Commit Graph

48 Commits

Author SHA1 Message Date
Garrett Wollman
0fb7a0beb0 Fix preprocessor directive syntax (text after #endif).
The style of this program is still atrocious (not fixed).
2002-05-30 21:18:01 +00:00
Alfred Perlstein
e043516d4d Check for defined(__i386__) instead of just defined(i386) since the compiler
will be updated to only define(__i386__) for ANSI cleanliness.
2002-05-30 07:00:42 +00:00
Philippe Charnier
e8937ba009 Use `The .Nm utility' 2002-04-20 12:18:28 +00:00
Mike Barcroft
0ec63e60db Include <stdint.h> for definition of intmax_t and/or uintmax_t. 2002-03-25 05:23:45 +00:00
Jake Burkholder
9f5bcb6350 Make this build on sparc64. 2002-03-10 01:25:47 +00:00
Bruce Evans
275ffbc6e6 Support runtime specification of the history counter type by whatever
generated the gmon data.  The support is currently limited to what is
easy to implement and/or needed:

    signedess: signed or insigned
    size: 8, 16, 32 or 64 bits
    format: a binary integer in gprof's format (gprof is not a cross-tool).

High-resolution kernel profiling uses signed 64-bit counters.  Normal
kernel profiling and user profiling use unsigned 16-bit counters but
should use 32-bit ones.
2002-03-06 11:25:58 +00:00
Bruce Evans
8109a9af1f Fixed some misspellings of 2 as sizeof(UNIT) so that they won't break
things when sizeof(UNIT) becomes a runtime parameter.  The relevant 2
is the one in profil(2)'s scaling of pc's to bucket numbers:

	   bucket = (pc - offset) / 2 * profil_scale / 65536

gprof(1) must duplicate this scaling, bug for bug compatibly, so it
must first do an integer division by 2 although this mainly makes
scales larger than 65536 useless.  sizeof(UNIT) was already wrong in
gprof4, but there were no problems because the fake profil scale is a
multiple of 2.

There are also some rounding bugs in the scaling, but these are only
problems if profil(2) is used directly to create unusual (and not
useful) scales.
2002-03-06 09:47:36 +00:00
Bruce Evans
8d57b8d3d2 Fixed printf format errors. In printgprof.c, also convert the scale
without possibly losing lots of precision, and print the scale using
%g instead of %d in case it is non-integral.  %g might not be the best
format for this.
2002-02-21 12:07:21 +00:00
Bruce Evans
312a529b81 Added missing copyright. Obtain one and a vendor id from gprof.c. aout.c
was split off from gprof.c in rev.1.7 of the latter.  elf.c is mostly new,
but the old copyright sort of applies to it and is better than none.

Use __FBSDID() for the FreeBSD id.
2002-02-21 12:03:06 +00:00
Bruce Evans
07ac83f0b0 Fixed a missing variable declaration so that gprof compiles with -DDEBUG. 2002-02-21 10:34:52 +00:00
Bruce Evans
4cd016e8fd Moved the definition of the machine-independent macro UNITS_TO_CODE
from <number of machines> machine-dependent headers to the one
non-header here it is used so that it is easier to fix.  This macro
just divides the machine-dependent offset OFFSET_OF_CODE by the
machine-independent scale factor sizeof(UNIT), as required for bug
for bug compatibility with the scaling of pc's in gprof.c.  UNIT is
the type of a profiling counter, and its size has nothing to do with
the correct scale factor except both are usually 2.
2002-02-21 07:12:57 +00:00
Ruslan Ermilov
2bdd8946cb Oops, forgot to remove the `beforeinstall' block. 2001-12-17 16:04:40 +00:00
Ruslan Ermilov
ccaec6d133 Install files via FILES. 2001-12-17 16:02:15 +00:00
Ruslan Ermilov
681c4d3c61 mdoc(7) police: kill whitespace at EOL. 2001-11-20 15:49:18 +00:00
Brian Feldman
3fc980b135 Add -K support to gprof(1), which enables dynamic symbol resolution from
the currently-running kernel (and supercedes an executable file argument
given).  With this change, properly-compiled KLD modules are now
able to be profiled.

