.\" $FreeBSD: head/lib/libc/sys/ptrace.2 297355 2016-03-28 16:48:28Z trasz $
.\"	$NetBSD: ptrace.2,v 1.2 1995/02/27 12:35:37 cgd Exp $
.\"
.\" This file is in the public domain.
.Dd December 29, 2015
.Dt PTRACE 2
.Os
.Sh NAME
.Nm ptrace
.Nd process tracing and debugging
.Sh LIBRARY
.Lb libc
.Sh SYNOPSIS
.In sys/types.h
.In sys/ptrace.h
.Ft int
.Fn ptrace "int request" "pid_t pid" "caddr_t addr" "int data"
.Sh DESCRIPTION
The
.Fn ptrace
system call
provides tracing and debugging facilities.
It allows one process
(the
.Em tracing
process)
to control another
(the
.Em traced
process).
The tracing process must first attach to the traced process, and then
issue a series of
.Fn ptrace
system calls to control the execution of the process, as well as access
process memory and register state.
For the duration of the tracing session, the traced process will be
.Dq re-parented ,
with its parent process ID (and resulting behavior)
changed to the tracing process.
It is permissible for a tracing process to attach to more than one
other process at a time.
When the tracing process has completed its work, it must detach the
traced process; if a tracing process exits without first detaching all
processes it has attached, those processes will be killed.
.Pp
Most of the time, the traced process runs normally, but when it
receives a signal
(see
.Xr sigaction 2 ) ,
it stops.
The tracing process is expected to notice this via
.Xr wait 2
or the delivery of a
.Dv SIGCHLD
signal, examine the state of the stopped process, and cause it to
terminate or continue as appropriate.
The signal may be a normal process signal, generated as a result of
traced process behavior, or use of the
.Xr kill 2
system call; alternatively, it may be generated by the tracing facility
as a result of attaching, system calls, or stepping by the tracing
process.
The tracing process may choose to intercept the signal, using it to
observe process behavior (such as
.Dv SIGTRAP ) ,
or forward the signal to the process if appropriate.
The
.Fn ptrace
system call
is the mechanism by which all this happens.
.Pp
The
.Fa request
argument specifies what operation is being performed; the meaning of
the rest of the arguments depends on the operation, but except for one
special case noted below, all
.Fn ptrace
calls are made by the tracing process, and the
.Fa pid
argument specifies the process ID of the traced process
or a corresponding thread ID.
The
.Fa request
argument
can be:
.Bl -tag -width 12n
.It Dv PT_TRACE_ME
This request is the only one used by the traced process; it declares
that the process expects to be traced by its parent.
All the other arguments are ignored.
(If the parent process does not expect to trace the child, it will
probably be rather confused by the results; once the traced process
stops, it cannot be made to continue except via
.Fn ptrace . )
When a process has used this request and calls
.Xr execve 2
or any of the routines built on it
(such as
.Xr execv 3 ) ,
it will stop before executing the first instruction of the new image.
Also, any setuid or setgid bits on the executable being executed will
be ignored.
If the child was created by
.Xr vfork 2
system call or
.Xr rfork 2
call with the
.Dv RFMEM
flag specified, the debugging events are reported to the parent
only after the
.Xr execve 2
is executed.
.It Dv PT_READ_I , Dv PT_READ_D
These requests read a single
.Vt int
of data from the traced process's address space.
Traditionally,
.Fn ptrace
has allowed for machines with distinct address spaces for instruction
and data, which is why there are two requests: conceptually,
.Dv PT_READ_I
reads from the instruction space and
.Dv PT_READ_D
reads from the data space.
In the current
.Fx
implementation, these two requests are completely identical.
The
.Fa addr
argument specifies the address
(in the traced process's virtual address space)
at which the read is to be done.
This address does not have to meet any alignment constraints.
The value read is returned as the return value from
.Fn ptrace .
.It Dv PT_WRITE_I , Dv PT_WRITE_D
These requests parallel
.Dv PT_READ_I
and
.Dv PT_READ_D ,
except that they write rather than read.
