This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
Must store pointers with the right width.
[perl5.git] / pod / perlhack.pod
CommitLineData
e8cd7eae
GS
1=head1 NAME
2
3perlhack - How to hack at the Perl internals
4
5=head1 DESCRIPTION
6
7This document attempts to explain how Perl development takes place,
8and ends with some suggestions for people wanting to become bona fide
9porters.
10
11The perl5-porters mailing list is where the Perl standard distribution
12is maintained and developed. The list can get anywhere from 10 to 150
13messages a day, depending on the heatedness of the debate. Most days
14there are two or three patches, extensions, features, or bugs being
15discussed at a time.
16
f8e3975a 17A searchable archive of the list is at either:
e8cd7eae
GS
18
19 http://www.xray.mpe.mpg.de/mailing-lists/perl5-porters/
20
f8e3975a
IP
21or
22
23 http://archive.develooper.com/perl5-porters@perl.org/
24
e8cd7eae
GS
25List subscribers (the porters themselves) come in several flavours.
26Some are quiet curious lurkers, who rarely pitch in and instead watch
27the ongoing development to ensure they're forewarned of new changes or
28features in Perl. Some are representatives of vendors, who are there
29to make sure that Perl continues to compile and work on their
30platforms. Some patch any reported bug that they know how to fix,
31some are actively patching their pet area (threads, Win32, the regexp
32engine), while others seem to do nothing but complain. In other
33words, it's your usual mix of technical people.
34
35Over this group of porters presides Larry Wall. He has the final word
f6c51b38
GS
36in what does and does not change in the Perl language. Various
37releases of Perl are shepherded by a ``pumpking'', a porter
38responsible for gathering patches, deciding on a patch-by-patch
39feature-by-feature basis what will and will not go into the release.
40For instance, Gurusamy Sarathy is the pumpking for the 5.6 release of
41Perl.
e8cd7eae
GS
42
43In addition, various people are pumpkings for different things. For
44instance, Andy Dougherty and Jarkko Hietaniemi share the I<Configure>
45pumpkin, and Tom Christiansen is the documentation pumpking.
46
47Larry sees Perl development along the lines of the US government:
48there's the Legislature (the porters), the Executive branch (the
49pumpkings), and the Supreme Court (Larry). The legislature can
50discuss and submit patches to the executive branch all they like, but
51the executive branch is free to veto them. Rarely, the Supreme Court
52will side with the executive branch over the legislature, or the
53legislature over the executive branch. Mostly, however, the
54legislature and the executive branch are supposed to get along and
55work out their differences without impeachment or court cases.
56
57You might sometimes see reference to Rule 1 and Rule 2. Larry's power
58as Supreme Court is expressed in The Rules:
59
60=over 4
61
62=item 1
63
64Larry is always by definition right about how Perl should behave.
65This means he has final veto power on the core functionality.
66
67=item 2
68
69Larry is allowed to change his mind about any matter at a later date,
70regardless of whether he previously invoked Rule 1.
71
72=back
73
74Got that? Larry is always right, even when he was wrong. It's rare
75to see either Rule exercised, but they are often alluded to.
76
77New features and extensions to the language are contentious, because
78the criteria used by the pumpkings, Larry, and other porters to decide
79which features should be implemented and incorporated are not codified
80in a few small design goals as with some other languages. Instead,
81the heuristics are flexible and often difficult to fathom. Here is
82one person's list, roughly in decreasing order of importance, of
83heuristics that new features have to be weighed against:
84
85=over 4
86
87=item Does concept match the general goals of Perl?
88
89These haven't been written anywhere in stone, but one approximation
90is:
91
92 1. Keep it fast, simple, and useful.
93 2. Keep features/concepts as orthogonal as possible.
94 3. No arbitrary limits (platforms, data sizes, cultures).
95 4. Keep it open and exciting to use/patch/advocate Perl everywhere.
96 5. Either assimilate new technologies, or build bridges to them.
97
98=item Where is the implementation?
99
100All the talk in the world is useless without an implementation. In
101almost every case, the person or people who argue for a new feature
102will be expected to be the ones who implement it. Porters capable
103of coding new features have their own agendas, and are not available
104to implement your (possibly good) idea.
105
106=item Backwards compatibility
107
108It's a cardinal sin to break existing Perl programs. New warnings are
109contentious--some say that a program that emits warnings is not
110broken, while others say it is. Adding keywords has the potential to
111break programs, changing the meaning of existing token sequences or
112functions might break programs.
113
114=item Could it be a module instead?
115
116Perl 5 has extension mechanisms, modules and XS, specifically to avoid
117the need to keep changing the Perl interpreter. You can write modules
118that export functions, you can give those functions prototypes so they
119can be called like built-in functions, you can even write XS code to
120mess with the runtime data structures of the Perl interpreter if you
121want to implement really complicated things. If it can be done in a
122module instead of in the core, it's highly unlikely to be added.
123
124=item Is the feature generic enough?
125
126Is this something that only the submitter wants added to the language,
127or would it be broadly useful? Sometimes, instead of adding a feature
128with a tight focus, the porters might decide to wait until someone
129implements the more generalized feature. For instance, instead of
130implementing a ``delayed evaluation'' feature, the porters are waiting
131for a macro system that would permit delayed evaluation and much more.
132
133=item Does it potentially introduce new bugs?
134
135Radical rewrites of large chunks of the Perl interpreter have the
136potential to introduce new bugs. The smaller and more localized the
137change, the better.
138
139=item Does it preclude other desirable features?
140
141A patch is likely to be rejected if it closes off future avenues of
142development. For instance, a patch that placed a true and final
143interpretation on prototypes is likely to be rejected because there
144are still options for the future of prototypes that haven't been
145addressed.
146
147=item Is the implementation robust?
148
149Good patches (tight code, complete, correct) stand more chance of
150going in. Sloppy or incorrect patches might be placed on the back
151burner until the pumpking has time to fix, or might be discarded
152altogether without further notice.
153
154=item Is the implementation generic enough to be portable?
155
156The worst patches make use of a system-specific features. It's highly
157unlikely that nonportable additions to the Perl language will be
158accepted.
159
160=item Is there enough documentation?
161
162Patches without documentation are probably ill-thought out or
163incomplete. Nothing can be added without documentation, so submitting
164a patch for the appropriate manpages as well as the source code is
165always a good idea. If appropriate, patches should add to the test
166suite as well.
167
168=item Is there another way to do it?
169
170Larry said ``Although the Perl Slogan is I<There's More Than One Way
171to Do It>, I hesitate to make 10 ways to do something''. This is a
172tricky heuristic to navigate, though--one man's essential addition is
173another man's pointless cruft.
174
175=item Does it create too much work?
176
177Work for the pumpking, work for Perl programmers, work for module
178authors, ... Perl is supposed to be easy.
179
f6c51b38
GS
180=item Patches speak louder than words
181
182Working code is always preferred to pie-in-the-sky ideas. A patch to
183add a feature stands a much higher chance of making it to the language
184than does a random feature request, no matter how fervently argued the
185request might be. This ties into ``Will it be useful?'', as the fact
186that someone took the time to make the patch demonstrates a strong
187desire for the feature.
188
e8cd7eae
GS
189=back
190
191If you're on the list, you might hear the word ``core'' bandied
192around. It refers to the standard distribution. ``Hacking on the
193core'' means you're changing the C source code to the Perl
194interpreter. ``A core module'' is one that ships with Perl.
195
a1f349fd
MB
196=head2 Keeping in sync
197
e8cd7eae
GS
198The source code to the Perl interpreter, in its different versions, is
199kept in a repository managed by a revision control system (which is
200currently the Perforce program, see http://perforce.com/). The
201pumpkings and a few others have access to the repository to check in
202changes. Periodically the pumpking for the development version of Perl
203will release a new version, so the rest of the porters can see what's
2be4c08b
GS
204changed. The current state of the main trunk of repository, and patches
205that describe the individual changes that have happened since the last
206public release are available at this location:
207
208 ftp://ftp.linux.activestate.com/pub/staff/gsar/APC/
209
a1f349fd
MB
210If you are a member of the perl5-porters mailing list, it is a good
211thing to keep in touch with the most recent changes. If not only to
212verify if what you would have posted as a bug report isn't already
213solved in the most recent available perl development branch, also
214known as perl-current, bleading edge perl, bleedperl or bleadperl.
