1 .\" Automatically generated by Pod::Man version 1.16
2 .\" Mon Jun 25 00:14:36 2001
5 .\" ======================================================================
6 .de Sh \" Subsection heading
14 .de Sp \" Vertical space (when we can't use .PP)
20 .ie \\n(.$>=3 .ne \\$3
24 .de Vb \" Begin verbatim text
29 .de Ve \" End verbatim text
34 .\" Set up some character translations and predefined strings. \*(-- will
35 .\" give an unbreakable dash, \*(PI will give pi, \*(L" will give a left
36 .\" double quote, and \*(R" will give a right double quote. | will give a
37 .\" real vertical bar. \*(C+ will give a nicer C++. Capital omega is used
38 .\" to do unbreakable dashes and therefore won't be available. \*(C` and
39 .\" \*(C' expand to `' in nroff, nothing in troff, for use with C<>
41 .ds C+ C\v'-.1v'\h'-1p'\s-2+\h'-1p'+\s0\v'.1v'\h'-1p'
45 . if (\n(.H=4u)&(1m=24u) .ds -- \(*W\h'-12u'\(*W\h'-12u'-\" diablo 10 pitch
46 . if (\n(.H=4u)&(1m=20u) .ds -- \(*W\h'-12u'\(*W\h'-8u'-\" diablo 12 pitch
59 .\" If the F register is turned on, we'll generate index entries on stderr
60 .\" for titles (.TH), headers (.SH), subsections (.Sh), items (.Ip), and
61 .\" index entries marked with X<> in POD. Of course, you'll have to process
62 .\" the output yourself in some meaningful fashion.
65 . tm Index:\\$1\t\\n%\t"\\$2"
71 .\" For nroff, turn off justification. Always turn off hyphenation; it
72 .\" makes way too many mistakes in technical documents.
76 .\" Accent mark definitions (@(#)ms.acc 1.5 88/02/08 SMI; from UCB 4.2).
77 .\" Fear. Run. Save yourself. No user-serviceable parts.
79 . \" fudge factors for nroff and troff
88 . ds #H ((1u-(\\\\n(.fu%2u))*.13m)
94 . \" simple accents for nroff and troff
104 . ds ' \\k:\h'-(\\n(.wu*8/10-\*(#H)'\'\h"|\\n:u"
105 . ds ` \\k:\h'-(\\n(.wu*8/10-\*(#H)'\`\h'|\\n:u'
106 . ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'^\h'|\\n:u'
107 . ds , \\k:\h'-(\\n(.wu*8/10)',\h'|\\n:u'
108 . ds ~ \\k:\h'-(\\n(.wu-\*(#H-.1m)'~\h'|\\n:u'
109 . ds / \\k:\h'-(\\n(.wu*8/10-\*(#H)'\z\(sl\h'|\\n:u'
111 . \" troff and (daisy-wheel) nroff accents
112 .ds : \\k:\h'-(\\n(.wu*8/10-\*(#H+.1m+\*(#F)'\v'-\*(#V'\z.\h'.2m+\*(#F'.\h'|\\n:u'\v'\*(#V'
113 .ds 8 \h'\*(#H'\(*b\h'-\*(#H'
114 .ds o \\k:\h'-(\\n(.wu+\w'\(de'u-\*(#H)/2u'\v'-.3n'\*(#[\z\(de\v'.3n'\h'|\\n:u'\*(#]
115 .ds d- \h'\*(#H'\(pd\h'-\w'~'u'\v'-.25m'\f2\(hy\fP\v'.25m'\h'-\*(#H'
116 .ds D- D\\k:\h'-\w'D'u'\v'-.11m'\z\(hy\v'.11m'\h'|\\n:u'
117 .ds th \*(#[\v'.3m'\s+1I\s-1\v'-.3m'\h'-(\w'I'u*2/3)'\s-1o\s+1\*(#]
118 .ds Th \*(#[\s+2I\s-2\h'-\w'I'u*3/5'\v'-.3m'o\v'.3m'\*(#]
119 .ds ae a\h'-(\w'a'u*4/10)'e
120 .ds Ae A\h'-(\w'A'u*4/10)'E
121 . \" corrections for vroff
122 .if v .ds ~ \\k:\h'-(\\n(.wu*9/10-\*(#H)'\s-2\u~\d\s+2\h'|\\n:u'
123 .if v .ds ^ \\k:\h'-(\\n(.wu*10/11-\*(#H)'\v'-.4m'^\v'.4m'\h'|\\n:u'
124 . \" for low resolution devices (crt and lpr)
125 .if \n(.H>23 .if \n(.V>19 \
138 .\" ======================================================================
141 .TH GCOV 1 "gcc-3.1" "2001-06-25" "GNU"
144 gcov \- coverage testing tool
146 .IX Header "SYNOPSIS"
147 gcov [\fB\-b\fR] [\fB\-c\fR] [\fB\-v\fR] [\fB\-n\fR] [\fB\-l\fR] [\fB\-f\fR] [\fB\-o\fR \fIdirectory\fR] \fIsourcefile\fR
149 .IX Header "DESCRIPTION"
150 \&\fBgcov\fR is a test coverage program. Use it in concert with \s-1GCC\s0
151 to analyze your programs to help create more efficient, faster
152 running code. You can use \fBgcov\fR as a profiling tool to help
153 discover where your optimization efforts will best affect your code. You
154 can also use \fBgcov\fR along with the other profiling tool,
155 \&\fBgprof\fR, to assess which parts of your code use the greatest amount
158 Profiling tools help you analyze your code's performance. Using a
159 profiler such as \fBgcov\fR or \fBgprof\fR, you can find out some
160 basic performance statistics, such as:
162 how often each line of code executes
164 what lines of code are actually executed
166 how much computing time each section of code uses
168 Once you know these things about how your code works when compiled, you
169 can look at each module to see which modules should be optimized.