The
.Fa data
argument supplies the value to be written.
.It Dv PT_IO
This request allows reading and writing arbitrary amounts of data in
the traced process's address space.
The
.Fa addr
argument specifies a pointer to a
.Vt "struct ptrace_io_desc" ,
which is defined as follows:
.Bd -literal
struct ptrace_io_desc {
	int	piod_op;	/* I/O operation */
	void	*piod_offs;	/* child offset */
	void	*piod_addr;	/* parent offset */
	size_t	piod_len;	/* request length */
};

/*
 * Operations in piod_op.
 */
#define PIOD_READ_D	1	/* Read from D space */
#define PIOD_WRITE_D	2	/* Write to D space */
#define PIOD_READ_I	3	/* Read from I space */
#define PIOD_WRITE_I	4	/* Write to I space */
.Ed
.Pp
The
.Fa data
argument is ignored.
The actual number of bytes read or written is stored in
.Va piod_len
upon return.
.It Dv PT_CONTINUE
The traced process continues execution.
The
.Fa addr
argument
is an address specifying the place where execution is to be resumed
(a new value for the program counter),
or
.Po Vt caddr_t Pc Ns 1
to indicate that execution is to pick up where it left off.
The
.Fa data
argument
provides a signal number to be delivered to the traced process as it
resumes execution, or 0 if no signal is to be sent.
.It Dv PT_STEP
The traced process is single stepped one instruction.
The
.Fa addr
argument
should be passed
.Po Vt caddr_t Pc Ns 1 .
The
.Fa data
argument
provides a signal number to be delivered to the traced process as it
resumes execution, or 0 if no signal is to be sent.
.It Dv PT_KILL
The traced process terminates, as if
.Dv PT_CONTINUE
had been used with
.Dv SIGKILL
given as the signal to be delivered.
.It Dv PT_ATTACH
This request allows a process to gain control of an otherwise
unrelated process and begin tracing it.
It does not need any cooperation from the to-be-traced process.
In
this case,
.Fa pid
specifies the process ID of the to-be-traced process, and the other
two arguments are ignored.
This request requires that the target process must have the same real
UID as the tracing process, and that it must not be executing a setuid
or setgid executable.
(If the tracing process is running as root, these restrictions do not
apply.)
The tracing process will see the newly-traced process stop and may
then control it as if it had been traced all along.
.It Dv PT_DETACH
This request is like PT_CONTINUE, except that it does not allow
specifying an alternate place to continue execution, and after it
succeeds, the traced process is no longer traced and continues
execution normally.
.It Dv PT_GETREGS
This request reads the traced process's machine registers into the
.Do
.Vt "struct reg"
.Dc
(defined in
.In machine/reg.h )
pointed to by
.Fa addr .
.It Dv PT_SETREGS
This request is the converse of
.Dv PT_GETREGS ;
it loads the traced process's machine registers from the
.Do
.Vt "struct reg"
.Dc
(defined in
.In machine/reg.h )
pointed to by
.Fa addr .
.It Dv PT_GETFPREGS
This request reads the traced process's floating-point registers into
the
.Do
.Vt "struct fpreg"
.Dc
(defined in
.In machine/reg.h )
pointed to by
.Fa addr .
.It Dv PT_SETFPREGS
This request is the converse of
.Dv PT_GETFPREGS ;
it loads the traced process's floating-point registers from the
.Do
.Vt "struct fpreg"
.Dc
(defined in
.In machine/reg.h )
pointed to by
.Fa addr .
.It Dv PT_GETDBREGS
This request reads the traced process's debug registers into
the
.Do
.Vt "struct dbreg"
.Dc
(defined in
.In machine/reg.h )
pointed to by
.Fa addr .
.It Dv PT_SETDBREGS
This request is the converse of
.Dv PT_GETDBREGS ;
it loads the traced process's debug registers from the
.Do
.Vt "struct dbreg"
.Dc
(defined in
.In machine/reg.h )
pointed to by
.Fa addr .
