perf_event_open.2: grfix

[thirdparty/man-pages.git] / man2 / ptrace.2

.\" Copyright (c) 1993 Michael Haardt <michael@moria.de>
.\" Fri Apr  2 11:32:09 MET DST 1993
.\"
.\" and changes Copyright (C) 1999 Mike Coleman (mkc@acm.org)
.\" -- major revision to fully document ptrace semantics per recent Linux
.\"    kernel (2.2.10) and glibc (2.1.2)
.\" Sun Nov  7 03:18:35 CST 1999
.\"
.\" and Copyright (c) 2011, Denys Vlasenko <vda.linux@googlemail.com>
.\"
.\" %%%LICENSE_START(GPLv2+_DOC_FULL)
.\" This is free documentation; you can redistribute it and/or
.\" modify it under the terms of the GNU General Public License as
.\" published by the Free Software Foundation; either version 2 of
.\" the License, or (at your option) any later version.
.\"
.\" The GNU General Public License's references to "object code"
.\" and "executables" are to be interpreted as the output of any
.\" document formatting or typesetting system, including
.\" intermediate and printed output.
.\"
.\" This manual is distributed in the hope that it will be useful,
.\" but WITHOUT ANY WARRANTY; without even the implied warranty of
.\" MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
.\" GNU General Public License for more details.
.\"
.\" You should have received a copy of the GNU General Public
.\" License along with this manual; if not, see
.\" <http://www.gnu.org/licenses/>.
.\" %%%LICENSE_END
.\"
.\" Modified Fri Jul 23 23:47:18 1993 by Rik Faith <faith@cs.unc.edu>
.\" Modified Fri Jan 31 16:46:30 1997 by Eric S. Raymond <esr@thyrsus.com>
.\" Modified Thu Oct  7 17:28:49 1999 by Andries Brouwer <aeb@cwi.nl>
.\" Modified, 27 May 2004, Michael Kerrisk <mtk.manpages@gmail.com>
.\"     Added notes on capability requirements
.\"
.\" 2006-03-24, Chuck Ebbert <76306.1226@compuserve.com>
.\"    Added    PTRACE_SETOPTIONS, PTRACE_GETEVENTMSG, PTRACE_GETSIGINFO,
.\"        PTRACE_SETSIGINFO, PTRACE_SYSEMU, PTRACE_SYSEMU_SINGLESTEP
.\"    (Thanks to Blaisorblade, Daniel Jacobowitz and others who helped.)
.\" 2011-09, major update by Denys Vlasenko <vda.linux@googlemail.com>
.\"
.TH PTRACE 2 2014-02-20 "Linux" "Linux Programmer's Manual"
.SH NAME
ptrace \- process trace
.SH SYNOPSIS
.nf
.B #include <sys/ptrace.h>
.sp
.BI "long ptrace(enum __ptrace_request " request ", pid_t " pid ", "
.BI "            void *" addr ", void *" data );
.fi
.SH DESCRIPTION
The
.BR ptrace ()
system call provides a means by which one process (the "tracer")
may observe and control the execution of another process (the "tracee"),
and examine and change the tracee's memory and registers.
It is primarily used to implement breakpoint debugging and system
call tracing.
.LP
A tracee first needs to be attached to the tracer.
Attachment and subsequent commands are per thread:
in a multithreaded process,
every thread can be individually attached to a
(potentially different) tracer,
or left not attached and thus not debugged.
Therefore, "tracee" always means "(one) thread",
never "a (possibly multithreaded) process".
Ptrace commands are always sent to
a specific tracee using a call of the form

    ptrace(PTRACE_foo, pid, ...)

where
.I pid
is the thread ID of the corresponding Linux thread.
.LP
(Note that in this page, a "multithreaded process"
means a thread group consisting of threads created using the
.BR clone (2)
.B CLONE_THREAD
flag.)
.LP
A process can initiate a trace by calling
.BR fork (2)
and having the resulting child do a
.BR PTRACE_TRACEME ,
followed (typically) by an
.BR execve (2).
Alternatively, one process may commence tracing another process using
.B PTRACE_ATTACH
or
.BR PTRACE_SEIZE .
.LP
While being traced, the tracee will stop each time a signal is delivered,
even if the signal is being ignored.
(An exception is
.BR SIGKILL ,
which has its usual effect.)
The tracer will be notified at its next call to
.BR waitpid (2)
(or one of the related "wait" system calls); that call will return a
.I status
value containing information that indicates
the cause of the stop in the tracee.
While the tracee is stopped,
the tracer can use various ptrace requests to inspect and modify the tracee.
The tracer then causes the tracee to continue,
optionally ignoring the delivered signal
(or even delivering a different signal instead).
.LP
If the
.B PTRACE_O_TRACEEXEC
option is not in effect, all successful calls to
.BR execve (2)
by the traced process will cause it to be sent a
.B SIGTRAP
signal,
giving the parent a chance to gain control before the new program
begins execution.
.LP
When the tracer is finished tracing, it can cause the tracee to continue
executing in a normal, untraced mode via
.BR PTRACE_DETACH .
.LP
The value of
.I request
determines the action to be performed:
.TP
.B PTRACE_TRACEME
Indicate that this process is to be traced by its parent.
A process probably shouldn't make this request if its parent
isn't expecting to trace it.
.RI ( pid ,
.IR addr ,
and
.IR data
are ignored.)
.IP
The
.B PTRACE_TRACEME
request is used only by the tracee;
the remaining requests are used only by the tracer.
In the following requests,
.I pid
specifies the thread ID of the tracee to be acted on.
For requests other than
.BR PTRACE_ATTACH ,
.BR PTRACE_SEIZE ,
.BR PTRACE_INTERRUPT ,
and
.BR PTRACE_KILL ,
the tracee must be stopped.
.TP
.BR PTRACE_PEEKTEXT ", " PTRACE_PEEKDATA
Read a word at the address
.I addr
in the tracee's memory, returning the word as the result of the
.BR ptrace ()
call.
Linux does not have separate text and data address spaces,
so these two requests are currently equivalent.
.RI ( data
is ignored; but see NOTES.)
.TP
.B PTRACE_PEEKUSER
.\" PTRACE_PEEKUSR in kernel source, but glibc uses PTRACE_PEEKUSER,
.\" and that is the name that seems common on other systems.
Read a word at offset
.I addr
in the tracee's USER area,
which holds the registers and other information about the process
(see
.IR <sys/user.h> ).
The word is returned as the result of the
.BR ptrace ()
call.
Typically, the offset must be word-aligned, though this might vary by
architecture.
See NOTES.
.RI ( data
is ignored; but see NOTES.)
.TP
.BR PTRACE_POKETEXT ", " PTRACE_POKEDATA
Copy the word
.I data
to the address
.I addr
in the tracee's memory.
As for
.BR PTRACE_PEEKTEXT
and
.BR PTRACE_PEEKDATA ,
these two requests are currently equivalent.
.TP
.B PTRACE_POKEUSER
.\" PTRACE_POKEUSR in kernel source, but glibc uses PTRACE_POKEUSER,
.\" and that is the name that seems common on other systems.
Copy the word
.I data
to offset
.I addr
in the tracee's USER area.
As for
.BR PTRACE_PEEKUSER ,
the offset must typically be word-aligned.
In order to maintain the integrity of the kernel,
some modifications to the USER area are disallowed.
.\" FIXME In the preceding sentence, which modifications are disallowed,
.\" and when they are disallowed, how does user space discover that fact?
.TP
.BR PTRACE_GETREGS ", " PTRACE_GETFPREGS
Copy the tracee's general-purpose or floating-point registers,
respectively, to the address
.I data
in the tracer.
See
.I <sys/user.h>
for information on the format of this data.
.RI ( addr
is ignored.)
Note that SPARC systems have the meaning of
.I data
and
.I addr
reversed; that is,
.I data
is ignored and the registers are copied to the address
.IR addr .
.B PTRACE_GETREGS
and
.B PTRACE_GETFPREGS
are not present on all architectures.
.TP
.BR PTRACE_GETREGSET " (since Linux 2.6.34)"
Read the tracee's registers.
.I addr
specifies, in an architecture-dependent way, the type of registers to be read.
.B NT_PRSTATUS
(with numerical value 1)
usually results in reading of general-purpose registers.
If the CPU has, for example,
floating-point and/or vector registers, they can be retrieved by setting
.I addr
to the corresponding
.B NT_foo
constant.
.I data
points to a
.BR "struct iovec" ,
which describes the destination buffer's location and length.
On return, the kernel modifies
.B iov.len
to indicate the actual number of bytes returned.
.TP
.BR PTRACE_SETREGS ", " PTRACE_SETFPREGS
Modify the tracee's general-purpose or floating-point registers,
respectively, from the address
.I data
in the tracer.
As for
.BR PTRACE_POKEUSER ,
some general-purpose register modifications may be disallowed.
.\" FIXME In the preceding sentence, which modifications are disallowed,
.\" and when they are disallowed, how does user space discover that fact?
.RI ( addr
is ignored.)
Note that SPARC systems have the meaning of
.I data
and
.I addr
reversed; that is,
.I data
is ignored and the registers are copied from the address
.IR addr .
.B PTRACE_SETREGS
and
.B PTRACE_SETFPREGS
are not present on all architectures.
.TP
.BR PTRACE_SETREGSET " (since Linux 2.6.34)"
Modify the tracee's registers.
The meaning of
.I addr
and
.I data
is analogous to
.BR PTRACE_GETREGSET .
.TP
.BR PTRACE_GETSIGINFO " (since Linux 2.3.99-pre6)"
Retrieve information about the signal that caused the stop.
Copy a
.I siginfo_t
structure (see
.BR sigaction (2))
from the tracee to the address
.I data
in the tracer.
.RI ( addr
is ignored.)
.TP
.BR PTRACE_SETSIGINFO " (since Linux 2.3.99-pre6)"
Set signal information:
copy a
.I siginfo_t
structure from the address
.I data
in the tracer to the tracee.
This will affect only signals that would normally be delivered to
the tracee and were caught by the tracer.
It may be difficult to tell
these normal signals from synthetic signals generated by
.BR ptrace ()
itself.
.RI ( addr
is ignored.)
.TP
.BR PTRACE_PEEKSIGINFO " (since Linux 3.10)"
.\" commit 84c751bd4aebbaae995fe32279d3dba48327bad4
Retrieve
.I siginfo_t
structures without removing signals from a queue.
.I addr
points to a
.I ptrace_peeksiginfo_args
structure that specifies the ordinal position from which
copying of signals should start,
and the number of signals to copy.
.I siginfo_t
structures are copied into the buffer pointed to by
.IR data .
The return value contains the number of copied signals (zero indicates
that there is no signal corresponding to the specified ordinal position).
Within the returned
.I siginfo
structures,
the
.IR si_code
field includes information
.RB ( __SI_CHLD ,
.BR __SI_FAULT ,
etc.) that are not otherwise exposed to user space.
.PP
.in +10n
.nf
struct ptrace_peeksiginfo_args {
    u64 off;    /* Ordinal position in queue at which
                   to start copying signals */
    u32 flags;  /* PTRACE_PEEKSIGINFO_SHARED or 0 */
    s32 nr;     /* Number of signals to copy */
};
.fi

