freebsd-dev/usr.bin/gprof/gprof.1
2008-12-25 08:48:08 +00:00

333 lines
9.7 KiB
Groff

.\" Copyright (c) 1983, 1990, 1993
.\" The Regents of the University of California. 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. All advertising materials mentioning features or use of this software
.\" must display the following acknowledgement:
.\" This product includes software developed by the University of
.\" California, Berkeley and its contributors.
.\" 4. Neither the name of the University nor the names of its contributors
.\" may be used to endorse or promote products derived from this software
.\" without specific prior written permission.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE REGENTS AND CONTRIBUTORS ``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 REGENTS OR CONTRIBUTORS 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.
.\"
.\" @(#)gprof.1 8.1 (Berkeley) 6/6/93
.\" $FreeBSD$
.\"
.Dd December 25, 2008
.Dt GPROF 1
.Os
.Sh NAME
.Nm gprof
.Nd display call graph profile data
.Sh SYNOPSIS
.Nm
.Op Fl abKlLsuz
.Op Fl C Ar count
.Op Fl e Ar name
.Op Fl E Ar name
.Op Fl f Ar name
.Op Fl F Ar name
.Op Fl k Ar fromname toname
.Op Ar a.out Op Ar a.out.gmon ...
.Sh DESCRIPTION
The
.Nm
utility produces an execution profile of C, Pascal, or Fortran77 programs.
The effect of called routines is incorporated in the profile of each caller.
The profile data is taken from the call graph profile file
which is created by programs that are compiled with the
.Fl pg
option of
.Xr cc 1 ,
.Xr pc 1 ,
and
.Xr f77 1 .
The
.Fl pg
option also links in versions of the library routines
that are compiled for profiling.
By convention these libraries have their name suffixed with
.Pa _p ,
i.e., the profiled version of
.Pa libc.a
is
.Pa libc_p.a
and if you specify libraries directly to the
compiler or linker you can use
.Fl l Ns Ar c_p
instead of
.Fl l Ns Ar c .
Read the given object file (the default is
.Pa a.out)
and establishes the relation between its symbol table
and the call graph profile.
The default graph profile file name is the name
of the executable with the suffix
.Pa .gmon
appended.
If more than one profile file is specified,
the
.Nm
output shows the sum of the profile information in the given profile files.
.Pp
The
.Nm
utility calculates the amount of time spent in each routine.
Next, these times are propagated along the edges of the call graph.
Cycles are discovered, and calls into a cycle are made to share the time
of the cycle.
The first listing shows the functions
sorted according to the time they represent
including the time of their call graph descendants.
Below each function entry is shown its (direct) call graph children,
and how their times are propagated to this function.
A similar display above the function shows how this function's time and the
time of its descendants is propagated to its (direct) call graph parents.
.Pp
Cycles are also shown, with an entry for the cycle as a whole and
a listing of the members of the cycle and their contributions to the
time and call counts of the cycle.
.Pp
Second, a flat profile is given,
similar to that provided by
.Xr prof 1 .
This listing gives the total execution times, the call counts,
the time that the call spent in the routine itself, and
the time that the call spent in the routine itself including
its descendants.
The units for the per-call times are normally milliseconds,
but they are nanoseconds if the profiling clock frequency
is 10 million or larger,
and if a function appears to be never called then its total self time
is printed as a percentage in the self time per call column.
The very high profiling clock frequencies needed to get sufficient
accuracy in the per-call times for short-lived programs are only
implemented for
.Dq high resolution
(non-statistical) kernel profiling.
.Pp
Finally, an index of the function names is provided.
.Pp
The following options are available:
.Bl -tag -width indent
.It Fl a
Suppress the printing of statically declared functions.
If this option is given, all relevant information about the static function
(e.g., time samples, calls to other functions, calls from other functions)
belongs to the function loaded just before the static function in the
.Pa a.out
file.
.It Fl b
Suppress the printing of a description of each field in the profile.
.It Fl C Ar count
Find a minimal set of arcs that can be broken to eliminate all cycles with
.Ar count
or more members.
Caution: the algorithm used to break cycles is exponential,
so using this option may cause
.Nm
to run for a very long time.
.It Fl e Ar name
Suppress the printing of the graph profile entry for routine
.Ar name
and all its descendants
(unless they have other ancestors that are not suppressed).
