
[pid 28772] select(4, [3], NULL, NULL, NULL <unfinished ...>
[pid 28779] clock_gettime(CLOCK_REALTIME, {1130322148, 939977000}) = 0
[pid 28772] <... select resumed> )      = 1 (in [3])
.CE
Interruption of a (restartable) system call by a signal delivery is processed
differently as kernel terminates the system call and also arranges its
immediate reexecution after the signal handler completes.
.CW
read(0, 0x7ffff72cf5cf, 1)              = ? ERESTARTSYS (To be restarted)
--- SIGALRM ... ---
rt_sigreturn(0xe)                       = 0
read(0, "", 1)                          = 0
.CE
Arguments are printed in symbolic form with passion.
This example shows the shell performing ">>xyzzy" output redirection:
.CW
open("xyzzy", O_WRONLY|O_APPEND|O_CREAT, 0666) = 3
.CE
Here, the third argument of open is decoded by breaking down the
flag argument into its three bitwise-OR constituents and printing the
mode value in octal by tradition.  Where the traditional or native
usage differs from ANSI or POSIX, the latter forms are preferred.
In some cases,
.B strace
output is proven to be more readable than the source.
.LP
Structure pointers are dereferenced and the members are displayed
as appropriate.  In most cases, arguments are formatted in the most C-like
fashion possible.
For example, the essence of the command "ls \-l /dev/null" is captured as:
.CW
lstat("/dev/null", {st_mode=S_IFCHR|0666, st_rdev=makedev(1, 3), ...}) = 0
.CE
Notice how the 'struct stat' argument is dereferenced and how each member is
displayed symbolically.  In particular, observe how the
.B st_mode
member is carefully decoded into a bitwise-OR of symbolic and numeric values.
Also notice in this example that the first argument to
.B lstat
is an input to the system call and the second argument is an output.
Since output arguments are not modified if the system call fails, arguments may
not always be dereferenced.  For example, retrying the "ls \-l" example
with a non-existent file produces the following line:
.CW
lstat("/foo/bar", 0xb004) = \-1 ENOENT (No such file or directory)
.CE
In this case the porch light is on but nobody is home.
.LP
Syscalls unknown to
.B strace
are printed raw, with the unknown system call number printed in hexadecimal form
and prefixed with "syscall_":
.CW
syscall_0xbad(0xfedcba9876543210, 0xfedcba9876543211, 0xfedcba9876543212,
0xfedcba9876543213, 0xfedcba9876543214, 0xfedcba9876543215) = -1 (errno 38)
.CE
.LP
Character pointers are dereferenced and printed as C strings.
Non-printing characters in strings are normally represented by
ordinary C escape codes.
Only the first
.I strsize
(32 by default) bytes of strings are printed;
longer strings have an ellipsis appended following the closing quote.
Here is a line from "ls \-l" where the
.B getpwuid
library routine is reading the password file:
.CW
read(3, "root::0:0:System Administrator:/"..., 1024) = 422
.CE
While structures are annotated using curly braces, simple pointers
and arrays are printed using square brackets with commas separating
elements.  Here is an example from the command "id" on a system with
supplementary group ids:
.CW
getgroups(32, [100, 0]) = 2
.CE
On the other hand, bit-sets are also shown using square brackets
but set elements are separated only by a space.  Here is the shell,
preparing to execute an external command:
.CW
sigprocmask(SIG_BLOCK, [CHLD TTOU], []) = 0
.CE
Here, the second argument is a bit-set of two signals,
.BR SIGCHLD " and " SIGTTOU .
In some cases, the bit-set is so full that printing out the unset
elements is more valuable.  In that case, the bit-set is prefixed by
a tilde like this:
.CW
sigprocmask(SIG_UNBLOCK, ~[], NULL) = 0
.CE
Here, the second argument represents the full set of all signals.
.SH OPTIONS
.SS Output format
.TP 12
.BI "\-a " column
Align return values in a specific column (default column 40).
.TP
.B \-i
Print the instruction pointer at the time of the system call.
.TP
.B \-k
Print the execution stack trace of the traced processes after each system call (experimental).
This option is available only if
.B strace
is built with libunwind.
.TP
.BI "\-o " filename
Write the trace output to the file
.I filename
rather than to stderr.
.IR filename . pid
form is used if
.B \-ff
option is supplied.
If the argument begins with '|' or '!', the rest of the
argument is treated as a command and all output is piped to it.
This is convenient for piping the debugging output to a program
without affecting the redirections of executed programs.
The latter is not compatible with
.B \-ff
option currently.
.TP
.B \-q
Suppress messages about attaching, detaching etc.  This happens