2be4c08b
GS
215
216Needless to say, the source code in perl-current is usually in a perpetual
217state of evolution. You should expect it to be very buggy. Do B<not> use
218it for any purpose other than testing and development.
e8cd7eae 219
3e148164
JH
220Keeping in sync with the most recent branch can be done in several ways,
221but the most convenient and reliable way is using B<rsync>, available at
222ftp://rsync.samba.org/pub/rsync/ . (You can also get the most recent
223branch by FTP.)
a1f349fd
MB
224
225If you choose to keep in sync using rsync, there are two approaches
3e148164 226to doing so:
a1f349fd
MB
227
228=over 4
229
230=item rsync'ing the source tree
231
3e148164 232Presuming you are in the directory where your perl source resides
a1f349fd
MB
233and you have rsync installed and available, you can `upgrade' to
234the bleadperl using:
235
236 # rsync -avz rsync://ftp.linux.activestate.com/perl-current/ .
237
238This takes care of updating every single item in the source tree to
239the latest applied patch level, creating files that are new (to your
240distribution) and setting date/time stamps of existing files to
241reflect the bleadperl status.
242
c6d0653e
LC
243Note that this will not delete any files that were in '.' before
244the rsync. Once you are sure that the rsync is running correctly,
245run it with the --delete and the --dry-run options like this:
246
247 # rsync -avz --delete --dry-run rsync://ftp.linux.activestate.com/perl-current/ .
248
249This will I<simulate> an rsync run that also deletes files not
250present in the bleadperl master copy. Observe the results from
251this run closely. If you are sure that the actual run would delete
252no files precious to you, you could remove the '--dry-run' option.
253
a1f349fd
MB
254You can than check what patch was the latest that was applied by
255looking in the file B<.patch>, which will show the number of the
256latest patch.
257
258If you have more than one machine to keep in sync, and not all of
259them have access to the WAN (so you are not able to rsync all the
260source trees to the real source), there are some ways to get around
261this problem.
262
263=over 4
264
265=item Using rsync over the LAN
266
267Set up a local rsync server which makes the rsynced source tree
3e148164 268available to the LAN and sync the other machines against this
a1f349fd
MB
269directory.
270
271From http://rsync.samba.org/README.html:
272
273 "Rsync uses rsh or ssh for communication. It does not need to be
274 setuid and requires no special privileges for installation. It
3958b146 275 does not require an inetd entry or a daemon. You must, however,
a1f349fd
MB
276 have a working rsh or ssh system. Using ssh is recommended for
277 its security features."
278
279=item Using pushing over the NFS
280
281Having the other systems mounted over the NFS, you can take an
3e148164
JH
282active pushing approach by checking the just updated tree against
283the other not-yet synced trees. An example would be
284
285 #!/usr/bin/perl -w
286
287 use strict;
288 use File::Copy;
289
290 my %MF = map {
291 m/(\S+)/;
292 $1 => [ (stat $1)[2, 7, 9] ]; # mode, size, mtime
293 } `cat MANIFEST`;
294
295 my %remote = map { $_ => "/$_/pro/3gl/CPAN/perl-5.7.1" } qw(host1 host2);
296
297 foreach my $host (keys %remote) {
298 unless (-d $remote{$host}) {
299 print STDERR "Cannot Xsync for host $host\n";
300 next;
301 }
302 foreach my $file (keys %MF) {
303 my $rfile = "$remote{$host}/$file";
304 my ($mode, $size, $mtime) = (stat $rfile)[2, 7, 9];
305 defined $size or ($mode, $size, $mtime) = (0, 0, 0);
306 $size == $MF{$file}[1] && $mtime == $MF{$file}[2] and next;
307 printf "%4s %-34s %8d %9d %8d %9d\n",
308 $host, $file, $MF{$file}[1], $MF{$file}[2], $size, $mtime;
309 unlink $rfile;
310 copy ($file, $rfile);
311 utime time, $MF{$file}[2], $rfile;
312 chmod $MF{$file}[0], $rfile;
313 }
314 }
315
316though this is not perfect. It could be improved with checking
a1f349fd
MB
317file checksums before updating. Not all NFS systems support
318reliable utime support (when used over the NFS).
319
320=back
321
322=item rsync'ing the patches
323
324The source tree is maintained by the pumpking who applies patches to
325the files in the tree. These patches are either created by the
326pumpking himself using C<diff -c> after updating the file manually or
327by applying patches sent in by posters on the perl5-porters list.
328These patches are also saved and rsync'able, so you can apply them
329yourself to the source files.
330
331Presuming you are in a directory where your patches reside, you can
3e148164 332get them in sync with
a1f349fd
MB
333
334 # rsync -avz rsync://ftp.linux.activestate.com/perl-current-diffs/ .
335
336This makes sure the latest available patch is downloaded to your
337patch directory.
338
3e148164 339It's then up to you to apply these patches, using something like
a1f349fd
MB
340
341 # last=`ls -rt1 *.gz | tail -1`
342 # rsync -avz rsync://ftp.linux.activestate.com/perl-current-diffs/ .
343 # find . -name '*.gz' -newer $last -exec gzcat {} \; >blead.patch
344 # cd ../perl-current
345 # patch -p1 -N <../perl-current-diffs/blead.patch
346
347or, since this is only a hint towards how it works, use CPAN-patchaperl
348from Andreas K├Ânig to have better control over the patching process.
349
350=back
351
f7e1e956 352=head2 Why rsync the source tree
a1f349fd
MB
353
354=over 4
355
10f58044 356=item It's easier to rsync the source tree
a1f349fd
MB
357
358Since you don't have to apply the patches yourself, you are sure all
359files in the source tree are in the right state.
360
361=item It's more recent
362
363According to Gurusamy Sarathy:
364
365 "... The rsync mirror is automatic and syncs with the repository
366 every five minutes.
367
3e148164 368 "Updating the patch area still requires manual intervention
a1f349fd
MB
369 (with all the goofiness that implies, which you've noted) and
370 is typically on a daily cycle. Making this process automatic
371 is on my tuit list, but don't ask me when."
372
373=item It's more reliable
374
3e148164 375Well, since the patches are updated by hand, I don't have to say any
a1f349fd
MB
376more ... (see Sarathy's remark).
377
378=back
379
f7e1e956 380=head2 Why rsync the patches
a1f349fd
MB
381
382=over 4
383
10f58044 384=item It's easier to rsync the patches
a1f349fd
MB
385
386If you have more than one machine that you want to keep in track with
3e148164 387bleadperl, it's easier to rsync the patches only once and then apply
a1f349fd
MB
388them to all the source trees on the different machines.
389
390In case you try to keep in pace on 5 different machines, for which
391only one of them has access to the WAN, rsync'ing all the source
3e148164 392trees should than be done 5 times over the NFS. Having
a1f349fd 393rsync'ed the patches only once, I can apply them to all the source
3e148164 394trees automatically. Need you say more ;-)
a1f349fd
MB
395
396=item It's a good reference
397
398If you do not only like to have the most recent development branch,
399but also like to B<fix> bugs, or extend features, you want to dive
400into the sources. If you are a seasoned perl core diver, you don't
401need no manuals, tips, roadmaps, perlguts.pod or other aids to find
402your way around. But if you are a starter, the patches may help you
403in finding where you should start and how to change the bits that
404bug you.
405
406The file B<Changes> is updated on occasions the pumpking sees as his
407own little sync points. On those occasions, he releases a tar-ball of
408the current source tree (i.e. perl@7582.tar.gz), which will be an
409excellent point to start with when choosing to use the 'rsync the
410patches' scheme. Starting with perl@7582, which means a set of source
411files on which the latest applied patch is number 7582, you apply all
f18956b7 412succeeding patches available from then on (7583, 7584, ...).
a1f349fd
MB
413
414You can use the patches later as a kind of search archive.
415
416=over 4
417
418=item Finding a start point
419
420If you want to fix/change the behaviour of function/feature Foo, just
421scan the patches for patches that mention Foo either in the subject,
3e148164 422the comments, or the body of the fix. A good chance the patch shows
a1f349fd
MB
423you the files that are affected by that patch which are very likely
424to be the starting point of your journey into the guts of perl.