Currently, there is only one flag,
.BR PTRACE_PEEKSIGINFO_SHARED ,
for dumping signals from the process-wide signal queue.
If this flag is not set,
signals are read from the per-thread queue of the specified thread.
.in
.PP
.TP
.BR PTRACE_GETSIGMASK " (since Linux 3.11)"
.\" commit 29000caecbe87b6b66f144f72111f0d02fbbf0c1
Place a copy of the mask of blocked signals (see
.BR sigprocmask (2))
in the buffer pointed to by
.IR data ,
which should be a pointer to a buffer of type
.IR sigset_t .
The
.I addr
argument contains the size of the buffer pointed to by
.IR data
(i.e.,
.IR sizeof(sigset_t) ).
.TP
.BR PTRACE_SETSIGMASK " (since Linux 3.11)"
Change the mask of blocked signals (see
.BR sigprocmask (2))
to the value specified in the buffer pointed to by
.IR data ,
which should be a pointer to a buffer of type
.IR sigset_t .
The
.I addr
argument contains the size of the buffer pointed to by
.IR data
(i.e.,
.IR sizeof(sigset_t) ).
.TP
.BR PTRACE_SETOPTIONS " (since Linux 2.4.6; see BUGS for caveats)"
Set ptrace options from
.IR data .
.RI ( addr
is ignored.)
.IR data
is interpreted as a bit mask of options,
which are specified by the following flags:
.RS
.TP
.BR PTRACE_O_EXITKILL " (since Linux 3.8)"
.\" commit 992fb6e170639b0849bace8e49bf31bd37c4123
If a tracer sets this flag, a
.B SIGKILL
signal will be sent to every tracee if the tracer exits.
This option is useful for ptrace jailers that
want to ensure that tracees can never escape the tracer's control.
.TP
.BR PTRACE_O_TRACECLONE " (since Linux 2.5.46)"
Stop the tracee at the next
.BR clone (2)
and automatically start tracing the newly cloned process,
which will start with a
.BR SIGSTOP ,
or
.B PTRACE_EVENT_STOP
if
.B PTRACE_SEIZE
was used.
A
.BR waitpid (2)
by the tracer will return a
.I status
value such that

.nf
  status>>8 == (SIGTRAP | (PTRACE_EVENT_CLONE<<8))
.fi

The PID of the new process can be retrieved with
.BR PTRACE_GETEVENTMSG .
.IP
This option may not catch
.BR clone (2)
calls in all cases.
If the tracee calls
.BR clone (2)
with the
.B CLONE_VFORK
flag,
.B PTRACE_EVENT_VFORK
will be delivered instead
if
.B PTRACE_O_TRACEVFORK
is set; otherwise if the tracee calls
.BR clone (2)
with the exit signal set to
.BR SIGCHLD ,
.B PTRACE_EVENT_FORK
will be delivered if
.B PTRACE_O_TRACEFORK
is set.
.TP
.BR PTRACE_O_TRACEEXEC " (since Linux 2.5.46)"
Stop the tracee at the next
.BR execve (2).
A
.BR waitpid (2)
by the tracer will return a
.I status
value such that

.nf
  status>>8 == (SIGTRAP | (PTRACE_EVENT_EXEC<<8))
.fi

If the execing thread is not a thread group leader,
the thread ID is reset to thread group leader's ID before this stop.
Since Linux 3.0, the former thread ID can be retrieved with
.BR PTRACE_GETEVENTMSG .
.TP
.BR PTRACE_O_TRACEEXIT " (since Linux 2.5.60)"
Stop the tracee at exit.
A
.BR waitpid (2)
by the tracer will return a
.I status
value such that

.nf
  status>>8 == (SIGTRAP | (PTRACE_EVENT_EXIT<<8))
.fi

The tracee's exit status can be retrieved with
.BR PTRACE_GETEVENTMSG .
.IP
The tracee is stopped early during process exit,
when registers are still available,
allowing the tracer to see where the exit occurred,
whereas the normal exit notification is done after the process
is finished exiting.
Even though context is available,
the tracer cannot prevent the exit from happening at this point.
.TP
.BR PTRACE_O_TRACEFORK " (since Linux 2.5.46)"
Stop the tracee at the next
.BR fork (2)
and automatically start tracing the newly forked process,
which will start with a
.BR SIGSTOP ,
or
.B PTRACE_EVENT_STOP
if
.B PTRACE_SEIZE
was used.
A
.BR waitpid (2)
by the tracer will return a
.I status
value such that