170 \&\fBgcov\fR helps you determine where to work on optimization.
172 Software developers also use coverage testing in concert with
173 testsuites, to make sure software is actually good enough for a release.
174 Testsuites can verify that a program works as expected; a coverage
175 program tests to see how much of the program is exercised by the
176 testsuite. Developers can then determine what kinds of test cases need
177 to be added to the testsuites to create both better testing and a better
180 You should compile your code without optimization if you plan to use
181 \&\fBgcov\fR because the optimization, by combining some lines of code
182 into one function, may not give you as much information as you need to
183 look for `hot spots' where the code is using a great deal of computer
184 time. Likewise, because \fBgcov\fR accumulates statistics by line (at
185 the lowest resolution), it works best with a programming style that
186 places only one statement on each line. If you use complicated macros
187 that expand to loops or to other control structures, the statistics are
188 less helpful\-\-\-they only report on the line where the macro call
189 appears. If your complex macros behave like functions, you can replace
190 them with inline functions to solve this problem.
192 \&\fBgcov\fR creates a logfile called \fI\fIsourcefile\fI.gcov\fR which
193 indicates how many times each line of a source file \fI\fIsourcefile\fI.c\fR
194 has executed. You can use these logfiles along with \fBgprof\fR to aid
195 in fine-tuning the performance of your programs. \fBgprof\fR gives
196 timing information you can use along with the information you get from
199 \&\fBgcov\fR works only on code compiled with \s-1GCC\s0. It is not
200 compatible with any other profiling or test coverage mechanism.
205 Write branch frequencies to the output file, and write branch summary
206 info to the standard output. This option allows you to see how often
207 each branch in your program was taken.
210 Write branch frequencies as the number of branches taken, rather than
211 the percentage of branches taken.
214 Display the \fBgcov\fR version number (on the standard error stream).
217 Do not create the \fBgcov\fR output file.
220 Create long file names for included source files. For example, if the
221 header file \fIx.h\fR contains code, and was included in the file
222 \&\fIa.c\fR, then running \fBgcov\fR on the file \fIa.c\fR will produce
223 an output file called \fIa.c.x.h.gcov\fR instead of \fIx.h.gcov\fR.
224 This can be useful if \fIx.h\fR is included in multiple source files.
227 Output summaries for each function in addition to the file level summary.
230 The directory where the object files live. Gcov will search for \fI.bb\fR,
231 \&\fI.bbg\fR, and \fI.da\fR files in this directory.
233 When using \fBgcov\fR, you must first compile your program with two
234 special \s-1GCC\s0 options: \fB\-fprofile-arcs \-ftest-coverage\fR.
235 This tells the compiler to generate additional information needed by
236 gcov (basically a flow graph of the program) and also includes
237 additional code in the object files for generating the extra profiling
238 information needed by gcov. These additional files are placed in the
239 directory where the source code is located.
241 Running the program will cause profile output to be generated. For each
242 source file compiled with \fB\-fprofile-arcs\fR, an accompanying \fI.da\fR
243 file will be placed in the source directory.
245 Running \fBgcov\fR with your program's source file names as arguments
246 will now produce a listing of the code along with frequency of execution
247 for each line. For example, if your program is called \fItmp.c\fR, this
248 is what you see when you use the basic \fBgcov\fR facility:
251 \& $ gcc -fprofile-arcs -ftest-coverage tmp.c
254 \& 87.50% of 8 source lines executed in file tmp.c
255 \& Creating tmp.c.gcov.