425
426=item Finding how to fix a bug
427
428If you've found I<where> the function/feature Foo misbehaves, but you
429don't know how to fix it (but you do know the change you want to
430make), you can, again, peruse the patches for similar changes and
431look how others apply the fix.
432
433=item Finding the source of misbehaviour
434
435When you keep in sync with bleadperl, the pumpking would love to
3958b146 436I<see> that the community efforts really work. So after each of his
a1f349fd
MB
437sync points, you are to 'make test' to check if everything is still
438in working order. If it is, you do 'make ok', which will send an OK
439report to perlbug@perl.org. (If you do not have access to a mailer
3e148164 440from the system you just finished successfully 'make test', you can
a1f349fd
MB
441do 'make okfile', which creates the file C<perl.ok>, which you can
442than take to your favourite mailer and mail yourself).
443
3958b146 444But of course, as always, things will not always lead to a success
a1f349fd
MB
445path, and one or more test do not pass the 'make test'. Before
446sending in a bug report (using 'make nok' or 'make nokfile'), check
447the mailing list if someone else has reported the bug already and if
448so, confirm it by replying to that message. If not, you might want to
449trace the source of that misbehaviour B<before> sending in the bug,
450which will help all the other porters in finding the solution.
451
3e148164
JH
452Here the saved patches come in very handy. You can check the list of
453patches to see which patch changed what file and what change caused
454the misbehaviour. If you note that in the bug report, it saves the
455one trying to solve it, looking for that point.
a1f349fd
MB
456
457=back
458
459If searching the patches is too bothersome, you might consider using
460perl's bugtron to find more information about discussions and
461ramblings on posted bugs.
462
463=back
464
3e148164
JH
465If you want to get the best of both worlds, rsync both the source
466tree for convenience, reliability and ease and rsync the patches
467for reference.
468
a1f349fd
MB
469=head2 Submitting patches
470
f7e1e956
MS
471Always submit patches to I<perl5-porters@perl.org>. If you're
472patching a core module and there's an author listed, send the author a
473copy (see L<Patching a core module>). This lets other porters review
474your patch, which catches a surprising number of errors in patches.
475Either use the diff program (available in source code form from
476I<ftp://ftp.gnu.org/pub/gnu/>), or use Johan Vromans' I<makepatch>
477(available from I<CPAN/authors/id/JV/>). Unified diffs are preferred,
478but context diffs are accepted. Do not send RCS-style diffs or diffs
479without context lines. More information is given in the
480I<Porting/patching.pod> file in the Perl source distribution. Please
481patch against the latest B<development> version (e.g., if you're
482fixing a bug in the 5.005 track, patch against the latest 5.005_5x
483version). Only patches that survive the heat of the development
484branch get applied to maintenance versions.
485
486Your patch should update the documentation and test suite. See
487L<Writing a test>.
e8cd7eae
GS
488
489To report a bug in Perl, use the program I<perlbug> which comes with
490Perl (if you can't get Perl to work, send mail to the address
f18956b7 491I<perlbug@perl.org> or I<perlbug@perl.com>). Reporting bugs through
e8cd7eae
GS
492I<perlbug> feeds into the automated bug-tracking system, access to
493which is provided through the web at I<http://bugs.perl.org/>. It
494often pays to check the archives of the perl5-porters mailing list to
495see whether the bug you're reporting has been reported before, and if
496so whether it was considered a bug. See above for the location of
497the searchable archives.
498
499The CPAN testers (I<http://testers.cpan.org/>) are a group of
500volunteers who test CPAN modules on a variety of platforms. Perl Labs
f6c51b38
GS
501(I<http://labs.perl.org/>) automatically tests Perl source releases on
502platforms and gives feedback to the CPAN testers mailing list. Both
503efforts welcome volunteers.
e8cd7eae 504
e8cd7eae
GS
505It's a good idea to read and lurk for a while before chipping in.
506That way you'll get to see the dynamic of the conversations, learn the
507personalities of the players, and hopefully be better prepared to make
508a useful contribution when do you speak up.
509
510If after all this you still think you want to join the perl5-porters
f6c51b38
GS
511mailing list, send mail to I<perl5-porters-subscribe@perl.org>. To
512unsubscribe, send mail to I<perl5-porters-unsubscribe@perl.org>.
e8cd7eae 513
a422fd2d
SC
514To hack on the Perl guts, you'll need to read the following things:
515
516=over 3
517
518=item L<perlguts>
519
520This is of paramount importance, since it's the documentation of what
521goes where in the Perl source. Read it over a couple of times and it
522might start to make sense - don't worry if it doesn't yet, because the
523best way to study it is to read it in conjunction with poking at Perl
524source, and we'll do that later on.
525
526You might also want to look at Gisle Aas's illustrated perlguts -
527there's no guarantee that this will be absolutely up-to-date with the
528latest documentation in the Perl core, but the fundamentals will be
529right. (http://gisle.aas.no/perl/illguts/)
530
531=item L<perlxstut> and L<perlxs>
532
533A working knowledge of XSUB programming is incredibly useful for core
534hacking; XSUBs use techniques drawn from the PP code, the portion of the
535guts that actually executes a Perl program. It's a lot gentler to learn
536those techniques from simple examples and explanation than from the core
537itself.
538
539=item L<perlapi>
540
541The documentation for the Perl API explains what some of the internal
542functions do, as well as the many macros used in the source.
543
544=item F<Porting/pumpkin.pod>
545
546This is a collection of words of wisdom for a Perl porter; some of it is
547only useful to the pumpkin holder, but most of it applies to anyone
548wanting to go about Perl development.
549
550=item The perl5-porters FAQ
551
552This is posted to perl5-porters at the beginning on every month, and
553should be available from http://perlhacker.org/p5p-faq; alternatively,
554you can get the FAQ emailed to you by sending mail to
555C<perl5-porters-faq@perl.org>. It contains hints on reading
556perl5-porters, information on how perl5-porters works and how Perl
557development in general works.
558
559=back
560
561=head2 Finding Your Way Around
562
563Perl maintenance can be split into a number of areas, and certain people
564(pumpkins) will have responsibility for each area. These areas sometimes
565correspond to files or directories in the source kit. Among the areas are:
566
567=over 3
568
569=item Core modules
570
571Modules shipped as part of the Perl core live in the F<lib/> and F<ext/>
572subdirectories: F<lib/> is for the pure-Perl modules, and F<ext/>
573contains the core XS modules.
574
f7e1e956
MS
575=item Tests
576
577There are tests for nearly all the modules, built-ins and major bits
578of functionality. Test files all have a .t suffix. Module tests live
579in the F<lib/> and F<ext/> directories next to the module being
580tested. Others live in F<t/>. See L<Writing a test>
581
a422fd2d
SC
582=item Documentation
583
584Documentation maintenance includes looking after everything in the
585F<pod/> directory, (as well as contributing new documentation) and
586the documentation to the modules in core.
587
588=item Configure
589
590The configure process is the way we make Perl portable across the
591myriad of operating systems it supports. Responsibility for the
592configure, build and installation process, as well as the overall
593portability of the core code rests with the configure pumpkin - others
594help out with individual operating systems.
595
596The files involved are the operating system directories, (F<win32/>,
597F<os2/>, F<vms/> and so on) the shell scripts which generate F<config.h>
598and F<Makefile>, as well as the metaconfig files which generate
599F<Configure>. (metaconfig isn't included in the core distribution.)
600
601=item Interpreter
602
603And of course, there's the core of the Perl interpreter itself. Let's
604have a look at that in a little more detail.
605
606=back
607
608Before we leave looking at the layout, though, don't forget that
609F<MANIFEST> contains not only the file names in the Perl distribution,
610but short descriptions of what's in them, too. For an overview of the
611important files, try this:
612
613 perl -lne 'print if /^[^\/]+\.[ch]\s+/' MANIFEST
614
615=head2 Elements of the interpreter
616
617The work of the interpreter has two main stages: compiling the code
618into the internal representation, or bytecode, and then executing it.
619L<perlguts/Compiled code> explains exactly how the compilation stage
620happens.