.nf
  status>>8 == (SIGTRAP | (PTRACE_EVENT_FORK<<8))
.fi

The PID of the new process can be retrieved with
.BR PTRACE_GETEVENTMSG .
.TP
.BR PTRACE_O_TRACESYSGOOD " (since Linux 2.4.6)"
When delivering system call traps, set bit 7 in the signal number
(i.e., deliver
.IR "SIGTRAP|0x80" ).
This makes it easy for the tracer to distinguish
normal traps from those caused by a system call.
.RB ( PTRACE_O_TRACESYSGOOD
may not work on all architectures.)
.TP
.BR PTRACE_O_TRACEVFORK " (since Linux 2.5.46)"
Stop the tracee at the next
.BR vfork (2)
and automatically start tracing the newly vforked process,
which will start with a
.BR SIGSTOP ,
or
.B PTRACE_EVENT_STOP
if
.B PTRACE_SEIZE
was used.
A
.BR waitpid (2)
by the tracer will return a
.I status
value such that

.nf
  status>>8 == (SIGTRAP | (PTRACE_EVENT_VFORK<<8))
.fi

The PID of the new process can be retrieved with
.BR PTRACE_GETEVENTMSG .
.TP
.BR PTRACE_O_TRACEVFORKDONE " (since Linux 2.5.60)"
Stop the tracee at the completion of the next
.BR vfork (2).
A
.BR waitpid (2)
by the tracer will return a
.I status
value such that

.nf
  status>>8 == (SIGTRAP | (PTRACE_EVENT_VFORK_DONE<<8))
.fi

The PID of the new process can (since Linux 2.6.18) be retrieved with
.BR PTRACE_GETEVENTMSG .
.RE
.TP
.BR PTRACE_GETEVENTMSG " (since Linux 2.5.46)"
Retrieve a message (as an
.IR "unsigned long" )
about the ptrace event
that just happened, placing it at the address
.I data
in the tracer.
For
.BR PTRACE_EVENT_EXIT ,
this is the tracee's exit status.
For
.BR PTRACE_EVENT_FORK ,
.BR PTRACE_EVENT_VFORK ,
.BR PTRACE_EVENT_VFORK_DONE ,
and
.BR PTRACE_EVENT_CLONE ,
this is the PID of the new process.
.RI ( addr
is ignored.)
.TP
.B PTRACE_CONT
Restart the stopped tracee process.
If
.I data
is nonzero,
it is interpreted as the number of a signal to be delivered to the tracee;
otherwise, no signal is delivered.
Thus, for example, the tracer can control
whether a signal sent to the tracee is delivered or not.
.RI ( addr
is ignored.)
.TP
.BR PTRACE_SYSCALL ", " PTRACE_SINGLESTEP
Restart the stopped tracee as for
.BR PTRACE_CONT ,
but arrange for the tracee to be stopped at
the next entry to or exit from a system call,
or after execution of a single instruction, respectively.
(The tracee will also, as usual, be stopped upon receipt of a signal.)
From the tracer's perspective, the tracee will appear to have been
stopped by receipt of a
.BR SIGTRAP .
So, for
.BR PTRACE_SYSCALL ,
for example, the idea is to inspect
the arguments to the system call at the first stop,
then do another
.B PTRACE_SYSCALL
and inspect the return value of the system call at the second stop.
The
.I data
argument is treated as for
.BR PTRACE_CONT .
.RI ( addr
is ignored.)
.TP
.BR PTRACE_SYSEMU ", " PTRACE_SYSEMU_SINGLESTEP " (since Linux 2.6.14)"
For
.BR PTRACE_SYSEMU ,
continue and stop on entry to the next system call,
which will not be executed.
For
.BR PTRACE_SYSEMU_SINGLESTEP ,
do the same but also singlestep if not a system call.
This call is used by programs like
User Mode Linux that want to emulate all the tracee's system calls.
The
.I data
argument is treated as for
.BR PTRACE_CONT .
The
.I addr
argument is ignored.
These requests are currently
.\" As at 3.7
supported only on x86.
.TP
.BR PTRACE_LISTEN " (since Linux 3.4)"
Restart the stopped tracee, but prevent it from executing.
The resulting state of the tracee is similar to a process which
has been stopped by a
.B SIGSTOP
(or other stopping signal).
See the "group-stop" subsection for additional information.
.B PTRACE_LISTEN
works only on tracees attached by
.BR PTRACE_SEIZE .
.TP
.B PTRACE_KILL
Send the tracee a
.B SIGKILL
to terminate it.
.RI ( addr
and
.I data
are ignored.)
.IP
.I This operation is deprecated; do not use it!
Instead, send a
.BR SIGKILL
directly using
.BR kill (2)
or
.BR tgkill (2).
The problem with
.B PTRACE_KILL
is that it requires the tracee to be in signal-delivery-stop,
otherwise it may not work
(i.e., may complete successfully but won't kill the tracee).
By contrast, sending a
.B SIGKILL
directly has no such limitation.
.\" [Note from Denys Vlasenko:
.\"     deprecation suggested by Oleg Nesterov. He prefers to deprecate it
.\"     instead of describing (and needing to support) PTRACE_KILL's quirks.]
.TP
.BR PTRACE_INTERRUPT " (since Linux 3.4)"
Stop a tracee.
If the tracee is running or sleeping in kernel space and
.B PTRACE_SYSCALL
is in effect,
the system call is interrupted and syscall-exit-stop is reported.
(The interrupted system call is restarted when the tracee is restarted.)
If the tracee was already stopped by a signal and
.B PTRACE_LISTEN
was sent to it,
the tracee stops with
.B PTRACE_EVENT_STOP
and
.I WSTOPSIG(status)
returns the stop signal.
If any other ptrace-stop is generated at the same time (for example,
if a signal is sent to the tracee), this ptrace-stop happens.
If none of the above applies (for example, if the tracee is running in user
space), it stops with
.B PTRACE_EVENT_STOP
with
.I WSTOPSIG(status)
==
.BR SIGTRAP .
.B PTRACE_INTERRUPT
only works on tracees attached by
.BR PTRACE_SEIZE .
.TP
.B PTRACE_ATTACH
Attach to the process specified in
.IR pid ,
making it a tracee of the calling process.
.\" No longer true (removed by Denys Vlasenko, 2011, who remarks:
.\"        "I think it isn't true in non-ancient 2.4 and in 2.6/3.x.
.\"         Basically, it's not true for any Linux in practical use.
.\" ; the behavior of the tracee is as if it had done a
.\" .BR PTRACE_TRACEME .
.\" The calling process actually becomes the parent of the tracee
.\" process for most purposes (e.g., it will receive
.\" notification of tracee events and appears in
.\" .BR ps (1)
.\" output as the tracee's parent), but a
.\" .BR getppid (2)
.\" by the tracee will still return the PID of the original parent.
The tracee is sent a
.BR SIGSTOP ,
but will not necessarily have stopped
by the completion of this call; use
.BR waitpid (2)
to wait for the tracee to stop.
See the "Attaching and detaching" subsection for additional information.
.RI ( addr
and
.I data
are ignored.)
.TP
.BR PTRACE_SEIZE " (since Linux 3.4)"
Attach to the process specified in
.IR pid ,
making it a tracee of the calling process.
Unlike
.BR PTRACE_ATTACH ,
.B PTRACE_SEIZE
does not stop the process.
Only a
.BR PTRACE_SEIZE d
process can accept
.B PTRACE_INTERRUPT
and
.B PTRACE_LISTEN
commands.
.I addr
must be zero.
.I data
contains a bit mask of ptrace options to activate immediately.
.TP
.B PTRACE_DETACH
Restart the stopped tracee as for
.BR PTRACE_CONT ,
but first detach from it.
Under Linux, a tracee can be detached in this way regardless
of which method was used to initiate tracing.
.RI ( addr
is ignored.)
.SS Death under ptrace
When a (possibly multithreaded) process receives a killing signal
(one whose disposition is set to
.B SIG_DFL
and whose default action is to kill the process),
all threads exit.
Tracees report their death to their tracer(s).
Notification of this event is delivered via
.BR waitpid (2).
.LP
Note that the killing signal will first cause signal-delivery-stop
(on one tracee only),
and only after it is injected by the tracer
(or after it was dispatched to a thread which isn't traced),
will death from the signal happen on
.I all
tracees within a multithreaded process.
(The term "signal-delivery-stop" is explained below.)
.LP
.B SIGKILL
does not generate signal-delivery-stop and
therefore the tracer can't suppress it.
.B SIGKILL
kills even within system calls
(syscall-exit-stop is not generated prior to death by
.BR SIGKILL ).
The net effect is that
.B SIGKILL
always kills the process (all its threads),
even if some threads of the process are ptraced.
.LP
When the tracee calls
.BR _exit (2),
it reports its death to its tracer.
Other threads are not affected.
.LP
When any thread executes
.BR exit_group (2),
every tracee in its thread group reports its death to its tracer.
.LP
If the
.B PTRACE_O_TRACEEXIT
option is on,
.B PTRACE_EVENT_EXIT
will happen before actual death.
This applies to exits via
.BR exit (2),
.BR exit_group (2),
and signal deaths (except
.BR SIGKILL ),
and when threads are torn down on
.BR execve (2)
in a multithreaded process.
.LP
The tracer cannot assume that the ptrace-stopped tracee exists.
There are many scenarios when the tracee may die while stopped (such as
.BR SIGKILL ).
Therefore, the tracer must be prepared to handle an
.B ESRCH
error on any ptrace operation.
Unfortunately, the same error is returned if the tracee
exists but is not ptrace-stopped
(for commands which require a stopped tracee),
or if it is not traced by the process which issued the ptrace call.
The tracer needs to keep track of the stopped/running state of the tracee,
and interpret
.B ESRCH
as "tracee died unexpectedly" only if it knows that the tracee has
been observed to enter ptrace-stop.
Note that there is no guarantee that
.I waitpid(WNOHANG)
will reliably report the tracee's death status if a
ptrace operation returned
.BR ESRCH .
.I waitpid(WNOHANG)
may return 0 instead.
In other words, the tracee may be "not yet fully dead",
but already refusing ptrace requests.
.LP
The tracer can't assume that the tracee
.I always
ends its life by reporting
.I WIFEXITED(status)
or
.IR WIFSIGNALED(status) ;
there are cases where this does not occur.
For example, if a thread other than thread group leader does an
.BR execve (2),
it disappears;
its PID will never be seen again,
and any subsequent ptrace stops will be reported under
the thread group leader's PID.
.SS Stopped states
A tracee can be in two states: running or stopped.
For the purposes of ptrace, a tracee which is blocked in a system call
(such as
.BR read (2),
.BR pause (2),
etc.)
is nevertheless considered to be running, even if the tracee is blocked
for a long time.
The state of the tracee after
.BR PTRACE_LISTEN
is somewhat of a gray area: it is not in any ptrace-stop (ptrace commands
won't work on it, and it will deliver
.BR waitpid (2)
notifications),
but it also may be considered "stopped" because
it is not executing instructions (is not scheduled), and if it was
in group-stop before
.BR PTRACE_LISTEN ,
it will not respond to signals until
.B SIGCONT
is received.
.LP
There are many kinds of states when the tracee is stopped, and in ptrace
discussions they are often conflated.
Therefore, it is important to use precise terms.
.LP
In this manual page, any stopped state in which the tracee is ready
to accept ptrace commands from the tracer is called
.IR ptrace-stop .
Ptrace-stops can
be further subdivided into
.IR signal-delivery-stop ,
.IR group-stop ,
.IR syscall-stop ,
and so on.
These stopped states are described in detail below.
.LP
When the running tracee enters ptrace-stop, it notifies its tracer using
.BR waitpid (2)
(or one of the other "wait" system calls).
Most of this manual page assumes that the tracer waits with:
.LP
    pid = waitpid(pid_or_minus_1, &status, __WALL);
.LP
Ptrace-stopped tracees are reported as returns with
.I pid
greater than 0 and
.I WIFSTOPPED(status)
true.
.\" Denys Vlasenko:
.\"     Do we require __WALL usage, or will just using 0 be ok? (With 0,
.\"     I am not 100% sure there aren't ugly corner cases.) Are the
.\"     rules different if user wants to use waitid? Will waitid require
.\"     WEXITED?
.\"
.LP
The
.B __WALL
flag does not include the
.B WSTOPPED
and
.B WEXITED
flags, but implies their functionality.
.LP
Setting the
.B WCONTINUED
flag when calling
.BR waitpid (2)
is not recommended: the "continued" state is per-process and
consuming it can confuse the real parent of the tracee.
.LP
Use of the
.B WNOHANG
flag may cause
.BR waitpid (2)
to return 0 ("no wait results available yet")
even if the tracer knows there should be a notification.
Example:
.nf

