ktrcsw, ktremul, ktrgenio, ktrnamei, ktrpsig, ktrsyscall,
ktrsysret,
KTRPOINT - process tracing kernel interface
#include <sys/param.h>
#include <sys/proc.h>
#include <sys/ktrace.h>
KTRPOINT(struct proc *p, int type);
void
ktrcsw(struct proc *p, int out, int user);
void
ktremul(struct proc *p, char *emul);
void
ktrgenio(struct proc *p, int fd, enum uio_rw rw, struct
iovec *iov,
int len, int error);
void
ktrnamei(struct proc *p, char *path);
void
ktrpsig(struct proc *p, int sig, sig_t action, int mask, int
code,
siginfo_t *si);
void
ktrsyscall(struct proc *p, register_t code, size_t argsize,
register_t args[]);
void
ktrsysret(struct proc *p, register_t code, int error,
register_t retval);
This interface is meant for kernel subsystems and machine
dependent code
to inform the user about the events occurring to the process
should tracing
of such be enabled using the ktrace(2) system call.
Each of the
functions (except for KTRPOINT) is meant for a particular
type of event
and is described below.
The KTRPOINT() macro should be used before calling any of
the other tracing
functions to verify that tracing for that particular
type of events
has been enabled. Possible values for the type argument are
a mask of
the KTRFAC_ values described in ktrace(2).
ktrcsw() is called during the context switching. The user
argument is a
boolean value specifying whether the process has been put
into a waiting
state (true) or placed onto a running queue (false). Furthermore the
user argument indicates whether a voluntary (false) or an
involuntary
(true) switching has happened.
ktremul() should be called every time emulation for the execution environment
is changed and thus the name of which is given in
the name argument.
ktrgenio() should be called for each generic input/output
transaction
that is described by the fd file descriptor, rw transaction
type (consult
sys/sys/uio.h for the enum uio_rw definition), iov input/output data vector,
len size of the iov vector, and, lastly, error status
of the transaction.
ktrnamei() should be called every time a namei(9) operation
is performed
over the path name.
ktrpsig() should be called for each signal sig posted for
the traced process.
The action taken is one of SIG_DFL, SIG_IGN, or
SIG_ERR as described
in the sigaction(2) document. mask is the current
traced process'
signal mask. Signal-specific code and siginfo_t
structure as described
in <sys/siginfo.h> are given in the code and si arguments respectively.
ktrsyscall() should be called for each system call number
code executed
with arguments in args of total count of argsize.
ktrsysret() should be called for a return from each system
call number
code and error number of error as described in errno(2) and
a return value
in retval that is syscall dependent.
The process tracing facility is implemented in
sys/kern/kern_ktrace.c.
errno(2), ktrace(2), syscall(2), namei(9), syscall(9)
The process tracing facility first appeared in 4.4BSD.
The ktrace section manual page appeared in OpenBSD 3.4.
OpenBSD 3.6 July 21, 2003
[ Back ] |