621
622Here is a short breakdown of perl's operation:
623
624=over 3
625
626=item Startup
627
628The action begins in F<perlmain.c>. (or F<miniperlmain.c> for miniperl)
629This is very high-level code, enough to fit on a single screen, and it
630resembles the code found in L<perlembed>; most of the real action takes
631place in F<perl.c>
632
633First, F<perlmain.c> allocates some memory and constructs a Perl
634interpreter:
635
636 1 PERL_SYS_INIT3(&argc,&argv,&env);
637 2
638 3 if (!PL_do_undump) {
639 4 my_perl = perl_alloc();
640 5 if (!my_perl)
641 6 exit(1);
642 7 perl_construct(my_perl);
643 8 PL_perl_destruct_level = 0;
644 9 }
645
646Line 1 is a macro, and its definition is dependent on your operating
647system. Line 3 references C<PL_do_undump>, a global variable - all
648global variables in Perl start with C<PL_>. This tells you whether the
649current running program was created with the C<-u> flag to perl and then
650F<undump>, which means it's going to be false in any sane context.
651
652Line 4 calls a function in F<perl.c> to allocate memory for a Perl
653interpreter. It's quite a simple function, and the guts of it looks like
654this:
655
656 my_perl = (PerlInterpreter*)PerlMem_malloc(sizeof(PerlInterpreter));
657
658Here you see an example of Perl's system abstraction, which we'll see
659later: C<PerlMem_malloc> is either your system's C<malloc>, or Perl's
660own C<malloc> as defined in F<malloc.c> if you selected that option at
661configure time.
662
663Next, in line 7, we construct the interpreter; this sets up all the
664special variables that Perl needs, the stacks, and so on.
665
666Now we pass Perl the command line options, and tell it to go:
667
668 exitstatus = perl_parse(my_perl, xs_init, argc, argv, (char **)NULL);
669 if (!exitstatus) {
670 exitstatus = perl_run(my_perl);
671 }
672
673
674C<perl_parse> is actually a wrapper around C<S_parse_body>, as defined
675in F<perl.c>, which processes the command line options, sets up any
676statically linked XS modules, opens the program and calls C<yyparse> to
677parse it.
678
679=item Parsing
680
681The aim of this stage is to take the Perl source, and turn it into an op
682tree. We'll see what one of those looks like later. Strictly speaking,
683there's three things going on here.
684
685C<yyparse>, the parser, lives in F<perly.c>, although you're better off
686reading the original YACC input in F<perly.y>. (Yes, Virginia, there
687B<is> a YACC grammar for Perl!) The job of the parser is to take your
688code and `understand' it, splitting it into sentences, deciding which
689operands go with which operators and so on.
690
691The parser is nobly assisted by the lexer, which chunks up your input
692into tokens, and decides what type of thing each token is: a variable
693name, an operator, a bareword, a subroutine, a core function, and so on.
694The main point of entry to the lexer is C<yylex>, and that and its
695associated routines can be found in F<toke.c>. Perl isn't much like
696other computer languages; it's highly context sensitive at times, it can
697be tricky to work out what sort of token something is, or where a token
698ends. As such, there's a lot of interplay between the tokeniser and the
699parser, which can get pretty frightening if you're not used to it.
700
701As the parser understands a Perl program, it builds up a tree of
702operations for the interpreter to perform during execution. The routines
703which construct and link together the various operations are to be found
704in F<op.c>, and will be examined later.
705
706=item Optimization
707
708Now the parsing stage is complete, and the finished tree represents
709the operations that the Perl interpreter needs to perform to execute our
710program. Next, Perl does a dry run over the tree looking for
711optimisations: constant expressions such as C<3 + 4> will be computed
712now, and the optimizer will also see if any multiple operations can be
713replaced with a single one. For instance, to fetch the variable C<$foo>,
714instead of grabbing the glob C<*foo> and looking at the scalar
715component, the optimizer fiddles the op tree to use a function which
716directly looks up the scalar in question. The main optimizer is C<peep>
717in F<op.c>, and many ops have their own optimizing functions.
718
719=item Running
720
721Now we're finally ready to go: we have compiled Perl byte code, and all
722that's left to do is run it. The actual execution is done by the
723C<runops_standard> function in F<run.c>; more specifically, it's done by
724these three innocent looking lines:
725
726 while ((PL_op = CALL_FPTR(PL_op->op_ppaddr)(aTHX))) {
727 PERL_ASYNC_CHECK();
728 }
729
730You may be more comfortable with the Perl version of that:
731
732 PERL_ASYNC_CHECK() while $Perl::op = &{$Perl::op->{function}};
733
734Well, maybe not. Anyway, each op contains a function pointer, which
735stipulates the function which will actually carry out the operation.
736This function will return the next op in the sequence - this allows for
737things like C<if> which choose the next op dynamically at run time.
738The C<PERL_ASYNC_CHECK> makes sure that things like signals interrupt
739execution if required.
740
741The actual functions called are known as PP code, and they're spread
742between four files: F<pp_hot.c> contains the `hot' code, which is most
743often used and highly optimized, F<pp_sys.c> contains all the
744system-specific functions, F<pp_ctl.c> contains the functions which
745implement control structures (C<if>, C<while> and the like) and F<pp.c>
746contains everything else. These are, if you like, the C code for Perl's
747built-in functions and operators.
748
749=back
750
751=head2 Internal Variable Types
752
753You should by now have had a look at L<perlguts>, which tells you about
754Perl's internal variable types: SVs, HVs, AVs and the rest. If not, do
755that now.
756
757These variables are used not only to represent Perl-space variables, but
758also any constants in the code, as well as some structures completely
759internal to Perl. The symbol table, for instance, is an ordinary Perl
760hash. Your code is represented by an SV as it's read into the parser;
761any program files you call are opened via ordinary Perl filehandles, and
762so on.
763
764The core L<Devel::Peek|Devel::Peek> module lets us examine SVs from a
765Perl program. Let's see, for instance, how Perl treats the constant
766C<"hello">.
767
768 % perl -MDevel::Peek -e 'Dump("hello")'
769 1 SV = PV(0xa041450) at 0xa04ecbc
770 2 REFCNT = 1
771 3 FLAGS = (POK,READONLY,pPOK)
772 4 PV = 0xa0484e0 "hello"\0
773 5 CUR = 5
774 6 LEN = 6
775
776Reading C<Devel::Peek> output takes a bit of practise, so let's go
777through it line by line.
778
779Line 1 tells us we're looking at an SV which lives at C<0xa04ecbc> in
780memory. SVs themselves are very simple structures, but they contain a
781pointer to a more complex structure. In this case, it's a PV, a
782structure which holds a string value, at location C<0xa041450>. Line 2
783is the reference count; there are no other references to this data, so
784it's 1.
785
786Line 3 are the flags for this SV - it's OK to use it as a PV, it's a
787read-only SV (because it's a constant) and the data is a PV internally.
788Next we've got the contents of the string, starting at location
789C<0xa0484e0>.
790
791Line 5 gives us the current length of the string - note that this does
792B<not> include the null terminator. Line 6 is not the length of the
793string, but the length of the currently allocated buffer; as the string
794grows, Perl automatically extends the available storage via a routine
795called C<SvGROW>.
796
797You can get at any of these quantities from C very easily; just add
798C<Sv> to the name of the field shown in the snippet, and you've got a
799macro which will return the value: C<SvCUR(sv)> returns the current
800length of the string, C<SvREFCOUNT(sv)> returns the reference count,
801C<SvPV(sv, len)> returns the string itself with its length, and so on.
802More macros to manipulate these properties can be found in L<perlguts>.
803
804Let's take an example of manipulating a PV, from C<sv_catpvn>, in F<sv.c>
805
806 1 void
807 2 Perl_sv_catpvn(pTHX_ register SV *sv, register const char *ptr, register STRLEN len)
808 3 {
809 4 STRLEN tlen;
810 5 char *junk;
811
812 6 junk = SvPV_force(sv, tlen);
813 7 SvGROW(sv, tlen + len + 1);
814 8 if (ptr == junk)
815 9 ptr = SvPVX(sv);
816 10 Move(ptr,SvPVX(sv)+tlen,len,char);
817 11 SvCUR(sv) += len;
818 12 *SvEND(sv) = '\0';
819 13 (void)SvPOK_only_UTF8(sv); /* validate pointer */
820 14 SvTAINT(sv);
821 15 }
822
823This is a function which adds a string, C<ptr>, of length C<len> onto
824the end of the PV stored in C<sv>. The first thing we do in line 6 is
825make sure that the SV B<has> a valid PV, by calling the C<SvPV_force>
826macro to force a PV. As a side effect, C<tlen> gets set to the current
827value of the PV, and the PV itself is returned to C<junk>.