    errno = 0;
    ptrace(PTRACE_CONT, pid, 0L, 0L);
    if (errno == ESRCH) {
        /* tracee is dead */
        r = waitpid(tracee, &status, __WALL | WNOHANG);
        /* r can still be 0 here! */
    }
.fi
.\" FIXME:
.\"     waitid usage? WNOWAIT?
.\"     describe how wait notifications queue (or not queue)
.LP
The following kinds of ptrace-stops exist: signal-delivery-stops,
group-stops,
.B PTRACE_EVENT
stops, syscall-stops.
They all are reported by
.BR waitpid (2)
with
.I WIFSTOPPED(status)
true.
They may be differentiated by examining the value
.IR status>>8 ,
and if there is ambiguity in that value, by querying
.BR PTRACE_GETSIGINFO .
(Note: the
.I WSTOPSIG(status)
macro can't be used to perform this examination,
because it returns the value
.IR "(status>>8)\ &\ 0xff" .)
.SS Signal-delivery-stop
When a (possibly multithreaded) process receives any signal except
.BR SIGKILL ,
the kernel selects an arbitrary thread which handles the signal.
(If the signal is generated with
.BR tgkill (2),
the target thread can be explicitly selected by the caller.)
If the selected thread is traced, it enters signal-delivery-stop.
At this point, the signal is not yet delivered to the process,
and can be suppressed by the tracer.
If the tracer doesn't suppress the signal,
it passes the signal to the tracee in the next ptrace restart request.
This second step of signal delivery is called
.I "signal injection"
in this manual page.
Note that if the signal is blocked,
signal-delivery-stop doesn't happen until the signal is unblocked,
with the usual exception that
.B SIGSTOP
can't be blocked.
.LP
Signal-delivery-stop is observed by the tracer as
.BR waitpid (2)
returning with
.I WIFSTOPPED(status)
true, with the signal returned by
.IR WSTOPSIG(status) .
If the signal is
.BR SIGTRAP ,
this may be a different kind of ptrace-stop;
see the "Syscall-stops" and "execve" sections below for details.
If
.I WSTOPSIG(status)
returns a stopping signal, this may be a group-stop; see below.
.SS Signal injection and suppression
After signal-delivery-stop is observed by the tracer,
the tracer should restart the tracee with the call
.LP
    ptrace(PTRACE_restart, pid, 0, sig)
.LP
where
.B PTRACE_restart
is one of the restarting ptrace requests.
If
.I sig
is 0, then a signal is not delivered.
Otherwise, the signal
.I sig
is delivered.
This operation is called
.I "signal injection"
in this manual page, to distinguish it from signal-delivery-stop.
.LP
The
.I sig
value may be different from the
.I WSTOPSIG(status)
value: the tracer can cause a different signal to be injected.
.LP
Note that a suppressed signal still causes system calls to return
prematurely.
In this case system calls will be restarted: the tracer will
observe the tracee to reexecute the interrupted system call (or
.BR restart_syscall (2)
system call for a few syscalls which use a different mechanism
for restarting) if the tracer uses
.BR PTRACE_SYSCALL .
Even system calls (such as
.BR poll (2))
which are not restartable after signal are restarted after
signal is suppressed;
however, kernel bugs exist which cause some syscalls to fail with
.B EINTR
even though no observable signal is injected to the tracee.
.LP
Restarting ptrace commands issued in ptrace-stops other than
signal-delivery-stop are not guaranteed to inject a signal, even if
.I sig
is nonzero.
No error is reported; a nonzero
.I sig
may simply be ignored.
Ptrace users should not try to "create a new signal" this way: use
.BR tgkill (2)
instead.
.LP
The fact that signal injection requests may be ignored
when restarting the tracee after
ptrace stops that are not signal-delivery-stops
is a cause of confusion among ptrace users.
One typical scenario is that the tracer observes group-stop,
mistakes it for signal-delivery-stop, restarts the tracee with