automatically when output is redirected to a file and the command
is run directly instead of attaching.
.TP
.B \-qq
If given twice, suppress messages about process exit status.
.TP
.B \-r
Print a relative timestamp upon entry to each system call.  This
records the time difference between the beginning of successive
system calls.
Note that since
.B \-r
option uses the monotonic clock time for measuring time difference and not the
wall clock time, its measurements can differ from the difference in time
reported by the
.B \-t
option.
.TP
.BI "\-s " strsize
Specify the maximum string size to print (the default is 32).  Note
that filenames are not considered strings and are always printed in
full.
.TP
.B \-t
Prefix each line of the trace with the wall clock time.
.TP
.B \-tt
If given twice, the time printed will include the microseconds.
.TP
.B \-ttt
If given thrice, the time printed will include the microseconds
and the leading portion will be printed as the number
of seconds since the epoch.
.TP
.B \-T
Show the time spent in system calls.  This records the time
difference between the beginning and the end of each system call.
.TP
.B \-x
Print all non-ASCII strings in hexadecimal string format.
.TP
.B \-xx
Print all strings in hexadecimal string format.
.TP
.B \-y
Print paths associated with file descriptor arguments.
.TP
.B \-yy
Print protocol specific information associated with socket file descriptors,
and block/character device number associated with device file descriptors.
.SS Statistics
.TP 12
.B \-c
Count time, calls, and errors for each system call and report a summary on
program exit, suppressing the regular output.
This attempts to show system time (CPU time spent running
in the kernel) independent of wall clock time.  If
.B \-c
is used with
.BR \-f ,
only aggregate totals for all traced processes are kept.
.TP
.B \-C
Like
.B \-c
but also print regular output while processes are running.
.TP
.BI "\-O " overhead
Set the overhead for tracing system calls to
.I overhead
microseconds.
This is useful for overriding the default heuristic for guessing
how much time is spent in mere measuring when timing system calls using
the
.B \-c
option.  The accuracy of the heuristic can be gauged by timing a given
program run without tracing (using
.BR time (1))
and comparing the accumulated
system call time to the total produced using
.BR \-c .
.TP
.BI "\-S " sortby
Sort the output of the histogram printed by the
.B \-c
option by the specified criterion.  Legal values are
.BR time ,
.BR calls ,
.BR name ,
and
.B nothing
(default is
.BR time ).
.TP
.B \-w
Summarise the time difference between the beginning and end of
each system call.  The default is to summarise the system time.
.SS Filtering
.TP 12
.BI "\-e " expr
A qualifying expression which modifies which events to trace
or how to trace them.  The format of the expression is:
.RS 15
.IP
[\,\fIqualifier\/\fB=\fR][\fB!\fR][\fB?\fR]\,\fIvalue1\/\fR[\fB,\fR[\fB?\fR]\,\fIvalue2\/\fR]...
.RE
.IP
where
.I qualifier
is one of
.BR trace ,
.BR abbrev ,
.BR verbose ,
.BR raw ,
.BR signal ,
.BR read ,
.BR write ,
.BR fault ,
or
.B inject
and
.I value
is a qualifier-dependent symbol or number.  The default
qualifier is
.BR trace .
Using an exclamation mark negates the set of values.  For example,
.BR \-e "\ " open
means literally
.BR \-e "\ " trace = open
which in turn means trace only the
.B open
system call.  By contrast,
.BR \-e "\ " trace "=!" open
means to trace every system call except
.BR open .
Question mark before the syscall qualification allows suppression of error
in case no syscalls matched the qualification provided.
In addition, the special values
.B all
and
.B none
have the obvious meanings.
.IP
Note that some shells use the exclamation point for history
expansion even inside quoted arguments.  If so, you must escape
the exclamation point with a backslash.
.TP
\fB\-e\ trace\fR=\,\fIset\fR
Trace only the specified set of system calls.  The
.B \-c
option is useful for determining which system calls might be useful
to trace.  For example,
.BR trace = open,close,read,write
means to only
trace those four system calls.  Be careful when making inferences
about the user/kernel boundary if only a subset of system calls
are being monitored.  The default is
.BR trace = all .
.TP
\fB\-e\ trace\fR=/\,\fIregex\fR
Trace only those system calls that match the
.IR regex .
You can use
.B POSIX
Extended Regular Expression syntax (see
.BR regex (7)).
.TP
.BR "\-e\ trace" = %file
.TQ
.BR "\-e\ trace" = file " (deprecated)"
Trace all system calls which take a file name as an argument.  You