Obtained from:	NAI Labs CBOSS project
Funded by:	DARPA
2001-10-30 15:54:09 +00:00
Peter Wemm
5b29dc6b1f Add ia64 support (stubs, just like i386, alpha) 2001-10-23 10:12:10 +00:00
Jens Schweikhardt
fa7a1ab968 Add hint about how we name profiled libraries (append _p at the right place).
PR:		8061
Submitted by:	Brandon Gillespie <brandon@ice.cold.org>
Reviewed by:	joerg, ru
MFC after:	5 weeks
2001-08-14 17:48:29 +00:00
Dima Dorfman
f247324df7 Remove whitespace at EOL. 2001-07-15 08:06:20 +00:00
Ruslan Ermilov
625003720a mdoc(7) police: removed HISTORY info from the .Os call. 2001-07-10 14:16:33 +00:00
David E. O'Brien
5e6220d9d0 * include/elf.h has been repo copied to include/elf-hints.h, and it no
longer includes machine/elf.h.
* consumers of elf.h now use the minimalist elf header possible.

This change is motivated by Binutils 2.11.0 and too much clashing over
our base elf headers and the Binutils elf headers.
2001-05-02 23:56:21 +00:00
Dima Dorfman
5e75e35cca Grammar police: "its", not "it's", is the possessive form of "it". 2001-04-15 19:53:47 +00:00
Ruslan Ermilov
9b88faecd3 Prepare for mdoc(7)NG. 2000-12-19 16:00:12 +00:00
Ruslan Ermilov
8fe908ef0c mdoc(7) police: use the new features of the Nm macro. 2000-11-20 19:21:22 +00:00
Sheldon Hearn
87faa07bec Remove single-space hard sentence breaks. These degrade the quality
of the typeset output, tend to make diffs harder to read and provide
bad examples for new-comers to mdoc.
2000-03-01 12:20:22 +00:00
Peter Wemm
c3aac50f28 $Id$ -> $FreeBSD$ 1999-08-28 01:08:13 +00:00
Hidetoshi Shimokawa
13a29a8a17 Enable gprof on alpha.
* alpha.{c,h} are same as i386.{c,h}.
* Force address calculation to be done in long precision(64bit on alpha)
  rather than double precision(52bit).
1999-07-16 07:22:10 +00:00
Nik Clayton
3be5f1f5ce Add $Id$, to make it simpler for members of the translation teams to
track.

The $Id$ line is normally at the bottom of the main comment block in the
man page, separated from the rest of the manpage by an empty comment,
like so;

     .\"    $Id$
     .\"

If the immediately preceding comment is a @(#) format ID marker than the
the $Id$ will line up underneath it with no intervening blank lines.
Otherwise, an additional blank line is inserted.

Approved by:            bde
1999-07-12 20:24:20 +00:00
Dmitrij Tejblum
aac6614bdc Don't ignore weak symbols.
EGCS assign weak symbols to inline functions it couldn't inline (e.g. virtual
inline functions), template functions, etc. Omitting them result in quite bogus
profile.

Weak symbols created by __weak_reference are not really problem.

Caught by:	Ilya Segalovich <iseg@comptek.ru>
1999-07-03 12:30:04 +00:00
Jean-Marc Zucconi
7e74cac42e Construct the profile file name from the name of the executable. A program
compiled with -pg and run will generate a file <executable-filename>.gmon,
not gmon.out.

PR:		bin/8426
1999-05-23 00:37:56 +00:00
John Polstra
5584f22bb3 Make profiling work for ELF. gprof now autodetects the format of
the executable file, so it will work for both a.out and ELF format
files.  I have split the object format specific code into separate
source files.  It's cleaner than it was before, but it's still
pretty crufty.