    ptrace(PTRACE_restart, pid, 0, stopsig)

with the intention of injecting
.IR stopsig ,
but
.I stopsig
gets ignored and the tracee continues to run.
.LP
The
.B SIGCONT
signal has a side effect of waking up (all threads of)
a group-stopped process.
This side effect happens before signal-delivery-stop.
The tracer can't suppress this side effect (it can
only suppress signal injection, which only causes the
.BR SIGCONT
handler to not be executed in the tracee, if such a handler is installed).
In fact, waking up from group-stop may be followed by
signal-delivery-stop for signal(s)
.I other than
.BR SIGCONT ,
if they were pending when
.B SIGCONT
was delivered.
In other words,
.B SIGCONT
may be not the first signal observed by the tracee after it was sent.
.LP
Stopping signals cause (all threads of) a process to enter group-stop.
This side effect happens after signal injection, and therefore can be
suppressed by the tracer.
.LP
In Linux 2.4 and earlier, the
.B SIGSTOP
signal can't be injected.
.\" In the Linux 2.4 sources, in arch/i386/kernel/signal.c::do_signal(),
.\" there is:
.\"
.\"             /* The debugger continued.  Ignore SIGSTOP.  */
.\"             if (signr == SIGSTOP)
.\"                     continue;
.LP
.B PTRACE_GETSIGINFO
can be used to retrieve a
.I siginfo_t
structure which corresponds to the delivered signal.
.B PTRACE_SETSIGINFO
may be used to modify it.
If
.B PTRACE_SETSIGINFO
has been used to alter
.IR siginfo_t ,
the
.I si_signo
field and the
.I sig
parameter in the restarting command must match,
otherwise the result is undefined.
.SS Group-stop
When a (possibly multithreaded) process receives a stopping signal,
all threads stop.
If some threads are traced, they enter a group-stop.
Note that the stopping signal will first cause signal-delivery-stop
(on one tracee only), and only after it is injected by the tracer
(or after it was dispatched to a thread which isn't traced),
will group-stop be initiated on
.I all
tracees within the multithreaded process.
As usual, every tracee reports its group-stop separately
to the corresponding tracer.
.LP
Group-stop is observed by the tracer as
.BR waitpid (2)
returning with
.I WIFSTOPPED(status)
true, with the stopping signal available via
.IR WSTOPSIG(status) .
The same result is returned by some other classes of ptrace-stops,
therefore the recommended practice is to perform the call
.LP
    ptrace(PTRACE_GETSIGINFO, pid, 0, &siginfo)
.LP
The call can be avoided if the signal is not
.BR SIGSTOP ,
.BR SIGTSTP ,
.BR SIGTTIN ,
or
.BR SIGTTOU ;
only these four signals are stopping signals.
If the tracer sees something else, it can't be a group-stop.
Otherwise, the tracer needs to call
.BR PTRACE_GETSIGINFO .
If
.B PTRACE_GETSIGINFO
fails with
.BR EINVAL ,
then it is definitely a group-stop.
(Other failure codes are possible, such as
.B ESRCH
("no such process") if a
.B SIGKILL
killed the tracee.)
.LP
If tracee was attached using
.BR PTRACE_SEIZE ,
group-stop is indicated by
.BR PTRACE_EVENT_STOP :
.IR "status>>16 == PTRACE_EVENT_STOP" .
This allows detection of group-stops
without requiring an extra
.B PTRACE_GETSIGINFO
call.
.LP
As of Linux 2.6.38,
after the tracer sees the tracee ptrace-stop and until it
restarts or kills it, the tracee will not run,
and will not send notifications (except
.B SIGKILL
death) to the tracer, even if the tracer enters into another
.BR waitpid (2)
call.
.LP
The kernel behavior described in the previous paragraph
causes a problem with transparent handling of stopping signals.
If the tracer restarts the tracee after group-stop,
the stopping signal
is effectively ignored\(emthe tracee doesn't remain stopped, it runs.
If the tracer doesn't restart the tracee before entering into the next
.BR waitpid (2),
future
.B SIGCONT
signals will not be reported to the tracer;
this would cause the
.B SIGCONT
signals to have no effect on the tracee.
.LP
Since Linux 3.4, there is a method to overcome this problem: instead of
.BR PTRACE_CONT ,
a
.B PTRACE_LISTEN
command can be used to restart a tracee in a way where it does not execute,
but waits for a new event which it can report via
.BR waitpid (2)
(such as when
it is restarted by a
.BR SIGCONT ).
.SS PTRACE_EVENT stops
If the tracer sets
.B PTRACE_O_TRACE_*
options, the tracee will enter ptrace-stops called
.B PTRACE_EVENT
stops.
.LP
.B PTRACE_EVENT
stops are observed by the tracer as
.BR waitpid (2)
returning with
.IR WIFSTOPPED(status) ,
and
.I WSTOPSIG(status)
returns
.BR SIGTRAP .
An additional bit is set in the higher byte of the status word:
the value
.I status>>8
will be

    (SIGTRAP | PTRACE_EVENT_foo << 8).