can think of this as an abbreviation for
.BR "\-e\ trace" = open , stat , chmod , unlink ,...
which is useful to seeing what files the process is referencing.
Furthermore, using the abbreviation will ensure that you don't
accidentally forget to include a call like
.B lstat
in the list.  Betchya woulda forgot that one.
.TP
.BR "\-e\ trace" = %process
.TQ
.BR "\-e\ trace" = process " (deprecated)"
Trace all system calls which involve process management.  This
is useful for watching the fork, wait, and exec steps of a process.
.TP
.BR "\-e\ trace" = %network
.TQ
.BR "\-e\ trace" = network " (deprecated)"
Trace all the network related system calls.
.TP
.BR "\-e\ trace" = %signal
.TQ
.BR "\-e\ trace" = signal " (deprecated)"
Trace all signal related system calls.
.TP
.BR "\-e\ trace" = %ipc
.TQ
.BR "\-e\ trace" = ipc " (deprecated)"
Trace all IPC related system calls.
.TP
.BR "\-e\ trace" = %desc
.TQ
.BR "\-e\ trace" = desc " (deprecated)"
Trace all file descriptor related system calls.
.TP
.BR "\-e\ trace" = %memory
.TQ
.BR "\-e\ trace" = memory " (deprecated)"
Trace all memory mapping related system calls.
.TP
.BR "\-e\ trace" = %stat
Trace stat syscall variants.
.TP
.BR "\-e\ trace" = %lstat
Trace lstat syscall variants.
.TP
.BR "\-e\ trace" = %fstat
Trace fstat and fstatat syscall variants.
.TP
.BR "\-e\ trace" = %%stat
Trace syscalls used for requesting file status (stat, lstat, fstat, fstatat,
statx, and their variants).
.TP
.BR "\-e\ trace" = %statfs
Trace statfs, statfs64, statvfs, osf_statfs, and osf_statfs64 system calls.
The same effect can be achieved with
.BR "\-e\ trace" = /^(.*_)?statv?fs
regular expression.
.TP
.BR "\-e\ trace" = %fstatfs
Trace fstatfs, fstatfs64, fstatvfs, osf_fstatfs, and osf_fstatfs64 system calls.
The same effect can be achieved with
.BR "\-e\ trace" = /fstatv?fs
regular expression.
.TP
.BR "\-e\ trace" = %%statfs
Trace syscalls related to file system statistics (statfs-like, fstatfs-like,
and ustat).  The same effect can be achieved with
.BR "\-e\ trace" = /statv?fs|fsstat|ustat
regular expression.
.TP
.BR "\-e\ trace" = %pure
Trace syscalls that always succeed and have no arguments.
Currently, this list includes
.BR arc_gettls "(2), " getdtablesize "(2), " getegid "(2), " getegid32 "(2),"
.BR geteuid "(2), " geteuid32 "(2), " getgid "(2), " getgid32 "(2),"
.BR getpagesize "(2), " getpgrp "(2), " getpid "(2), " getppid "(2),"
.BR get_thread_area (2)
(on architectures other than x86),
.BR gettid "(2), " get_tls "(2), " getuid "(2), " getuid32 "(2),"
.BR getxgid "(2), " getxpid "(2), " getxuid "(2), " kern_features "(2), and"
.BR metag_get_tls "(2)"
syscalls.
.TP
\fB\-e\ abbrev\fR=\,\fIset\fR
Abbreviate the output from printing each member of large structures.
The default is
.BR abbrev = all .
The
.B \-v
option has the effect of
.BR abbrev = none .
.TP
\fB\-e\ verbose\fR=\,\fIset\fR
Dereference structures for the specified set of system calls.  The
default is
.BR verbose = all .
.TP
\fB\-e\ raw\fR=\,\fIset\fR
Print raw, undecoded arguments for the specified set of system calls.
This option has the effect of causing all arguments to be printed
in hexadecimal.  This is mostly useful if you don't trust the
decoding or you need to know the actual numeric value of an
argument.
.TP
\fB\-e\ signal\fR=\,\fIset\fR
Trace only the specified subset of signals.  The default is
.BR signal = all .
For example,
.BR signal "=!" SIGIO
(or
.BR signal "=!" io )
causes
.B SIGIO
signals not to be traced.
.TP
\fB\-e\ read\fR=\,\fIset\fR
Perform a full hexadecimal and ASCII dump of all the data read from
file descriptors listed in the specified set.  For example, to see
all input activity on file descriptors
.I 3
and
.I 5
use
\fB\-e\ read\fR=\,\fI3\fR,\fI5\fR.
Note that this is independent from the normal tracing of the
.BR read (2)
system call which is controlled by the option
.BR -e "\ " trace = read .
.TP
\fB\-e\ write\fR=\,\fIset\fR
Perform a full hexadecimal and ASCII dump of all the data written to
file descriptors listed in the specified set.  For example, to see
all output activity on file descriptors
.I 3
and
.I 5
use
\fB\-e\ write\fR=\,\fI3\fR,\,\fI5\fR.
Note that this is independent from the normal tracing of the
.BR write (2)
system call which is controlled by the option
.BR -e "\ " trace = write .
.TP
\fB\-e\ inject\fR=\,\fIset\/\fR[:\fBerror\fR=\,\fIerrno\/\fR|:\fBretval\fR=\,\fIvalue\/\fR][:\fBsignal\fR=\,\fIsig\/\fR][:\fBdelay_enter\fR=\,\fIusecs\/\fR][:\fBdelay_exit\fR=\,\fIusecs\/\fR][:\fBwhen\fR=\,\fIexpr\/\fR]
Perform syscall tampering for the specified set of syscalls.