828
b1866b2d 829In line 7, we make sure that the SV will have enough room to accommodate
a422fd2d
SC
830the old string, the new string and the null terminator. If C<LEN> isn't
831big enough, C<SvGROW> will reallocate space for us.
832
833Now, if C<junk> is the same as the string we're trying to add, we can
834grab the string directly from the SV; C<SvPVX> is the address of the PV
835in the SV.
836
837Line 10 does the actual catenation: the C<Move> macro moves a chunk of
838memory around: we move the string C<ptr> to the end of the PV - that's
839the start of the PV plus its current length. We're moving C<len> bytes
840of type C<char>. After doing so, we need to tell Perl we've extended the
841string, by altering C<CUR> to reflect the new length. C<SvEND> is a
842macro which gives us the end of the string, so that needs to be a
843C<"\0">.
844
845Line 13 manipulates the flags; since we've changed the PV, any IV or NV
846values will no longer be valid: if we have C<$a=10; $a.="6";> we don't
847want to use the old IV of 10. C<SvPOK_only_utf8> is a special UTF8-aware
848version of C<SvPOK_only>, a macro which turns off the IOK and NOK flags
849and turns on POK. The final C<SvTAINT> is a macro which launders tainted
850data if taint mode is turned on.
851
852AVs and HVs are more complicated, but SVs are by far the most common
853variable type being thrown around. Having seen something of how we
854manipulate these, let's go on and look at how the op tree is
855constructed.
856
857=head2 Op Trees
858
859First, what is the op tree, anyway? The op tree is the parsed
860representation of your program, as we saw in our section on parsing, and
861it's the sequence of operations that Perl goes through to execute your
862program, as we saw in L</Running>.
863
864An op is a fundamental operation that Perl can perform: all the built-in
865functions and operators are ops, and there are a series of ops which
866deal with concepts the interpreter needs internally - entering and
867leaving a block, ending a statement, fetching a variable, and so on.
868
869The op tree is connected in two ways: you can imagine that there are two
870"routes" through it, two orders in which you can traverse the tree.
871First, parse order reflects how the parser understood the code, and
872secondly, execution order tells perl what order to perform the
873operations in.
874
875The easiest way to examine the op tree is to stop Perl after it has
876finished parsing, and get it to dump out the tree. This is exactly what
877the compiler backends L<B::Terse|B::Terse> and L<B::Debug|B::Debug> do.
878
879Let's have a look at how Perl sees C<$a = $b + $c>:
880
881 % perl -MO=Terse -e '$a=$b+$c'
882 1 LISTOP (0x8179888) leave
883 2 OP (0x81798b0) enter
884 3 COP (0x8179850) nextstate
885 4 BINOP (0x8179828) sassign
886 5 BINOP (0x8179800) add [1]
887 6 UNOP (0x81796e0) null [15]
888 7 SVOP (0x80fafe0) gvsv GV (0x80fa4cc) *b
889 8 UNOP (0x81797e0) null [15]
890 9 SVOP (0x8179700) gvsv GV (0x80efeb0) *c
891 10 UNOP (0x816b4f0) null [15]
892 11 SVOP (0x816dcf0) gvsv GV (0x80fa460) *a
893
894Let's start in the middle, at line 4. This is a BINOP, a binary
895operator, which is at location C<0x8179828>. The specific operator in
896question is C<sassign> - scalar assignment - and you can find the code
897which implements it in the function C<pp_sassign> in F<pp_hot.c>. As a
898binary operator, it has two children: the add operator, providing the
899result of C<$b+$c>, is uppermost on line 5, and the left hand side is on
900line 10.
901
902Line 10 is the null op: this does exactly nothing. What is that doing
903there? If you see the null op, it's a sign that something has been
904optimized away after parsing. As we mentioned in L</Optimization>,
905the optimization stage sometimes converts two operations into one, for
906example when fetching a scalar variable. When this happens, instead of
907rewriting the op tree and cleaning up the dangling pointers, it's easier
908just to replace the redundant operation with the null op. Originally,
909the tree would have looked like this:
910
911 10 SVOP (0x816b4f0) rv2sv [15]
912 11 SVOP (0x816dcf0) gv GV (0x80fa460) *a
913
914That is, fetch the C<a> entry from the main symbol table, and then look
915at the scalar component of it: C<gvsv> (C<pp_gvsv> into F<pp_hot.c>)
916happens to do both these things.
917
918The right hand side, starting at line 5 is similar to what we've just
919seen: we have the C<add> op (C<pp_add> also in F<pp_hot.c>) add together
920two C<gvsv>s.
921
922Now, what's this about?
923
924 1 LISTOP (0x8179888) leave
925 2 OP (0x81798b0) enter
926 3 COP (0x8179850) nextstate
927
928C<enter> and C<leave> are scoping ops, and their job is to perform any
929housekeeping every time you enter and leave a block: lexical variables
930are tidied up, unreferenced variables are destroyed, and so on. Every
931program will have those first three lines: C<leave> is a list, and its
932children are all the statements in the block. Statements are delimited
933by C<nextstate>, so a block is a collection of C<nextstate> ops, with
934the ops to be performed for each statement being the children of
935C<nextstate>. C<enter> is a single op which functions as a marker.
936
937That's how Perl parsed the program, from top to bottom:
938
939 Program
940 |
941 Statement
942 |
943 =
944 / \
945 / \
946 $a +
947 / \
948 $b $c
949
950However, it's impossible to B<perform> the operations in this order:
951you have to find the values of C<$b> and C<$c> before you add them
952together, for instance. So, the other thread that runs through the op
953tree is the execution order: each op has a field C<op_next> which points
954to the next op to be run, so following these pointers tells us how perl
955executes the code. We can traverse the tree in this order using
956the C<exec> option to C<B::Terse>:
957
958 % perl -MO=Terse,exec -e '$a=$b+$c'
959 1 OP (0x8179928) enter
960 2 COP (0x81798c8) nextstate
961 3 SVOP (0x81796c8) gvsv GV (0x80fa4d4) *b
962 4 SVOP (0x8179798) gvsv GV (0x80efeb0) *c
963 5 BINOP (0x8179878) add [1]
964 6 SVOP (0x816dd38) gvsv GV (0x80fa468) *a
965 7 BINOP (0x81798a0) sassign
966 8 LISTOP (0x8179900) leave
967
968This probably makes more sense for a human: enter a block, start a
969statement. Get the values of C<$b> and C<$c>, and add them together.
970Find C<$a>, and assign one to the other. Then leave.
971
972The way Perl builds up these op trees in the parsing process can be
973unravelled by examining F<perly.y>, the YACC grammar. Let's take the
974piece we need to construct the tree for C<$a = $b + $c>
975
976 1 term : term ASSIGNOP term
977 2 { $$ = newASSIGNOP(OPf_STACKED, $1, $2, $3); }
978 3 | term ADDOP term
979 4 { $$ = newBINOP($2, 0, scalar($1), scalar($3)); }
980
981If you're not used to reading BNF grammars, this is how it works: You're
982fed certain things by the tokeniser, which generally end up in upper
983case. Here, C<ADDOP>, is provided when the tokeniser sees C<+> in your
984code. C<ASSIGNOP> is provided when C<=> is used for assigning. These are
985`terminal symbols', because you can't get any simpler than them.
986
987The grammar, lines one and three of the snippet above, tells you how to
988build up more complex forms. These complex forms, `non-terminal symbols'
989are generally placed in lower case. C<term> here is a non-terminal
990symbol, representing a single expression.
991
992The grammar gives you the following rule: you can make the thing on the
993left of the colon if you see all the things on the right in sequence.
994This is called a "reduction", and the aim of parsing is to completely
995reduce the input. There are several different ways you can perform a
996reduction, separated by vertical bars: so, C<term> followed by C<=>
997followed by C<term> makes a C<term>, and C<term> followed by C<+>
998followed by C<term> can also make a C<term>.