.It Dv PT_LWPINFO
This request can be used to obtain information about the kernel thread,
also known as light-weight process, that caused the traced process to stop.
The
.Fa addr
argument specifies a pointer to a
.Vt "struct ptrace_lwpinfo" ,
which is defined as follows:
.Bd -literal
struct ptrace_lwpinfo {
	lwpid_t pl_lwpid;
	int	pl_event;
	int	pl_flags;
	sigset_t pl_sigmask;
	sigset_t pl_siglist;
	siginfo_t pl_siginfo;
	char	pl_tdname[MAXCOMLEN + 1];
	pid_t	pl_child_pid;
	u_int	pl_syscall_code;
	u_int	pl_syscall_narg;
};
.Ed
.Pp
The
.Fa data
argument is to be set to the size of the structure known to the caller.
This allows the structure to grow without affecting older programs.
.Pp
The fields in the
.Vt "struct ptrace_lwpinfo"
have the following meaning:
.Bl -tag -width indent -compact
.It pl_lwpid
LWP id of the thread
.It pl_event
Event that caused the stop.
Currently defined events are
.Bl -tag -width indent -compact
.It PL_EVENT_NONE
No reason given
.It PL_EVENT_SIGNAL
Thread stopped due to the pending signal
.El
.It pl_flags
Flags that specify additional details about observed stop.
Currently defined flags are:
.Bl -tag -width indent -compact
.It PL_FLAG_SCE
The thread stopped due to system call entry, right after the kernel is entered.
The debugger may examine syscall arguments that are stored in memory and
registers according to the ABI of the current process, and modify them,
if needed.
.It PL_FLAG_SCX
The thread is stopped immediately before syscall is returning to the usermode.
The debugger may examine system call return values in the ABI-defined registers
and/or memory.
.It PL_FLAG_EXEC
When
.Dv PL_FLAG_SCX
is set, this flag may be additionally specified to inform that the
program being executed by debuggee process has been changed by successful
execution of a system call from the
.Fn execve 2
family.
.It PL_FLAG_SI
Indicates that
.Va pl_siginfo
member of
.Vt "struct ptrace_lwpinfo"
contains valid information.
.It PL_FLAG_FORKED
Indicates that the process is returning from a call to
.Fn fork 2
that created a new child process.
The process identifier of the new process is available in the
.Va pl_child_pid
member of
.Vt "struct ptrace_lwpinfo" .
.It PL_FLAG_CHILD
The flag is set for first event reported from a new child, which is
automatically attached due to
.Dv PT_FOLLOW_FORK
enabled.
.It PL_FLAG_BORN
This flag is set for the first event reported from a new LWP when LWP
events are enabled via
.Dv PT_LWP_EVENTS .
It is reported along with
.Dv PL_FLAG_SCX
and is always reported if LWP events are enabled.
.It PL_FLAG_EXITED
This flag is set for the last event reported by an exiting LWP when
LWP events are enabled via
.Dv PT_LWP_EVENTS .
Note that this event is not reported when the last LWP in a process exits.
The termination of the last thread is reported via a normal process exit
event.
.El
.It pl_sigmask
The current signal mask of the LWP
.It pl_siglist
The current pending set of signals for the LWP.
Note that signals that are delivered to the process would not appear
on an LWP siglist until the thread is selected for delivery.
.It pl_siginfo
The siginfo that accompanies the signal pending.
Only valid for
.Dv PL_EVENT_SIGNAL
stop when
.Dv PL_FLAG_SI
is set in
.Va pl_flags .
.It pl_tdname
The name of the thread.
.It pl_child_pid
The process identifier of the new child process.
Only valid for a
.Dv PL_EVENT_SIGNAL
stop when
.Dv PL_FLAG_FORKED
is set in
.Va pl_flags .
.It pl_syscall_code
The ABI-specific identifier of the current system call.
Note that for indirect system calls this field reports the indirected
system call.
Only valid when
.Dv PL_FLAG_SCE
or
.Dv PL_FLAG_SCX
is set in
.Va pl_flags.