At least one of
.BR error ,
.BR retval ,
.BR signal ,
.BR delay_enter ,
or
.B delay_exit
options has to be specified.
.B error
and
.B retval
are mutually exclusive.

If :\fBerror\fR=\,\fIerrno\/\fR option is specified,
a fault is injected into a syscall invocation:
the syscall number is replaced by -1 which corresponds to an invalid syscall,
and the error code is specified using a symbolic
.I errno
value like
.B ENOSYS
or a numeric value within 1..4095 range.

If :\fBretval\fR=\,\fIvalue\/\fR option is specified,
success injection is performed: the syscall number is replaced by -1,
but a bogus success value is returned to the callee.

If :\fBsignal\fR=\,\fIsig\/\fR option is specified with either a symbolic value
like
.B SIGSEGV
or a numeric value within 1..\fBSIGRTMAX\fR range,
that signal is delivered on entering every syscall specified by the
.IR set .

If :\fBdelay_enter\fR=\,\fIusecs\/\fR or :\fBdelay_exit\fR=\,\fIusecs\/\fR
options are specified, delay injection is performed: the tracee is delayed
by at least
.IR usecs
microseconds on entering or exiting the syscall.

If :\fBsignal\fR=\,\fIsig\/\fR option is specified without
:\fBerror\fR=\,\fIerrno\/\fR, :\fBretval\fR=\,\fIvalue\/\fR or
:\fBdelay_{enter,exit}\fR=\,\fIusecs\/\fR options,
then only a signal
.I sig
is delivered without a syscall fault or delay injection.
Conversely, :\fBerror\fR=\,\fIerrno\/\fR or
:\fBretval\fR=\,\fIvalue\/\fR option without
:\fBdelay_enter\fR=\,\fIusecs\/\fR,
:\fBdelay_exit\fR=\,\fIusecs\/\fR or
:\fBsignal\fR=\,\fIsig\/\fR options injects a fault without delivering a signal
or injecting a delay, etc.

If both :\fBerror\fR=\,\fIerrno\/\fR or :\fBretval\fR=\,\fIvalue\/\fR
and :\fBsignal\fR=\,\fIsig\/\fR options are specified, then both
a fault or success is injected and a signal is delivered.

Unless a :\fBwhen\fR=\,\fIexpr\fR subexpression is specified,
an injection is being made into every invocation of each syscall from the
.IR set .