Don't cheat on your make world for this update.  A lot of things
have to be rebuilt for it to work, including the compiler and all
of the profiled libraries.
1998-09-07 23:32:00 +00:00
John Birrell
db36ef8f02 Change MACHINE to MACHINE_ARCH to support MACHINE=pc98. 1998-09-05 08:22:30 +00:00
John Polstra
43ab5f3877 Don't ignore symbols containing "$". C++ destructors and other
special functions have names containing dollar signs, and ignoring
them causes gprof to produce incorrect and sometimes bizarre results.
The comment in the original code said that dollar signs were excluded
because they are used in Pascal labels.  That's not much of an
issue these days.
1998-08-08 17:48:26 +00:00
Philippe Charnier
297b94929a Revert to original style. 1997-07-15 08:04:40 +00:00
Bruce Evans
ffbce11fea 32-bit counters aren't large enough for 100+MHz clocks. Use 64-bit
counters.  `4' in GPROF4 and gprof4 now means 8.  gprof4 needs to be
recompiled to match the kernel.
1997-07-13 16:38:39 +00:00
Philippe Charnier
6451b2bfcb Fix typo I introduced during last commit. 1997-07-11 06:11:35 +00:00
Philippe Charnier
b34f7debd8 Use err(3). 1997-07-10 06:45:02 +00:00
Wolfram Schneider
da1ff3cb8a Sort cross references. 1997-01-15 23:25:55 +00:00
Bruce Evans
db6a4b8826 Use a (signed) int32_t counter instead of an `unsigned int' counter
for the GPROF4 case.  This allows a simpler method to be used for
non-statistical profiling (it allows overhead adjustments to be
subtracted from one counter without harm if that counter goes
negative; otherwise the adjustment would have to be distributed).

32 bit counters were already too small for GPROF4 with a 200MHz
clock.  int64_t counters should be used.
1996-10-16 21:02:49 +00:00
Bruce Evans
d4aa237228 Print times/call in ns if hz >= 10e7. hz will be this large for high
resolution profiling on Pentiums.  On a 100MHz Pentium, the resolution
is at best 10 ns and actually a few hundred ns, but units of 10's or
100's of ns would be inconvenient and the current units of 1 us are a
bit too coarse.
1996-06-08 12:29:57 +00:00
Mike Pritchard
906c1e27fc Correct some man page cross references and file location references. 1996-04-07 00:06:21 +00:00
Joerg Wunsch
9cb8a1050f Xref clocks(7). 1996-04-05 08:53:38 +00:00
Mike Pritchard
bcff8e2ae4 Another round of man page cleanups.
Down to only about 100 items left to cleanup! :-)
1996-02-12 04:57:03 +00:00
Bruce Evans
e6c645fad2 Implemented non-statistical kernel profiling. This is based on
looking at a high resolution clock for each of the following events:
function call, function return, interrupt entry, interrupt exit,
and interesting branches.  The differences between the times of
these events are added at appropriate places in a ordinary histogram
(as if very fast statistical profiling sampled the pc at those
places) so that ordinary gprof can be used to analyze the times.

gmon.h:
Histogram counters need to be 4 bytes for microsecond resolutions.
They will need to be larger for the 586 clock.
The comments were vax-centric and wrong even on vaxes.  Does anyone
disagree?

gprof4.c:
The standard gprof should support counters of all integral sizes
and the size of the counter should be in the gmon header.  This
hack will do until then.  (Use gprof4 -u to examine the results
of non-statistical profiling.)

config/*:
Non-statistical profiling is configured with `config -pp'.
`config -p' still gives ordinary profiling.

kgmon/*:
Non-statistical profiling is enabled with `kgmon -B'.  `kgmon -b'
still enables ordinary profiling (and distables non-statistical
profiling) if non-statistical profiling is configured.
1995-12-29 15:46:59 +00:00
Bruce Evans
9c0dc173cc Change install' to ${INSTALL}' so that default install flags can be
specified in the top level Makefiles.
1995-07-25 00:37:58 +00:00
Rodney W. Grimes
7799f52a32 Remove trailing whitespace. 1995-05-30 06:41:30 +00:00
Garrett Wollman
d87bdc80f2 Added much-needed new options -[lL], to suppress printing of either the
call-graph or the flat profile, since often you only want one of them.
1994-12-22 20:52:15 +00:00
Bruce Evans
ff18d3781d New flag -u to suppress functions whose name does not begin with an
underscore.  Use it to avoid seeing badsw when profiling the kernel.

Print times more accurately (e.g. usec in %8.0f format instead of
msec in %8.2f format for averages) if hz >= 10000.  This should have
no effect now since profhz is only 1024.
1994-09-05 16:14:54 +00:00
Rodney W. Grimes
9b50d90275 BSD 4.4 Lite Usr.bin Sources 1994-05-27 12:33:43 +00:00