.It pl_syscall_narg
The number of arguments passed to the current system call not counting
the system call identifier.
Note that for indirect system calls this field reports the arguments
passed to the indirected system call.
Only valid when
.Dv PL_FLAG_SCE
or
.Dv PL_FLAG_SCX
is set in
.Va pl_flags.
.El
.It PT_GETNUMLWPS
This request returns the number of kernel threads associated with the
traced process.
.It PT_GETLWPLIST
This request can be used to get the current thread list.
A pointer to an array of type
.Vt lwpid_t
should be passed in
.Fa addr ,
with the array size specified by
.Fa data .
The return value from
.Fn ptrace
is the count of array entries filled in.
.It PT_SETSTEP
This request will turn on single stepping of the specified process.
.It PT_CLEARSTEP
This request will turn off single stepping of the specified process.
.It PT_SUSPEND
This request will suspend the specified thread.
.It PT_RESUME
This request will resume the specified thread.
.It PT_TO_SCE
This request will trace the specified process on each system call entry.
.It PT_TO_SCX
This request will trace the specified process on each system call exit.
.It PT_SYSCALL
This request will trace the specified process
on each system call entry and exit.
.It PT_FOLLOW_FORK
This request controls tracing for new child processes of a traced process.
If
.Fa data
is non-zero,
then new child processes will enable tracing and stop before executing their
first instruction.
If
.Fa data
is zero, then new child processes will execute without tracing enabled.
By default, tracing is not enabled for new child processes.
Child processes do not inherit this property.
The traced process will set the
.Dv PL_FLAG_FORKED
flag upon exit from a system call that creates a new process.
.It PT_LWP_EVENTS
This request controls tracing of LWP creation and destruction.
If
.Fa data
is non-zero,
then LWPs will stop to report creation and destruction events.
If
.Fa data
is zero,
then LWP creation and destruction events will not be reported.
By default, tracing is not enabled for LWP events.
Child processes do not inherit this property.
New LWPs will stop to report an event with
.Dv PL_FLAG_BORN
set before executing their first instruction.
Exiting LWPs will stop to report an event with
.Dv PL_FLAG_EXITED
set before completing their termination.
.Pp
Note that new processes do not report an event for the creation of their
initial thread,
and exiting processes do not report an event for the termination of the
last thread.
.It PT_VM_TIMESTAMP
This request returns the generation number or timestamp of the memory map of
the traced process as the return value from
.Fn ptrace .
This provides a low-cost way for the tracing process to determine if the
VM map changed since the last time this request was made.
.It PT_VM_ENTRY
This request is used to iterate over the entries of the VM map of the traced
process.
The
.Fa addr
argument specifies a pointer to a
.Vt "struct ptrace_vm_entry" ,
which is defined as follows:
.Bd -literal
struct ptrace_vm_entry {
	int		pve_entry;
	int		pve_timestamp;
	u_long		pve_start;
	u_long		pve_end;
	u_long		pve_offset;
	u_int		pve_prot;
	u_int		pve_pathlen;
	long		pve_fileid;
	uint32_t	pve_fsid;
	char		*pve_path;
};
.Ed
.Pp
The first entry is returned by setting
.Va pve_entry
to zero.
Subsequent entries are returned by leaving
.Va pve_entry
unmodified from the value returned by previous requests.
The
.Va pve_timestamp
field can be used to detect changes to the VM map while iterating over the
entries.
The tracing process can then take appropriate action, such as restarting.
By setting
.Va pve_pathlen
to a non-zero value on entry, the pathname of the backing object is returned
in the buffer pointed to by
.Va pve_path ,
provided the entry is backed by a vnode.
The
.Va pve_pathlen
field is updated with the actual length of the pathname (including the
terminating null character).
The
.Va pve_offset
field is the offset within the backing object at which the range starts.
The range is located in the VM space at
.Va pve_start
and extends up to
.Va pve_end
(inclusive).
.Pp
The
.Fa data
argument is ignored.