The following events exist:
.TP
.B PTRACE_EVENT_VFORK
Stop before return from
.BR vfork (2)
or
.BR clone (2)
with the
.B CLONE_VFORK
flag.
When the tracee is continued after this stop, it will wait for child to
exit/exec before continuing its execution
(in other words, the usual behavior on
.BR vfork (2)).
.TP
.B PTRACE_EVENT_FORK
Stop before return from
.BR fork (2)
or
.BR clone (2)
with the exit signal set to
.BR SIGCHLD .
.TP
.B PTRACE_EVENT_CLONE
Stop before return from
.BR clone (2).
.TP
.B PTRACE_EVENT_VFORK_DONE
Stop before return from
.BR vfork (2)
or
.BR clone (2)
with the
.B CLONE_VFORK
flag,
but after the child unblocked this tracee by exiting or execing.
.LP
For all four stops described above,
the stop occurs in the parent (i.e., the tracee),
not in the newly created thread.
.BR PTRACE_GETEVENTMSG
can be used to retrieve the new thread's ID.
.TP
.B PTRACE_EVENT_EXEC
Stop before return from
.BR execve (2).
Since Linux 3.0,
.BR PTRACE_GETEVENTMSG
returns the former thread ID.
.TP
.B PTRACE_EVENT_EXIT
Stop before exit (including death from
.BR exit_group (2)),
signal death, or exit caused by
.BR execve (2)
in a multithreaded process.
.B PTRACE_GETEVENTMSG
returns the exit status.
Registers can be examined
(unlike when "real" exit happens).
The tracee is still alive; it needs to be
.BR PTRACE_CONT ed
or
.BR PTRACE_DETACH ed
to finish exiting.
.TP
.B PTRACE_EVENT_STOP
Stop induced by
.B PTRACE_INTERRUPT
command, or group-stop, or initial ptrace-stop when a new child is attached
(only if attached using
.BR PTRACE_SEIZE ),
or
.B PTRACE_EVENT_STOP
if
.B PTRACE_SEIZE
was used.
.LP
.B PTRACE_GETSIGINFO
on
.B PTRACE_EVENT
stops returns
.B SIGTRAP
in
.IR si_signo ,
with
.I si_code
set to
.IR "(event<<8)\ |\ SIGTRAP" .
.SS Syscall-stops
If the tracee was restarted by
.BR PTRACE_SYSCALL ,
the tracee enters
syscall-enter-stop just prior to entering any system call.
If the tracer restarts the tracee with
.BR PTRACE_SYSCALL ,
the tracee enters syscall-exit-stop when the system call is finished,
or if it is interrupted by a signal.
(That is, signal-delivery-stop never happens between syscall-enter-stop
and syscall-exit-stop; it happens
.I after
syscall-exit-stop.)
.LP
Other possibilities are that the tracee may stop in a
.B PTRACE_EVENT
stop, exit (if it entered
.BR _exit (2)
or
.BR exit_group (2)),
be killed by
.BR SIGKILL ,
or die silently (if it is a thread group leader, the
.BR execve (2)
happened in another thread,
and that thread is not traced by the same tracer;
this situation is discussed later).
.LP
Syscall-enter-stop and syscall-exit-stop are observed by the tracer as
.BR waitpid (2)
returning with
.I WIFSTOPPED(status)
true, and
.I WSTOPSIG(status)
giving
.BR SIGTRAP .
If the
.B PTRACE_O_TRACESYSGOOD
option was set by the tracer, then
.I WSTOPSIG(status)
will give the value
.IR "(SIGTRAP\ |\ 0x80)" .
.LP
Syscall-stops can be distinguished from signal-delivery-stop with
.B SIGTRAP
by querying
.BR PTRACE_GETSIGINFO
for the following cases:
.TP
.IR si_code " <= 0"
.B SIGTRAP
was delivered as a result of a user-space action,
for example, a system call
.RB ( tgkill (2),
.BR kill (2),
.BR sigqueue (3),
etc.),
expiration of a POSIX timer,
change of state on a POSIX message queue,
or completion of an asynchronous I/O request.
.TP
.IR si_code " == SI_KERNEL (0x80)"
.B SIGTRAP
was sent by the kernel.
.TP
.IR si_code " == SIGTRAP or " si_code " == (SIGTRAP|0x80)"
This is a syscall-stop.
.LP
However, syscall-stops happen very often (twice per system call),
and performing
.B PTRACE_GETSIGINFO
for every syscall-stop may be somewhat expensive.
.LP
Some architectures allow the cases to be distinguished
by examining registers.
For example, on x86,
.I rax
==
.RB - ENOSYS
in syscall-enter-stop.
Since
.B SIGTRAP
(like any other signal) always happens
.I after
syscall-exit-stop,
and at this point
.I rax
almost never contains
.RB - ENOSYS ,
the
.B SIGTRAP
looks like "syscall-stop which is not syscall-enter-stop";
in other words, it looks like a
"stray syscall-exit-stop" and can be detected this way.
But such detection is fragile and is best avoided.
.LP
Using the
.B PTRACE_O_TRACESYSGOOD
option is the recommended method to distinguish syscall-stops
from other kinds of ptrace-stops,
since it is reliable and does not incur a performance penalty.
.LP
Syscall-enter-stop and syscall-exit-stop are
indistinguishable from each other by the tracer.
The tracer needs to keep track of the sequence of
ptrace-stops in order to not misinterpret syscall-enter-stop as
syscall-exit-stop or vice versa.
The rule is that syscall-enter-stop is
always followed by syscall-exit-stop,
.B PTRACE_EVENT
stop or the tracee's death;
no other kinds of ptrace-stop can occur in between.
.LP
If after syscall-enter-stop,
the tracer uses a restarting command other than
.BR PTRACE_SYSCALL ,
syscall-exit-stop is not generated.
.LP
.B PTRACE_GETSIGINFO
on syscall-stops returns
.B SIGTRAP
in
.IR si_signo ,
with
.I si_code
set to
.B SIGTRAP
or
.IR (SIGTRAP|0x80) .
.SS PTRACE_SINGLESTEP, PTRACE_SYSEMU, PTRACE_SYSEMU_SINGLESTEP stops
[Details of these kinds of stops are yet to be documented.]
.\"
.\" FIXME
.\" document stops occurring with PTRACE_SINGLESTEP, PTRACE_SYSEMU,
.\" PTRACE_SYSEMU_SINGLESTEP
.SS Informational and restarting ptrace commands
Most ptrace commands (all except
.BR PTRACE_ATTACH ,
.BR PTRACE_SEIZE ,
.BR PTRACE_TRACEME ,
.BR PTRACE_INTERRUPT ,
and
.BR PTRACE_KILL )
require the tracee to be in a ptrace-stop, otherwise they fail with
.BR ESRCH .
.LP
When the tracee is in ptrace-stop,
the tracer can read and write data to
the tracee using informational commands.
These commands leave the tracee in ptrace-stopped state:
.LP
.nf
    ptrace(PTRACE_PEEKTEXT/PEEKDATA/PEEKUSER, pid, addr, 0);
    ptrace(PTRACE_POKETEXT/POKEDATA/POKEUSER, pid, addr, long_val);
    ptrace(PTRACE_GETREGS/GETFPREGS, pid, 0, &struct);
    ptrace(PTRACE_SETREGS/SETFPREGS, pid, 0, &struct);
    ptrace(PTRACE_GETREGSET, pid, NT_foo, &iov);
    ptrace(PTRACE_SETREGSET, pid, NT_foo, &iov);
    ptrace(PTRACE_GETSIGINFO, pid, 0, &siginfo);
    ptrace(PTRACE_SETSIGINFO, pid, 0, &siginfo);
    ptrace(PTRACE_GETEVENTMSG, pid, 0, &long_var);
    ptrace(PTRACE_SETOPTIONS, pid, 0, PTRACE_O_flags);
.fi
.LP
Note that some errors are not reported.
For example, setting signal information
.RI ( siginfo )
may have no effect in some ptrace-stops, yet the call may succeed
(return 0 and not set
.IR errno );
querying
.B PTRACE_GETEVENTMSG
may succeed and return some random value if current ptrace-stop
is not documented as returning a meaningful event message.
.LP
The call

    ptrace(PTRACE_SETOPTIONS, pid, 0, PTRACE_O_flags);

affects one tracee.
The tracee's current flags are replaced.
Flags are inherited by new tracees created and "auto-attached" via active
.BR PTRACE_O_TRACEFORK ,
.BR PTRACE_O_TRACEVFORK ,
or
.BR PTRACE_O_TRACECLONE
options.
.LP
Another group of commands makes the ptrace-stopped tracee run.
They have the form:
.LP
    ptrace(cmd, pid, 0, sig);
.LP
where
.I cmd
is
.BR PTRACE_CONT ,
.BR PTRACE_LISTEN ,
.BR PTRACE_DETACH ,
.BR PTRACE_SYSCALL ,
.BR PTRACE_SINGLESTEP ,
.BR PTRACE_SYSEMU ,
or
.BR PTRACE_SYSEMU_SINGLESTEP .
If the tracee is in signal-delivery-stop,
.I sig
is the signal to be injected (if it is nonzero).
Otherwise,
.I sig
may be ignored.
(When restarting a tracee from a ptrace-stop other than signal-delivery-stop,
recommended practice is to always pass 0 in
.IR sig .)
.SS Attaching and detaching
A thread can be attached to the tracer using the call

    ptrace(PTRACE_ATTACH, pid, 0, 0);

or

    ptrace(PTRACE_SEIZE, pid, 0, PTRACE_O_flags);