999
1000So, if you see two terms with an C<=> or C<+>, between them, you can
1001turn them into a single expression. When you do this, you execute the
1002code in the block on the next line: if you see C<=>, you'll do the code
1003in line 2. If you see C<+>, you'll do the code in line 4. It's this code
1004which contributes to the op tree.
1005
1006 | term ADDOP term
1007 { $$ = newBINOP($2, 0, scalar($1), scalar($3)); }
1008
1009What this does is creates a new binary op, and feeds it a number of
1010variables. The variables refer to the tokens: C<$1> is the first token in
1011the input, C<$2> the second, and so on - think regular expression
1012backreferences. C<$$> is the op returned from this reduction. So, we
1013call C<newBINOP> to create a new binary operator. The first parameter to
1014C<newBINOP>, a function in F<op.c>, is the op type. It's an addition
1015operator, so we want the type to be C<ADDOP>. We could specify this
1016directly, but it's right there as the second token in the input, so we
1017use C<$2>. The second parameter is the op's flags: 0 means `nothing
1018special'. Then the things to add: the left and right hand side of our
1019expression, in scalar context.
1020
1021=head2 Stacks
1022
1023When perl executes something like C<addop>, how does it pass on its
1024results to the next op? The answer is, through the use of stacks. Perl
1025has a number of stacks to store things it's currently working on, and
1026we'll look at the three most important ones here.
1027
1028=over 3
1029
1030=item Argument stack
1031
1032Arguments are passed to PP code and returned from PP code using the
1033argument stack, C<ST>. The typical way to handle arguments is to pop
1034them off the stack, deal with them how you wish, and then push the result
1035back onto the stack. This is how, for instance, the cosine operator
1036works:
1037
1038 NV value;
1039 value = POPn;
1040 value = Perl_cos(value);
1041 XPUSHn(value);
1042
1043We'll see a more tricky example of this when we consider Perl's macros
1044below. C<POPn> gives you the NV (floating point value) of the top SV on
1045the stack: the C<$x> in C<cos($x)>. Then we compute the cosine, and push
1046the result back as an NV. The C<X> in C<XPUSHn> means that the stack
1047should be extended if necessary - it can't be necessary here, because we
1048know there's room for one more item on the stack, since we've just
1049removed one! The C<XPUSH*> macros at least guarantee safety.
1050
1051Alternatively, you can fiddle with the stack directly: C<SP> gives you
1052the first element in your portion of the stack, and C<TOP*> gives you
1053the top SV/IV/NV/etc. on the stack. So, for instance, to do unary
1054negation of an integer:
1055
1056 SETi(-TOPi);
1057
1058Just set the integer value of the top stack entry to its negation.
1059
1060Argument stack manipulation in the core is exactly the same as it is in
1061XSUBs - see L<perlxstut>, L<perlxs> and L<perlguts> for a longer
1062description of the macros used in stack manipulation.
1063
1064=item Mark stack
1065
1066I say `your portion of the stack' above because PP code doesn't
1067necessarily get the whole stack to itself: if your function calls
1068another function, you'll only want to expose the arguments aimed for the
1069called function, and not (necessarily) let it get at your own data. The
1070way we do this is to have a `virtual' bottom-of-stack, exposed to each
1071function. The mark stack keeps bookmarks to locations in the argument
1072stack usable by each function. For instance, when dealing with a tied
1073variable, (internally, something with `P' magic) Perl has to call
1074methods for accesses to the tied variables. However, we need to separate
1075the arguments exposed to the method to the argument exposed to the
1076original function - the store or fetch or whatever it may be. Here's how
1077the tied C<push> is implemented; see C<av_push> in F<av.c>:
1078
1079 1 PUSHMARK(SP);
1080 2 EXTEND(SP,2);
1081 3 PUSHs(SvTIED_obj((SV*)av, mg));
1082 4 PUSHs(val);
1083 5 PUTBACK;
1084 6 ENTER;
1085 7 call_method("PUSH", G_SCALAR|G_DISCARD);
1086 8 LEAVE;
1087 9 POPSTACK;
13a2d996 1088
a422fd2d
SC
1089The lines which concern the mark stack are the first, fifth and last
1090lines: they save away, restore and remove the current position of the
1091argument stack.
1092
1093Let's examine the whole implementation, for practice:
1094
1095 1 PUSHMARK(SP);
1096
1097Push the current state of the stack pointer onto the mark stack. This is
1098so that when we've finished adding items to the argument stack, Perl
1099knows how many things we've added recently.
1100
1101 2 EXTEND(SP,2);
1102 3 PUSHs(SvTIED_obj((SV*)av, mg));
1103 4 PUSHs(val);
1104
1105We're going to add two more items onto the argument stack: when you have
1106a tied array, the C<PUSH> subroutine receives the object and the value
1107to be pushed, and that's exactly what we have here - the tied object,
1108retrieved with C<SvTIED_obj>, and the value, the SV C<val>.
1109
1110 5 PUTBACK;
1111
1112Next we tell Perl to make the change to the global stack pointer: C<dSP>
1113only gave us a local copy, not a reference to the global.
1114
1115 6 ENTER;
1116 7 call_method("PUSH", G_SCALAR|G_DISCARD);
1117 8 LEAVE;
1118
1119C<ENTER> and C<LEAVE> localise a block of code - they make sure that all
1120variables are tidied up, everything that has been localised gets
1121its previous value returned, and so on. Think of them as the C<{> and
1122C<}> of a Perl block.
1123
1124To actually do the magic method call, we have to call a subroutine in
1125Perl space: C<call_method> takes care of that, and it's described in
1126L<perlcall>. We call the C<PUSH> method in scalar context, and we're
1127going to discard its return value.
1128
1129 9 POPSTACK;
1130
1131Finally, we remove the value we placed on the mark stack, since we
1132don't need it any more.
1133
1134=item Save stack
1135
1136C doesn't have a concept of local scope, so perl provides one. We've
1137seen that C<ENTER> and C<LEAVE> are used as scoping braces; the save
1138stack implements the C equivalent of, for example:
1139
1140 {
1141 local $foo = 42;
1142 ...
1143 }
1144
1145See L<perlguts/Localising Changes> for how to use the save stack.
1146
1147=back
1148
1149=head2 Millions of Macros
1150
1151One thing you'll notice about the Perl source is that it's full of
1152macros. Some have called the pervasive use of macros the hardest thing
1153to understand, others find it adds to clarity. Let's take an example,
1154the code which implements the addition operator:
1155
1156 1 PP(pp_add)
1157 2 {
39644a26 1158 3 dSP; dATARGET; tryAMAGICbin(add,opASSIGN);
a422fd2d
SC
1159 4 {
1160 5 dPOPTOPnnrl_ul;
1161 6 SETn( left + right );
1162 7 RETURN;
1163 8 }
1164 9 }
1165
1166Every line here (apart from the braces, of course) contains a macro. The
1167first line sets up the function declaration as Perl expects for PP code;
1168line 3 sets up variable declarations for the argument stack and the
1169target, the return value of the operation. Finally, it tries to see if
1170the addition operation is overloaded; if so, the appropriate subroutine
1171is called.
1172
1173Line 5 is another variable declaration - all variable declarations start
1174with C<d> - which pops from the top of the argument stack two NVs (hence
1175C<nn>) and puts them into the variables C<right> and C<left>, hence the
1176C<rl>. These are the two operands to the addition operator. Next, we
1177call C<SETn> to set the NV of the return value to the result of adding
1178the two values. This done, we return - the C<RETURN> macro makes sure
1179that our return value is properly handled, and we pass the next operator
1180to run back to the main run loop.
1181
1182Most of these macros are explained in L<perlapi>, and some of the more
1183important ones are explained in L<perlxs> as well. Pay special attention
1184to L<perlguts/Background and PERL_IMPLICIT_CONTEXT> for information on
1185the C<[pad]THX_?> macros.
1186
1187
1188=head2 Poking at Perl
1189
1190To really poke around with Perl, you'll probably want to build Perl for
1191debugging, like this:
1192
1193 ./Configure -d -D optimize=-g
1194 make
1195
1196C<-g> is a flag to the C compiler to have it produce debugging
1197information which will allow us to step through a running program.