.El
.Sh x86 MACHINE-SPECIFIC REQUESTS
.Bl -tag -width "Dv PT_GETXSTATE_INFO"
.It Dv PT_GETXMMREGS
Copy the XMM FPU state into the buffer pointed to by the
argument
.Fa addr .
The buffer has the same layout as the 32-bit save buffer for the
machine instruction
.Dv FXSAVE .
.Pp
This request is only valid for i386 programs, both on native 32-bit
systems and on amd64 kernels.
For 64-bit amd64 programs, the XMM state is reported as part of
the FPU state returned by the
.Dv PT_GETFPREGS
request.
.Pp
The
.Fa data
argument is ignored.
.It Dv PT_SETXMMREGS
Load the XMM FPU state for the thread from the buffer pointed to
by the argument
.Fa addr .
The buffer has the same layout as the 32-bit load buffer for the
machine instruction
.Dv FXRSTOR .
.Pp
As with
.Dv PT_GETXMMREGS,
this request is only valid for i386 programs.
.Pp
The
.Fa data
argument is ignored.
.It Dv PT_GETXSTATE_INFO
Report which XSAVE FPU extensions are supported by the CPU
and allowed in userspace programs.
The
.Fa addr
argument must point to a variable of type
.Vt struct ptrace_xstate_info ,
which contains the information on the request return.
.Vt struct ptrace_xstate_info
is defined as follows:
.Bd -literal
struct ptrace_xstate_info {
	uint64_t	xsave_mask;
	uint32_t	xsave_len;
};
.Ed
The
.Dv xsave_mask
field is a bitmask of the currently enabled extensions.
The meaning of the bits is defined in the Intel and AMD
processor documentation.
The
.Dv xsave_len
field reports the length of the XSAVE area for storing the hardware
state for currently enabled extensions in the format defined by the x86
.Dv XSAVE
machine instruction.
.Pp
The
.Fa data
argument value must be equal to the size of the
.Vt struct ptrace_xstate_info .
.It Dv PT_GETXSTATE
Return the content of the XSAVE area for the thread.
The
.Fa addr
argument points to the buffer where the content is copied, and the
.Fa data
argument specifies the size of the buffer.
The kernel copies out as much content as allowed by the buffer size.
The buffer layout is specified by the layout of the save area for the
.Dv XSAVE
machine instruction.
.It Dv PT_SETXSTATE
Load the XSAVE state for the thread from the buffer specified by the
.Fa addr
pointer.
The buffer size is passed in the
.Fa data
argument.
The buffer must be at least as large as the
.Vt struct savefpu
(defined in
.Pa x86/fpu.h )
to allow the complete x87 FPU and XMM state load.
It must not be larger than the XSAVE state length, as reported by the
.Dv xsave_len
field from the
.Vt struct ptrace_xstate_info
of the
.Dv PT_GETXSTATE_INFO
request.
Layout of the buffer is identical to the layout of the load area for the
.Dv XRSTOR
machine instruction.
.It Dv PT_GETFSBASE
Return the value of the base used when doing segmented
memory addressing using the %fs segment register.
The
.Fa addr
argument points to an
.Vt unsigned long
variable where the base value is stored.
.Pp
The
.Fa data
argument is ignored.
.It Dv PT_GETGSBASE
Like the
.Dv PT_GETFSBASE
request, but returns the base for the %gs segment register.
.It Dv PT_SETFSBASE
Set the base for the %fs segment register to the value pointed to
by the
.Fa addr
argument.
.Fa addr
must point to the
.Vt unsigned long
variable containing the new base.
.Pp
The
.Fa data
argument is ignored.
.It Dv PT_SETGSBASE
Like the
.Dv PT_SETFSBASE
request, but sets the base for the %gs segment register.
.El
.Sh PowerPC MACHINE-SPECIFIC REQUESTS
.Bl -tag -width "Dv PT_SETVRREGS"
.It Dv PT_GETVRREGS
Return the thread's
.Dv ALTIVEC
machine state in the buffer pointed to by
.Fa addr .