The format of the subexpression is one of the following:
.RS
.IP "" 2
.I first
.RS 4
For every syscall from the
.IR set ,
perform an injection for the syscall invocation number
.I first
only.
.RE
.IP "" 2
\fIfirst\/\fB+\fR
.RS 4
For every syscall from the
.IR set ,
perform injections for the syscall invocation number
.I first
and all subsequent invocations.
.RE
.IP "" 2
\fIfirst\/\fB+\fIstep\fR
.RS 4
For every syscall from the
.IR set ,
perform injections for syscall invocations number
.IR first ,
.IR first + step ,
.IR first + step + step ,
and so on.
.RE
.RE
.IP
For example, to fail each third and subsequent chdir syscalls with
.BR ENOENT ,
use
\fB\-e\ inject\fR=\,\fIchdir\/\fR:\fBerror\fR=\,\fIENOENT\/\fR:\fBwhen\fR=\,\fI3\/\fB+\fR.

The valid range for numbers
.I first
and
.I step
is 1..65535.

An injection expression can contain only one
.BR error =
or
.BR retval =
specification, and only one
.BR signal =
specification.  If an injection expression contains multiple
.BR when =
specifications, the last one takes precedence.

Accounting of syscalls that are subject to injection
is done per syscall and per tracee.

Specification of syscall injection can be combined
with other syscall filtering options, for example,
\fB\-P \fI/dev/urandom \fB\-e inject\fR=\,\fIfile\/\fR:\fBerror\fR=\,\fIENOENT\fR.

.TP
\fB\-e\ fault\fR=\,\fIset\/\fR[:\fBerror\fR=\,\fIerrno\/\fR][:\fBwhen\fR=\,\fIexpr\/\fR]
Perform syscall fault injection for the specified set of syscalls.

This is equivalent to more generic
\fB\-e\ inject\fR= expression with default value of
.I errno
option set to
.IR ENOSYS .

.TP
.BI "\-P " path
Trace only system calls accessing
.IR path .
Multiple
.B \-P
options can be used to specify several paths.
.TP
.B \-v
Print unabbreviated versions of environment, stat, termios, etc.
calls.  These structures are very common in calls and so the default
behavior displays a reasonable subset of structure members.  Use
this option to get all of the gory details.
.SS Tracing
.TP 12
.BI "\-b " syscall
If specified syscall is reached, detach from traced process.
Currently, only
.I execve
syscall is supported.  This option is useful if you want to trace
multi-threaded process and therefore require -f, but don't want
to trace its (potentially very complex) children.
.TP
.B \-D
Run tracer process as a detached grandchild, not as parent of the
tracee.  This reduces the visible effect of
.B strace
by keeping the tracee a direct child of the calling process.
.TP
.B \-f
Trace child processes as they are created by currently traced
processes as a result of the
.BR fork (2),
.BR vfork (2)
and
.BR clone (2)
system calls.  Note that
.B \-p
.I PID
.B \-f
will attach all threads of process PID if it is multi-threaded,
not only thread with thread_id = PID.
.TP
.B \-ff
If the
.B \-o
.I filename
option is in effect, each processes trace is written to
.I filename.pid
where pid is the numeric process id of each process.
This is incompatible with
.BR \-c ,
since no per-process counts are kept.