More than one
.Fl e
option may be given.
Only one
.Ar name
may be given with each
.Fl e
option.
.It Fl E Ar name
Suppress the printing of the graph profile entry for routine
.Ar name
(and its descendants) as
.Fl e ,
above, and also excludes the time spent in
.Ar name
(and its descendants) from the total and percentage time computations.
(For example,
.Fl E
.Ar mcount
.Fl E
.Ar mcleanup
is the default.)
.It Fl f Ar name
Print the graph profile entry of only the specified routine
.Ar name
and its descendants.
More than one
.Fl f
option may be given.
Only one
.Ar name
may be given with each
.Fl f
option.
.It Fl F Ar name
Print the graph profile entry of only the routine
.Ar name
and its descendants (as
.Fl f ,
above) and also uses only the times of the printed routines
in total time and percentage computations.
More than one
.Fl F
option may be given.
Only one
.Ar name
may be given with each
.Fl F
option.
The
.Fl F
option
overrides
the
.Fl E
option.
.It Fl k Ar fromname Ar toname
Will delete any arcs from routine
.Ar fromname
to routine
.Ar toname .
This can be used to break undesired cycles.
More than one
.Fl k
option may be given.
Only one pair of routine names may be given with each
.Fl k
option.
.It Fl K
Gather information about symbols from the currently-running kernel using the
.Xr sysctl 3
and
.Xr kldsym 2
interfaces.
This forces the
.Pa a.out
argument to be ignored, and allows for symbols in
.Xr kld 4
modules to be used.
.It Fl l
Suppress the printing of the call-graph profile.
.It Fl L
Suppress the printing of the flat profile.
.It Fl s
A profile file
.Pa gmon.sum
is produced that represents
the sum of the profile information in all the specified profile files.
This summary profile file may be given to later
executions of gprof (probably also with a
.Fl s )
to accumulate profile data across several runs of an
.Pa a.out
file.
.It Fl u
Suppress the printing of functions whose names are not visible to
C programs.
For the ELF object format, this means names that
contain the
.Ql .\&
character.
For the a.out object format, it means names that do not
begin with a
.Ql _
character.
All relevant information about such functions belongs to the
(non-suppressed) function with the next lowest address.
This is useful for eliminating "functions" that are just labels
inside other functions.
.It Fl z
Display routines that have zero usage (as shown by call counts
and accumulated time).
.El
.Sh FILES
.Bl -tag -width a.out.gmon -compact
.It Pa a.out
The namelist and text space.
.It Pa a.out.gmon
Dynamic call graph and profile.
.It Pa gmon.sum
Summarized dynamic call graph and profile.
.El
.Sh SEE ALSO
.Xr cc 1 ,
.Xr profil 2 ,
.Xr clocks 7
.\" .Xr monitor 3 ,
.\" .Xr prof 1
.Rs
.%T "An Execution Profiler for Modular Programs"
.%A S. Graham
.%A P. Kessler
.%A M. McKusick
.%J "Software - Practice and Experience"
.%V 13
.%P pp. 671-685
.%D 1983
.Re
.Rs
.%T "gprof: A Call Graph Execution Profiler"
.%A S. Graham
.%A P. Kessler
.%A M. McKusick
.%J "Proceedings of the SIGPLAN '82 Symposium on Compiler Construction, SIGPLAN Notices"
.%V 17
.%N 6
.%P pp. 120-126
.%D June 1982
.Re
.Sh HISTORY
The
.Nm
profiler
appeared in
.Bx 4.2 .
.Sh BUGS
The granularity of the sampling is shown, but remains
statistical at best.
We assume that the time for each execution of a function
can be expressed by the total time for the function divided
by the number of times the function is called.
Thus the time propagated along the call graph arcs to the function's
parents is directly proportional to the number of times that
arc is traversed.
.Pp
Parents that are not themselves profiled will have the time of
their profiled children propagated to them, but they will appear
to be spontaneously invoked in the call graph listing, and will
not have their time propagated further.
Similarly, signal catchers, even though profiled, will appear
to be spontaneous (although for more obscure reasons).
Any profiled children of signal catchers should have their times
propagated properly, unless the signal catcher was invoked during
the execution of the profiling routine, in which case all is lost.
.Pp
The profiled program must call
.Xr exit 3
or return normally for the profiling information to be saved
in the graph profile file.