257 The file \fItmp.c.gcov\fR contains output from \fBgcov\fR.
269 \& 11 for (i = 0; i < 10; i++)
273 \& 1 if (total != 45)
274 \& ###### printf ("Failure\en");
276 \& 1 printf ("Success\en");
279 When you use the \fB\-b\fR option, your output looks like this:
283 \& 87.50% of 8 source lines executed in file tmp.c
284 \& 80.00% of 5 branches executed in file tmp.c
285 \& 80.00% of 5 branches taken at least once in file tmp.c
286 \& 50.00% of 2 calls executed in file tmp.c
287 \& Creating tmp.c.gcov.
289 Here is a sample of a resulting \fItmp.c.gcov\fR file:
300 \& 11 for (i = 0; i < 10; i++)
301 \& branch 0 taken = 91%
302 \& branch 1 taken = 100%
303 \& branch 2 taken = 100%
307 \& 1 if (total != 45)
308 \& branch 0 taken = 100%
309 \& ###### printf ("Failure\en");
310 \& call 0 never executed
311 \& branch 1 never executed
313 \& 1 printf ("Success\en");
314 \& call 0 returns = 100%
317 For each basic block, a line is printed after the last line of the basic
318 block describing the branch or call that ends the basic block. There can
319 be multiple branches and calls listed for a single source line if there
320 are multiple basic blocks that end on that line. In this case, the
321 branches and calls are each given a number. There is no simple way to map
322 these branches and calls back to source constructs. In general, though,
323 the lowest numbered branch or call will correspond to the leftmost construct
326 For a branch, if it was executed at least once, then a percentage
327 indicating the number of times the branch was taken divided by the
328 number of times the branch was executed will be printed. Otherwise, the
329 message ``never executed'' is printed.
331 For a call, if it was executed at least once, then a percentage
332 indicating the number of times the call returned divided by the number
333 of times the call was executed will be printed. This will usually be
334 100%, but may be less for functions call \f(CW\*(C`exit\*(C'\fR or \f(CW\*(C`longjmp\*(C'\fR,
335 and thus may not return every time they are called.
337 The execution counts are cumulative. If the example program were
338 executed again without removing the \fI.da\fR file, the count for the
339 number of times each line in the source was executed would be added to
340 the results of the previous run(s). This is potentially useful in
341 several ways. For example, it could be used to accumulate data over a
342 number of program runs as part of a test verification suite, or to
343 provide more accurate long-term information over a large number of
346 The data in the \fI.da\fR files is saved immediately before the program
347 exits. For each source file compiled with \fB\-fprofile-arcs\fR, the profiling
348 code first attempts to read in an existing \fI.da\fR file; if the file
349 doesn't match the executable (differing number of basic block counts) it
350 will ignore the contents of the file. It then adds in the new execution
351 counts and finally writes the data to the file.
352 .Sh "Using \fBgcov\fP with \s-1GCC\s0 Optimization"
353 .IX Subsection "Using gcov with GCC Optimization"
354 If you plan to use \fBgcov\fR to help optimize your code, you must
355 first compile your program with two special \s-1GCC\s0 options:
356 \&\fB\-fprofile-arcs \-ftest-coverage\fR. Aside from that, you can use any
357 other \s-1GCC\s0 options; but if you want to prove that every single line
358 in your program was executed, you should not compile with optimization
359 at the same time. On some machines the optimizer can eliminate some
360 simple code lines by combining them with other lines. For example, code
369 can be compiled into one instruction on some machines. In this case,
370 there is no way for \fBgcov\fR to calculate separate execution counts
371 for each line because there isn't separate code for each line. Hence
372 the \fBgcov\fR output looks like this if you compiled the program with
381 The output shows that this block of code, combined by optimization,
382 executed 100 times. In one sense this result is correct, because there
383 was only one instruction representing all four of these lines. However,
384 the output does not indicate how many times the result was 0 and how
385 many times the result was 1.
387 .IX Header "SEE ALSO"
388 \&\fIgcc\fR\|(1) and the Info entry for \fIgcc\fR.
390 .IX Header "COPYRIGHT"
391 Copyright (c) 1996, 1997, 1999, 2000 Free Software Foundation, Inc.
393 Permission is granted to make and distribute verbatim copies of this
394 manual provided the copyright notice and this permission notice are
395 preserved on all copies.
397 Permission is granted to copy and distribute modified versions of this
398 manual under the conditions for verbatim copying, provided also that the
399 entire resulting derived work is distributed under the terms of a
400 permission notice identical to this one.
402 Permission is granted to copy and distribute translations of this manual
403 into another language, under the above conditions for modified versions,
404 except that this permission notice may be included in translations
405 approved by the Free Software Foundation instead of in the original
OSDN Git Service