One might want to consider using
.BR strace-log-merge (1)
to obtain a combined strace log view.
.TP
.BI "\-I " interruptible
When strace can be interrupted by signals (such as pressing ^C).
1: no signals are blocked; 2: fatal signals are blocked while decoding syscall
(default); 3: fatal signals are always blocked (default if '-o FILE PROG');
4: fatal signals and SIGTSTP (^Z) are always blocked (useful to make
strace -o FILE PROG not stop on ^Z).
.SS Startup
.TP 12
\fB\-E\ \fIvar\fR=\,\fIval\fR
Run command with
.IR var = val
in its list of environment variables.
.TP
.BI "\-E " var
Remove
.IR var
from the inherited list of environment variables before passing it on to
the command.
.TP
.BI "\-p " pid
Attach to the process with the process
.SM ID
.I pid
and begin tracing.
The trace may be terminated
at any time by a keyboard interrupt signal (\c
.SM CTRL\s0-C).
.B strace
will respond by detaching itself from the traced process(es)
leaving it (them) to continue running.
Multiple
.B \-p
options can be used to attach to many processes in addition to
.I command
(which is optional if at least one
.B \-p
option is given).
.B \-p
"`pidof PROG`" syntax is supported.
.TP
.BI "\-u " username
Run command with the user \s-1ID\s0, group \s-2ID\s0, and
supplementary groups of
.IR username .
This option is only useful when running as root and enables the
correct execution of setuid and/or setgid binaries.
Unless this option is used setuid and setgid programs are executed
without effective privileges.
.SS Miscellaneous
.TP 12
.B \-d
Show some debugging output of
.B strace
itself on the standard error.
.TP
.B \-F
This option is deprecated.  It is retained for backward compatibility only
and may be removed in future releases.
Usage of multiple instances of
.B \-F
option is still equivalent to a single
.BR \-f ,
and it is ignored at all if used along with one or more instances of
.B \-f
option.
.TP
.B \-h
Print the help summary.
.TP
.B \-V
Print the version number of
.BR strace .
.SH DIAGNOSTICS
When
.I command
exits,
.B strace
exits with the same exit status.
If
.I command
is terminated by a signal,
.B strace
terminates itself with the same signal, so that
.B strace
can be used as a wrapper process transparent to the invoking parent process.
Note that parent-child relationship (signal stop notifications,
getppid() value, etc) between traced process and its parent are not preserved
unless
.B \-D
is used.
.LP
When using
.B \-p
without a
.IR command ,
the exit status of
.B strace
is zero unless no processes has been attached or there was an unexpected error
in doing the tracing.
.SH "SETUID INSTALLATION"
If
.B strace
is installed setuid to root then the invoking user will be able to
attach to and trace processes owned by any user.
In addition setuid and setgid programs will be executed and traced
with the correct effective privileges.
Since only users trusted with full root privileges should be allowed
to do these things,
it only makes sense to install
.B strace
as setuid to root when the users who can execute it are restricted
to those users who have this trust.
For example, it makes sense to install a special version of
.B strace
with mode 'rwsr-xr--', user
.B root
and group
.BR trace ,
where members of the
.B trace
group are trusted users.
If you do use this feature, please remember to install
a regular non-setuid version of
.B strace
for ordinary users to use.
.SH "MULTIPLE PERSONALITY SUPPORT"
On some architectures,
.B strace
supports decoding of syscalls for processes that use different ABI rather than
the one
.B strace
uses.
Specifically, in addition to decoding native ABI,
.B strace
can decode the following ABIs on the following architectures:
.TS H
allbox;
lb lb
l l.
Architecture    ABIs supported
x86_64  i386, x32 (when built as an x86_64 application); i386 (when built as an x32 application)
AArch64 ARM 32-bit EABI
PowerPC 64-bit  PowerPC 32-bit
RISC-V 64-bit   RISC-V 32-bit
s390x   s390
SPARC 64-bit    SPARC 32-bit
TILE 64-bit     TILE 32-bit
.TE
.PP
This support is optional and relies on ability to generate and parse structure
definitions during the build time.
Please refer to the output of the
.B strace \-V
command in order to figure out what support is available in your strace build
("non-native" refers to an ABI that differs from the ABI strace has):
.TP 15
.B m32-mpers
.B strace
can trace and properly decode non-native 32-bit binaries.
.TP
.B no-m32-mpers
.B strace
can trace, but cannot properly decode non-native 32-bit binaries.
.TP
.B mx32-mpers
.B strace
can trace and properly decode non-native 32-on-64-bit binaries.
.TP
.B no-mx32-mpers
.B strace
can trace, but cannot properly decode non-native 32-on-64-bit binaries.
.PP
If the output contains neither
.B m32-mpers
nor
.BR no-m32-mpers ,
then decoding of non-native 32-bit binaries is not implemented at all