.B PTRACE_ATTACH
sends
.B SIGSTOP
to this thread.
If the tracer wants this
.B SIGSTOP
to have no effect, it needs to suppress it.
Note that if other signals are concurrently sent to
this thread during attach,
the tracer may see the tracee enter signal-delivery-stop
with other signal(s) first!
The usual practice is to reinject these signals until
.B SIGSTOP
is seen, then suppress
.B SIGSTOP
injection.
The design bug here is that a ptrace attach and a concurrently delivered
.B SIGSTOP
may race and the concurrent
.B SIGSTOP
may be lost.
.\"
.\" FIXME: Describe how to attach to a thread which is already
.\"        group-stopped.
.LP
Since attaching sends
.B SIGSTOP
and the tracer usually suppresses it, this may cause a stray
.B EINTR
return from the currently executing system call in the tracee,
as described in the "Signal injection and suppression" section.
.LP
Since Linux 3.4,
.B PTRACE_SEIZE
can be used instead of
.BR PTRACE_ATTACH .
.B PTRACE_SEIZE
does not stop the attached process.
If you need to stop
it after attach (or at any other time) without sending it any signals,
use
.B PTRACE_INTERRUPT
command.
.LP
The request

    ptrace(PTRACE_TRACEME, 0, 0, 0);

turns the calling thread into a tracee.
The thread continues to run (doesn't enter ptrace-stop).
A common practice is to follow the
.B PTRACE_TRACEME
with

    raise(SIGSTOP);

and allow the parent (which is our tracer now) to observe our
signal-delivery-stop.
.LP
If the
.BR PTRACE_O_TRACEFORK ,
.BR PTRACE_O_TRACEVFORK ,
or
.BR PTRACE_O_TRACECLONE
options are in effect, then children created by, respectively,
.BR vfork (2)
or
.BR clone (2)
with the
.B CLONE_VFORK
flag,
.BR fork (2)
or
.BR clone (2)
with the exit signal set to
.BR SIGCHLD ,
and other kinds of
.BR clone (2),
are automatically attached to the same tracer which traced their parent.
.B SIGSTOP
is delivered to the children, causing them to enter
signal-delivery-stop after they exit the system call which created them.
.LP
Detaching of the tracee is performed by:

    ptrace(PTRACE_DETACH, pid, 0, sig);

.B PTRACE_DETACH
is a restarting operation;
therefore it requires the tracee to be in ptrace-stop.
If the tracee is in signal-delivery-stop, a signal can be injected.
Otherwise, the
.I sig
parameter may be silently ignored.
.LP
If the tracee is running when the tracer wants to detach it,
the usual solution is to send
.B SIGSTOP
(using
.BR tgkill (2),
to make sure it goes to the correct thread),
wait for the tracee to stop in signal-delivery-stop for
.B SIGSTOP
and then detach it (suppressing
.B SIGSTOP
injection).
A design bug is that this can race with concurrent
.BR SIGSTOP s.
Another complication is that the tracee may enter other ptrace-stops
and needs to be restarted and waited for again, until
.B SIGSTOP
is seen.
Yet another complication is to be sure that
the tracee is not already ptrace-stopped,
because no signal delivery happens while it is\(emnot even
.BR SIGSTOP .
.\" FIXME: Describe how to detach from a group-stopped tracee so that it
.\"        doesn't run, but continues to wait for SIGCONT.
.LP
If the tracer dies, all tracees are automatically detached and restarted,
unless they were in group-stop.
Handling of restart from group-stop is currently buggy,
but the "as planned" behavior is to leave tracee stopped and waiting for
.BR SIGCONT .
If the tracee is restarted from signal-delivery-stop,
the pending signal is injected.
.SS execve(2) under ptrace
.\" clone(2) CLONE_THREAD says:
.\"     If  any  of the threads in a thread group performs an execve(2),
.\"     then all threads other than the thread group leader are terminated,
.\"     and the new program is executed in the thread group leader.
.\"
When one thread in a multithreaded process calls
.BR execve (2),
the kernel destroys all other threads in the process,
.\" In kernel 3.1 sources, see fs/exec.c::de_thread()
and resets the thread ID of the execing thread to the
thread group ID (process ID).
(Or, to put things another way, when a multithreaded process does an
.BR execve (2),
at completion of the call, it appears as though the
.BR execve (2)
occurred in the thread group leader, regardless of which thread did the
.BR execve (2).)
This resetting of the thread ID looks very confusing to tracers:
.IP * 3
All other threads stop in
.B PTRACE_EVENT_EXIT
stop, if the
.BR PTRACE_O_TRACEEXIT
option was turned on.
Then all other threads except the thread group leader report
death as if they exited via
.BR _exit (2)
with exit code 0.
.IP *
The execing tracee changes its thread ID while it is in the
.BR execve (2).
(Remember, under ptrace, the "pid" returned from
.BR waitpid (2),
or fed into ptrace calls, is the tracee's thread ID.)
That is, the tracee's thread ID is reset to be the same as its process ID,
which is the same as the thread group leader's thread ID.
.IP *
Then a
.B PTRACE_EVENT_EXEC
stop happens, if the
.BR PTRACE_O_TRACEEXEC
option was turned on.
.IP *
If the thread group leader has reported its
.B PTRACE_EVENT_EXIT
stop by this time,
it appears to the tracer that
the dead thread leader "reappears from nowhere".
(Note: the thread group leader does not report death via
.I WIFEXITED(status)
until there is at least one other live thread.
This eliminates the possibility that the tracer will see
it dying and then reappearing.)
If the thread group leader was still alive,
for the tracer this may look as if thread group leader
returns from a different system call than it entered,
or even "returned from a system call even though
it was not in any system call".
If the thread group leader was not traced
(or was traced by a different tracer), then during
.BR execve (2)
it will appear as if it has become a tracee of
the tracer of the execing tracee.
.LP
All of the above effects are the artifacts of
the thread ID change in the tracee.
.LP
The
.B PTRACE_O_TRACEEXEC
option is the recommended tool for dealing with this situation.
First, it enables
.BR PTRACE_EVENT_EXEC
stop,
which occurs before
.BR execve (2)
returns.
In this stop, the tracer can use
.B PTRACE_GETEVENTMSG
to retrieve the tracee's former thread ID.
(This feature was introduced in Linux 3.0).
Second, the
.B PTRACE_O_TRACEEXEC
option disables legacy
.B SIGTRAP
generation on
.BR execve (2).
.LP
When the tracer receives
.B PTRACE_EVENT_EXEC
stop notification,
it is guaranteed that except this tracee and the thread group leader,
no other threads from the process are alive.
.LP
On receiving the
.B PTRACE_EVENT_EXEC
stop notification,
the tracer should clean up all its internal
data structures describing the threads of this process,
and retain only one data structure\(emone which
describes the single still running tracee, with