.Pp
The
.Fa data
argument is ignored.
.It Dv PT_SETVRREGS
Set the thread's
.Dv ALTIVEC
machine state from the buffer pointed to by
.Fa addr .
.Pp
The
.Fa data
argument is ignored.
.El
.Pp
Additionally, other machine-specific requests can exist.
.Sh RETURN VALUES
Some requests can cause
.Fn ptrace
to return
\-1
as a non-error value; to disambiguate,
.Va errno
can be set to 0 before the call and checked afterwards.
.Sh ERRORS
The
.Fn ptrace
system call may fail if:
.Bl -tag -width Er
.It Bq Er ESRCH
.Bl -bullet -compact
.It
No process having the specified process ID exists.
.El
.It Bq Er EINVAL
.Bl -bullet -compact
.It
A process attempted to use
.Dv PT_ATTACH
on itself.
.It
The
.Fa request
argument
was not one of the legal requests.
.It
The signal number
(in
.Fa data )
to
.Dv PT_CONTINUE
was neither 0 nor a legal signal number.
.It
.Dv PT_GETREGS ,
.Dv PT_SETREGS ,
.Dv PT_GETFPREGS ,
.Dv PT_SETFPREGS ,
.Dv PT_GETDBREGS ,
or
.Dv PT_SETDBREGS
was attempted on a process with no valid register set.
(This is normally true only of system processes.)
.It
.Dv PT_VM_ENTRY
was given an invalid value for
.Fa pve_entry .
This can also be caused by changes to the VM map of the process.
.It
The size (in
.Fa data )
provided to
.Dv PT_LWPINFO
was less than or equal to zero, or larger than the
.Vt ptrace_lwpinfo
structure known to the kernel.
.It
The size (in
.Fa data )
provided to the x86-specific
.Dv PT_GETXSTATE_INFO
request was not equal to the size of the
.Vt struct ptrace_xstate_info .
.It
The size (in
.Fa data )
provided to the x86-specific
.Dv PT_SETXSTATE
request was less than the size of the x87 plus the XMM save area.
.It
The size (in
.Fa data )
provided to the x86-specific
.Dv PT_SETXSTATE
request was larger than returned in the
.Dv xsave_len
member of the
.Vt struct ptrace_xstate_info
from the
.Dv PT_GETXSTATE_INFO
request.
.It
The base value, provided to the amd64-specific requests
.Dv PT_SETFSBASE
or
.Dv PT_SETGSBASE ,
pointed outside of the valid user address space.
This error will not occur in 32-bit programs.
.El
.It Bq Er EBUSY
.Bl -bullet -compact
.It
.Dv PT_ATTACH
was attempted on a process that was already being traced.
.It
A request attempted to manipulate a process that was being traced by
some process other than the one making the request.
.It
A request
(other than
.Dv PT_ATTACH )
specified a process that was not stopped.
.El
.It Bq Er EPERM
.Bl -bullet -compact
.It
A request
(other than
.Dv PT_ATTACH )
attempted to manipulate a process that was not being traced at all.
.It
An attempt was made to use
.Dv PT_ATTACH
on a process in violation of the requirements listed under
.Dv PT_ATTACH
above.
.El
.It Bq Er ENOENT
.Bl -bullet -compact
.It
.Dv PT_VM_ENTRY
previously returned the last entry of the memory map.
No more entries exist.
.El
.It Bq Er ENAMETOOLONG
.Bl -bullet -compact
.It
.Dv PT_VM_ENTRY
cannot return the pathname of the backing object because the buffer is not big
enough.
.Fa pve_pathlen
holds the minimum buffer size required on return.
.El
.El
.Sh SEE ALSO
.Xr execve 2 ,
.Xr sigaction 2 ,
.Xr wait 2 ,
.Xr execv 3 ,
.Xr i386_clr_watch 3 ,
.Xr i386_set_watch 3
.Sh HISTORY
The
.Fn ptrace
function appeared in
.At v7 .