or not applicable.
.PP
Likewise, if the output contains neither
.B mx32-mpers
nor
.BR no-mx32-mpers ,
then decoding of non-native 32-on-64-bit binaries is not implemented at all
or not applicable.
.SH NOTES
It is a pity that so much tracing clutter is produced by systems
employing shared libraries.
.LP
It is instructive to think about system call inputs and outputs
as data-flow across the user/kernel boundary.  Because user-space
and kernel-space are separate and address-protected, it is
sometimes possible to make deductive inferences about process
behavior using inputs and outputs as propositions.
.LP
In some cases, a system call will differ from the documented behavior
or have a different name.  For example, the
.BR faccessat (2)
system call does not have
.I flags
argument, and the
.BR setrlimit (2)
library function uses
.BR prlimit64 (2)
system call on modern (2.6.38+) kernels.  These
discrepancies are normal but idiosyncratic characteristics of the
system call interface and are accounted for by C library wrapper
functions.
.LP
Some system calls have different names in different architectures and
personalities.  In these cases, system call filtering and printing
uses the names that match corresponding
.BR __NR_ *
kernel macros of the tracee's architecture and personality.
There are two exceptions from this general rule:
.BR arm_fadvise64_64 (2)
ARM syscall and
.BR xtensa_fadvise64_64 (2)
Xtensa syscall are filtered and printed as
.BR fadvise64_64 (2).
.LP
On some platforms a process that is attached to with the
.B \-p
option may observe a spurious EINTR return from the current
system call that is not restartable.  (Ideally, all system calls
should be restarted on strace attach, making the attach invisible
to the traced process, but a few system calls aren't.
Arguably, every instance of such behavior is a kernel bug.)
This may have an unpredictable effect on the process
if the process takes no action to restart the system call.
.LP
As
.B strace
executes the specified
.I command
directly and does not employ a shell for that, scripts without shebang
that usually run just fine when invoked by shell fail to execute with
.B ENOEXEC
error.
It is advisable to manually supply a shell as a
.I command
with the script as its argument.
.SH BUGS
Programs that use the
.I setuid
bit do not have
effective user
.SM ID
privileges while being traced.
.LP
A traced process runs slowly.
.LP
Traced processes which are descended from
.I command
may be left running after an interrupt signal (\c
.SM CTRL\s0-C).
.SH HISTORY
The original
.B strace
was written by Paul Kranenburg
for SunOS and was inspired by its
.B trace
utility.
The SunOS version of
.B strace
was ported to Linux and enhanced
by Branko Lankester, who also wrote the Linux kernel support.
Even though Paul released
.B strace
2.5 in 1992,
Branko's work was based on Paul's
.B strace
1.5 release from 1991.
In 1993, Rick Sladkey merged
.B strace
2.5 for SunOS and the second release of
.B strace
for Linux, added many of the features of
.BR truss (1)
from SVR4, and produced an
.B strace
that worked on both platforms.  In 1994 Rick ported
.B strace
to SVR4 and Solaris and wrote the
automatic configuration support.  In 1995 he ported
.B strace
to Irix
and tired of writing about himself in the third person.
.PP
Beginning with 1996,
.B strace
was maintained by Wichert Akkerman.
During his tenure,
.B strace
development migrated to CVS; ports to FreeBSD and many architectures on Linux
(including ARM, IA-64, MIPS, PA-RISC, PowerPC, s390, SPARC) were introduced.
In 2002, the burden of
.B strace
maintainership was transferred to Ronald McGrath.
Since then,
.B strace
gained support for several new Linux architectures (AMD64, s390x, SuperH),
bi-architecture support for some of them, and received numerous additions and
improvements in syscalls decoders on Linux;
.B strace
development migrated to
.B git
during that period.
Since 2009,
.B strace
is actively maintained by Dmitry Levin.
.B strace
gained support for AArch64, ARC, AVR32, Blackfin, Meta, Nios II, OpenSISC 1000,
RISC-V, Tile/TileGx, Xtensa architectures since that time.
In 2012, unmaintained and apparently broken support for non-Linux operating
systems was removed.
Also, in 2012
.B strace
gained support for path tracing and file descriptor path decoding.
In 2014, support for stack traces printing was added.
In 2016, syscall fault injection was implemented.
.PP
For the additional information, please refer to the
.B NEWS
file and
.B strace
repository commit log.
.SH REPORTING BUGS
Problems with
.B strace
should be reported to the
.B strace
mailing list at <strace\-devel@lists.strace.io>.
.SH "SEE ALSO"
.BR strace-log-merge (1),
.BR ltrace (1),
.BR perf-trace (1),
.BR trace-cmd (1),
.BR time (1),
.BR ptrace (2),
.BR proc (5)