1198F<Configure> will also turn on the C<DEBUGGING> compilation symbol which
1199enables all the internal debugging code in Perl. There are a whole bunch
1200of things you can debug with this: L<perlrun> lists them all, and the
1201best way to find out about them is to play about with them. The most
1202useful options are probably
1203
1204 l Context (loop) stack processing
1205 t Trace execution
1206 o Method and overloading resolution
1207 c String/numeric conversions
1208
1209Some of the functionality of the debugging code can be achieved using XS
1210modules.
13a2d996 1211
a422fd2d
SC
1212 -Dr => use re 'debug'
1213 -Dx => use O 'Debug'
1214
1215=head2 Using a source-level debugger
1216
1217If the debugging output of C<-D> doesn't help you, it's time to step
1218through perl's execution with a source-level debugger.
1219
1220=over 3
1221
1222=item *
1223
1224We'll use C<gdb> for our examples here; the principles will apply to any
1225debugger, but check the manual of the one you're using.
1226
1227=back
1228
1229To fire up the debugger, type
1230
1231 gdb ./perl
1232
1233You'll want to do that in your Perl source tree so the debugger can read
1234the source code. You should see the copyright message, followed by the
1235prompt.
1236
1237 (gdb)
1238
1239C<help> will get you into the documentation, but here are the most
1240useful commands:
1241
1242=over 3
1243
1244=item run [args]
1245
1246Run the program with the given arguments.
1247
1248=item break function_name
1249
1250=item break source.c:xxx
1251
1252Tells the debugger that we'll want to pause execution when we reach
cea6626f 1253either the named function (but see L<perlguts/Internal Functions>!) or the given
a422fd2d
SC
1254line in the named source file.
1255
1256=item step
1257
1258Steps through the program a line at a time.
1259
1260=item next
1261
1262Steps through the program a line at a time, without descending into
1263functions.
1264
1265=item continue
1266
1267Run until the next breakpoint.
1268
1269=item finish
1270
1271Run until the end of the current function, then stop again.
1272
13a2d996 1273=item 'enter'
a422fd2d
SC
1274
1275Just pressing Enter will do the most recent operation again - it's a
1276blessing when stepping through miles of source code.
1277
1278=item print
1279
1280Execute the given C code and print its results. B<WARNING>: Perl makes
1281heavy use of macros, and F<gdb> is not aware of macros. You'll have to
1282substitute them yourself. So, for instance, you can't say
1283
1284 print SvPV_nolen(sv)
1285
1286but you have to say
1287
1288 print Perl_sv_2pv_nolen(sv)
1289
1290You may find it helpful to have a "macro dictionary", which you can
1291produce by saying C<cpp -dM perl.c | sort>. Even then, F<cpp> won't
1292recursively apply the macros for you.
1293
1294=back
1295
1296=head2 Dumping Perl Data Structures
1297
1298One way to get around this macro hell is to use the dumping functions in
1299F<dump.c>; these work a little like an internal
1300L<Devel::Peek|Devel::Peek>, but they also cover OPs and other structures
1301that you can't get at from Perl. Let's take an example. We'll use the
1302C<$a = $b + $c> we used before, but give it a bit of context:
1303C<$b = "6XXXX"; $c = 2.3;>. Where's a good place to stop and poke around?
1304
1305What about C<pp_add>, the function we examined earlier to implement the
1306C<+> operator:
1307
1308 (gdb) break Perl_pp_add
1309 Breakpoint 1 at 0x46249f: file pp_hot.c, line 309.
1310
cea6626f 1311Notice we use C<Perl_pp_add> and not C<pp_add> - see L<perlguts/Internal Functions>.
a422fd2d
SC
1312With the breakpoint in place, we can run our program:
1313
1314 (gdb) run -e '$b = "6XXXX"; $c = 2.3; $a = $b + $c'
1315
1316Lots of junk will go past as gdb reads in the relevant source files and
1317libraries, and then:
1318
1319 Breakpoint 1, Perl_pp_add () at pp_hot.c:309
39644a26 1320 309 dSP; dATARGET; tryAMAGICbin(add,opASSIGN);
a422fd2d
SC
1321 (gdb) step
1322 311 dPOPTOPnnrl_ul;
1323 (gdb)
1324
1325We looked at this bit of code before, and we said that C<dPOPTOPnnrl_ul>
1326arranges for two C<NV>s to be placed into C<left> and C<right> - let's
1327slightly expand it:
1328
1329 #define dPOPTOPnnrl_ul NV right = POPn; \
1330 SV *leftsv = TOPs; \
1331 NV left = USE_LEFT(leftsv) ? SvNV(leftsv) : 0.0
1332
1333C<POPn> takes the SV from the top of the stack and obtains its NV either
1334directly (if C<SvNOK> is set) or by calling the C<sv_2nv> function.
1335C<TOPs> takes the next SV from the top of the stack - yes, C<POPn> uses
1336C<TOPs> - but doesn't remove it. We then use C<SvNV> to get the NV from
1337C<leftsv> in the same way as before - yes, C<POPn> uses C<SvNV>.
1338
1339Since we don't have an NV for C<$b>, we'll have to use C<sv_2nv> to
1340convert it. If we step again, we'll find ourselves there:
1341
1342 Perl_sv_2nv (sv=0xa0675d0) at sv.c:1669
1343 1669 if (!sv)
1344 (gdb)
1345
1346We can now use C<Perl_sv_dump> to investigate the SV:
1347
1348 SV = PV(0xa057cc0) at 0xa0675d0
1349 REFCNT = 1
1350 FLAGS = (POK,pPOK)
1351 PV = 0xa06a510 "6XXXX"\0
1352 CUR = 5
1353 LEN = 6
1354 $1 = void
1355
1356We know we're going to get C<6> from this, so let's finish the
1357subroutine:
1358
1359 (gdb) finish
1360 Run till exit from #0 Perl_sv_2nv (sv=0xa0675d0) at sv.c:1671
1361 0x462669 in Perl_pp_add () at pp_hot.c:311
1362 311 dPOPTOPnnrl_ul;
1363
1364We can also dump out this op: the current op is always stored in
1365C<PL_op>, and we can dump it with C<Perl_op_dump>. This'll give us
1366similar output to L<B::Debug|B::Debug>.
1367
1368 {
1369 13 TYPE = add ===> 14
1370 TARG = 1
1371 FLAGS = (SCALAR,KIDS)
1372 {
1373 TYPE = null ===> (12)
1374 (was rv2sv)
1375 FLAGS = (SCALAR,KIDS)
1376 {
1377 11 TYPE = gvsv ===> 12
1378 FLAGS = (SCALAR)
1379 GV = main::b
1380 }
1381 }
1382
10f58044 1383# finish this later #
a422fd2d
SC
1384
1385=head2 Patching
1386
1387All right, we've now had a look at how to navigate the Perl sources and
1388some things you'll need to know when fiddling with them. Let's now get
1389on and create a simple patch. Here's something Larry suggested: if a
1390C<U> is the first active format during a C<pack>, (for example,
1391C<pack "U3C8", @stuff>) then the resulting string should be treated as
1392UTF8 encoded.
1393
1394How do we prepare to fix this up? First we locate the code in question -
1395the C<pack> happens at runtime, so it's going to be in one of the F<pp>
1396files. Sure enough, C<pp_pack> is in F<pp.c>. Since we're going to be
1397altering this file, let's copy it to F<pp.c~>.
1398
a6ec74c1
JH
1399[Well, it was in F<pp.c> when this tutorial was written. It has now been
1400split off with C<pp_unpack> to its own file, F<pp_pack.c>]
1401
a422fd2d
SC
1402Now let's look over C<pp_pack>: we take a pattern into C<pat>, and then
1403loop over the pattern, taking each format character in turn into
1404C<datum_type>. Then for each possible format character, we swallow up
1405the other arguments in the pattern (a field width, an asterisk, and so
1406on) and convert the next chunk input into the specified format, adding
1407it onto the output SV C<cat>.