    thread ID == thread group ID == process ID.
.LP
Example: two threads call
.BR execve (2)
at the same time:
.LP
.nf
*** we get syscall-enter-stop in thread 1: **
PID1 execve("/bin/foo", "foo" <unfinished ...>
*** we issue PTRACE_SYSCALL for thread 1 **
*** we get syscall-enter-stop in thread 2: **
PID2 execve("/bin/bar", "bar" <unfinished ...>
*** we issue PTRACE_SYSCALL for thread 2 **
*** we get PTRACE_EVENT_EXEC for PID0, we issue PTRACE_SYSCALL **
*** we get syscall-exit-stop for PID0: **
PID0 <... execve resumed> )             = 0
.fi
.LP
If the
.B PTRACE_O_TRACEEXEC
option is
.I not
in effect for the execing tracee, the kernel delivers an extra
.B SIGTRAP
to the tracee after
.BR execve (2)
returns.
This is an ordinary signal (similar to one which can be
generated by
.IR "kill -TRAP" ),
not a special kind of ptrace-stop.
Employing
.B PTRACE_GETSIGINFO
for this signal returns
.I si_code
set to 0
.RI ( SI_USER ).
This signal may be blocked by signal mask,
and thus may be delivered (much) later.
.LP
Usually, the tracer (for example,
.BR strace (1))
would not want to show this extra post-execve
.B SIGTRAP
signal to the user, and would suppress its delivery to the tracee (if
.B SIGTRAP
is set to
.BR SIG_DFL ,
it is a killing signal).
However, determining
.I which
.B SIGTRAP
to suppress is not easy.
Setting the
.B PTRACE_O_TRACEEXEC
option and thus suppressing this extra
.B SIGTRAP
is the recommended approach.
.SS Real parent
The ptrace API (ab)uses the standard UNIX parent/child signaling over
.BR waitpid (2).
This used to cause the real parent of the process to stop receiving
several kinds of
.BR waitpid (2)
notifications when the child process is traced by some other process.
.LP
Many of these bugs have been fixed, but as of Linux 2.6.38 several still
exist; see BUGS below.
.LP
As of Linux 2.6.38, the following is believed to work correctly:
.IP * 3
exit/death by signal is reported first to the tracer, then,
when the tracer consumes the
.BR waitpid (2)
result, to the real parent (to the real parent only when the
whole multithreaded process exits).
If the tracer and the real parent are the same process,
the report is sent only once.
.SH RETURN VALUE
On success, the
.B PTRACE_PEEK*
requests return the requested data (but see NOTES),
while other requests return zero.
.LP
On error, all requests return \-1, and
.I errno
is set appropriately.
Since the value returned by a successful
.B PTRACE_PEEK*
request may be \-1, the caller must clear
.I errno
before the call, and then check it afterward
to determine whether or not an error occurred.
.SH ERRORS
.TP
.B EBUSY
(i386 only) There was an error with allocating or freeing a debug register.
.TP
.B EFAULT
There was an attempt to read from or write to an invalid area in
the tracer's or the tracee's memory,
probably because the area wasn't mapped or accessible.
Unfortunately, under Linux, different variations of this fault
will return
.B EIO
or
.B EFAULT
more or less arbitrarily.
.TP
.B EINVAL
An attempt was made to set an invalid option.
.TP
.B EIO
.I request
is invalid, or an attempt was made to read from or
write to an invalid area in the tracer's or the tracee's memory,
or there was a word-alignment violation,
or an invalid signal was specified during a restart request.
.TP
.B EPERM
The specified process cannot be traced.
This could be because the
tracer has insufficient privileges (the required capability is
.BR CAP_SYS_PTRACE );
unprivileged processes cannot trace processes that they
cannot send signals to or those running
set-user-ID/set-group-ID programs, for obvious reasons.
Alternatively, the process may already be being traced,
or (on kernels before 2.6.26) be
.BR init (8)
(PID 1).
.TP
.B ESRCH
The specified process does not exist, or is not currently being traced
by the caller, or is not stopped
(for requests that require a stopped tracee).
.SH CONFORMING TO
SVr4, 4.3BSD.
.SH NOTES
Although arguments to
.BR ptrace ()
are interpreted according to the prototype given,
glibc currently declares
.BR ptrace ()
as a variadic function with only the
.I request
argument fixed.
It is recommended to always supply four arguments,
even if the requested operation does not use them,
setting unused/ignored arguments to
.I 0L
or
.IR "(void\ *)\ 0".
.LP
At the system call level, the
.BR PTRACE_PEEKTEXT ,
.BR PTRACE_PEEKDATA ,
and
.BR PTRACE_PEEKUSER
requests have a different API: they store the result
at the address specified by the
.I data
parameter, and the return value is the error flag.
The glibc wrapper function provides the API given in DESCRIPTION above,
with the result being returned via the function return value.
.LP
In Linux kernels before 2.6.26,
.\" See commit 00cd5c37afd5f431ac186dd131705048c0a11fdb
.BR init (8),
the process with PID 1, may not be traced.
.LP
The layout of the contents of memory and the USER area are
quite operating-system- and architecture-specific.
The offset supplied, and the data returned,
might not entirely match with the definition of
.IR "struct user" .
.\" See http://lkml.org/lkml/2008/5/8/375
.LP
The size of a "word" is determined by the operating-system variant
(e.g., for 32-bit Linux it is 32 bits).
.LP
This page documents the way the
.BR ptrace ()
call works currently in Linux.
Its behavior differs noticeably on other flavors of UNIX.
In any case, use of
.BR ptrace ()
is highly specific to the operating system and architecture.
.SH BUGS
On hosts with 2.6 kernel headers,
.B PTRACE_SETOPTIONS
is declared with a different value than the one for 2.4.
This leads to applications compiled with 2.6 kernel
headers failing when run on 2.4 kernels.
This can be worked around by redefining
.B PTRACE_SETOPTIONS
to
.BR PTRACE_OLDSETOPTIONS ,
if that is defined.
.LP
Group-stop notifications are sent to the tracer, but not to real parent.
Last confirmed on 2.6.38.6.
.LP
If a thread group leader is traced and exits by calling
.BR _exit (2),
.\" Note from Denys Vlasenko:
.\"     Here "exits" means any kind of death - _exit, exit_group,
.\"     signal death. Signal death and exit_group cases are trivial,
.\"     though: since signal death and exit_group kill all other threads
.\"     too, "until all other threads exit" thing happens rather soon
.\"     in these cases. Therefore, only _exit presents observably
.\"     puzzling behavior to ptrace users: thread leader _exit's,
.\"     but WIFEXITED isn't reported! We are trying to explain here
.\"     why it is so.
a
.B PTRACE_EVENT_EXIT
stop will happen for it (if requested), but the subsequent
.B WIFEXITED
notification will not be delivered until all other threads exit.
As explained above, if one of other threads calls
.BR execve (2),
the death of the thread group leader will
.I never
be reported.
If the execed thread is not traced by this tracer,
the tracer will never know that
.BR execve (2)
happened.
One possible workaround is to
.B PTRACE_DETACH
the thread group leader instead of restarting it in this case.
Last confirmed on 2.6.38.6.
.\"  FIXME: ^^^ need to test/verify this scenario
.LP
A
.B SIGKILL
signal may still cause a
.B PTRACE_EVENT_EXIT
stop before actual signal death.
This may be changed in the future;
.B SIGKILL
is meant to always immediately kill tasks even under ptrace.
Last confirmed on 2.6.38.6.
.LP
Some system calls return with
.B EINTR
if a signal was sent to a tracee, but delivery was suppressed by the tracer.
(This is very typical operation: it is usually
done by debuggers on every attach, in order to not introduce
a bogus
.BR SIGSTOP ).
As of Linux 3.2.9, the following system calls are affected
(this list is likely incomplete):
.BR epoll_wait (2),
and
.BR read (2)
from an
.BR inotify (7)
file descriptor.
The usual symptom of this bug is that when you attach to
a quiescent process with the command

    strace \-p <process-ID>

then, instead of the usual
and expected one-line output such as
.nf

    restart_syscall(<... resuming interrupted call ...>_

.fi
or
.nf

    select(6, [5], NULL, [5], NULL_

.fi
('_' denotes the cursor position), you observe more than one line.
For example:
.nf

    clock_gettime(CLOCK_MONOTONIC, {15370, 690928118}) = 0
    epoll_wait(4,_

.fi
What is not visible here is that the process was blocked in
.BR epoll_wait (2)
before
.BR strace (1)
has attached to it.
Attaching caused
.BR epoll_wait (2)
to return to user space with the error
.BR EINTR .
In this particular case, the program reacted to
.B EINTR
by checking the current time, and then executing
.BR epoll_wait (2)
again.
(Programs which do not expect such "stray"
.BR EINTR
errors may behave in an unintended way upon an
.BR strace (1)
attach.)
.SH SEE ALSO
.BR gdb (1),
.BR strace (1),
.BR clone (2),
.BR execve (2),
.BR fork (2),
.BR gettid (2),
.BR sigaction (2),
.BR tgkill (2),
.BR vfork (2),
.BR waitpid (2),
.BR exec (3),
.BR capabilities (7),
.BR signal (7)