1408
1409How do we know if the C<U> is the first format in the C<pat>? Well, if
1410we have a pointer to the start of C<pat> then, if we see a C<U> we can
1411test whether we're still at the start of the string. So, here's where
1412C<pat> is set up:
1413
1414 STRLEN fromlen;
1415 register char *pat = SvPVx(*++MARK, fromlen);
1416 register char *patend = pat + fromlen;
1417 register I32 len;
1418 I32 datumtype;
1419 SV *fromstr;
1420
1421We'll have another string pointer in there:
1422
1423 STRLEN fromlen;
1424 register char *pat = SvPVx(*++MARK, fromlen);
1425 register char *patend = pat + fromlen;
1426 + char *patcopy;
1427 register I32 len;
1428 I32 datumtype;
1429 SV *fromstr;
1430
1431And just before we start the loop, we'll set C<patcopy> to be the start
1432of C<pat>:
1433
1434 items = SP - MARK;
1435 MARK++;
1436 sv_setpvn(cat, "", 0);
1437 + patcopy = pat;
1438 while (pat < patend) {
1439
1440Now if we see a C<U> which was at the start of the string, we turn on
1441the UTF8 flag for the output SV, C<cat>:
1442
1443 + if (datumtype == 'U' && pat==patcopy+1)
1444 + SvUTF8_on(cat);
1445 if (datumtype == '#') {
1446 while (pat < patend && *pat != '\n')
1447 pat++;
1448
1449Remember that it has to be C<patcopy+1> because the first character of
1450the string is the C<U> which has been swallowed into C<datumtype!>
1451
1452Oops, we forgot one thing: what if there are spaces at the start of the
1453pattern? C<pack(" U*", @stuff)> will have C<U> as the first active
1454character, even though it's not the first thing in the pattern. In this
1455case, we have to advance C<patcopy> along with C<pat> when we see spaces:
1456
1457 if (isSPACE(datumtype))
1458 continue;
1459
1460needs to become
1461
1462 if (isSPACE(datumtype)) {
1463 patcopy++;
1464 continue;
1465 }
1466
1467OK. That's the C part done. Now we must do two additional things before
1468this patch is ready to go: we've changed the behaviour of Perl, and so
1469we must document that change. We must also provide some more regression
1470tests to make sure our patch works and doesn't create a bug somewhere
1471else along the line.
1472
b23b8711
MS
1473The regression tests for each operator live in F<t/op/>, and so we
1474make a copy of F<t/op/pack.t> to F<t/op/pack.t~>. Now we can add our
1475tests to the end. First, we'll test that the C<U> does indeed create
1476Unicode strings.
1477
1478t/op/pack.t has a sensible ok() function, but if it didn't we could
1479write one easily.
1480
1481 my $test = 1;
1482 sub ok {
812f5127 1483 my($ok, $name) = @_;
c274e827
MS
1484
1485 # You have to do it this way or VMS will get confused.
812f5127 1486 print $ok ? "ok $test - $name\n" : "not ok $test - $name\n";
c274e827 1487
17639bde
MS
1488 printf "# Failed test at line %d\n", (caller)[2] unless $ok;
1489
b23b8711
MS
1490 $test++;
1491 return $ok;
1492 }
1493
1494so instead of this:
a422fd2d
SC
1495
1496 print 'not ' unless "1.20.300.4000" eq sprintf "%vd", pack("U*",1,20,300,4000);
1497 print "ok $test\n"; $test++;
1498
b23b8711
MS
1499we can write the (somewhat) more sensible:
1500
812f5127
MS
1501 ok( "1.20.300.4000" eq sprintf "%vd", pack("U*",1,20,300,4000),
1502 "U* produces unicode" );
b23b8711 1503
a422fd2d
SC
1504Now we'll test that we got that space-at-the-beginning business right:
1505
812f5127
MS
1506 ok( "1.20.300.4000" eq sprintf "%vd", pack(" U*",1,20,300,4000),
1507 " with spaces at the beginning" );
a422fd2d
SC
1508
1509And finally we'll test that we don't make Unicode strings if C<U> is B<not>
1510the first active format:
1511
812f5127
MS
1512 ok( v1.20.300.4000 ne sprintf "%vd", pack("C0U*",1,20,300,4000),
1513 "U* not first isn't unicode" );
a422fd2d 1514
b1866b2d 1515Mustn't forget to change the number of tests which appears at the top, or
a422fd2d
SC
1516else the automated tester will get confused:
1517
1518 -print "1..156\n";
1519 +print "1..159\n";
1520
1521We now compile up Perl, and run it through the test suite. Our new
1522tests pass, hooray!
1523
1524Finally, the documentation. The job is never done until the paperwork is
1525over, so let's describe the change we've just made. The relevant place
1526is F<pod/perlfunc.pod>; again, we make a copy, and then we'll insert
1527this text in the description of C<pack>:
1528
1529 =item *
1530
1531 If the pattern begins with a C<U>, the resulting string will be treated
1532 as Unicode-encoded. You can force UTF8 encoding on in a string with an
1533 initial C<U0>, and the bytes that follow will be interpreted as Unicode
1534 characters. If you don't want this to happen, you can begin your pattern
1535 with C<C0> (or anything else) to force Perl not to UTF8 encode your
1536 string, and then follow this with a C<U*> somewhere in your pattern.
1537
1538All done. Now let's create the patch. F<Porting/patching.pod> tells us
1539that if we're making major changes, we should copy the entire directory
1540to somewhere safe before we begin fiddling, and then do
13a2d996 1541
a422fd2d
SC
1542 diff -ruN old new > patch
1543
1544However, we know which files we've changed, and we can simply do this:
1545
1546 diff -u pp.c~ pp.c > patch
1547 diff -u t/op/pack.t~ t/op/pack.t >> patch
1548 diff -u pod/perlfunc.pod~ pod/perlfunc.pod >> patch
1549
1550We end up with a patch looking a little like this:
1551
1552 --- pp.c~ Fri Jun 02 04:34:10 2000
1553 +++ pp.c Fri Jun 16 11:37:25 2000
1554 @@ -4375,6 +4375,7 @@
1555 register I32 items;
1556 STRLEN fromlen;
1557 register char *pat = SvPVx(*++MARK, fromlen);
1558 + char *patcopy;
1559 register char *patend = pat + fromlen;
1560 register I32 len;
1561 I32 datumtype;
1562 @@ -4405,6 +4406,7 @@
1563 ...
1564
1565And finally, we submit it, with our rationale, to perl5-porters. Job
1566done!
1567
f7e1e956
MS
1568=head2 Patching a core module
1569
1570This works just like patching anything else, with an extra
1571consideration. Many core modules also live on CPAN. If this is so,
1572patch the CPAN version instead of the core and send the patch off to
1573the module maintainer (with a copy to p5p). This will help the module
1574maintainer keep the CPAN version in sync with the core version without
1575constantly scanning p5p.
1576
acbe17fc
JP
1577=head2 Adding a new function to the core
1578
1579If, as part of a patch to fix a bug, or just because you have an
1580especially good idea, you decide to add a new function to the core,
1581discuss your ideas on p5p well before you start work. It may be that
1582someone else has already attempted to do what you are considering and
1583can give lots of good advice or even provide you with bits of code
1584that they already started (but never finished).
1585
1586You have to follow all of the advice given above for patching. It is
1587extremely important to test any addition thoroughly and add new tests
1588to explore all boundary conditions that your new function is expected
1589to handle. If your new function is used only by one module (e.g. toke),
1590then it should probably be named S_your_function (for static); on the
1591other hand, if you expect it to accessable from other functions in
1592Perl, you should name it Perl_your_function. See L<perlguts/Internal Functions>
1593for more details.
1594
1595The location of any new code is also an important consideration. Don't
1596just create a new top level .c file and put your code there; you would
1597have to make changes to Configure (so the Makefile is created properly),
1598as well as possibly lots of include files. This is strictly pumpking
1599business.
1600
1601It is better to add your function to one of the existing top level
1602source code files, but your choice is complicated by the nature of
1603the Perl distribution. Only the files that are marked as compiled
1604static are located in the perl executable. Everything else is located
1605in the shared library (or DLL if you are running under WIN32). So,
1606for example, if a function was only used by functions located in
1607toke.c, then your code can go in toke.c. If, however, you want to call
1608the function from universal.c, then you should put your code in another
1609location, for example util.c.
1610
1611In addition to writing your c-code, you will need to create an
1612appropriate entry in embed.pl describing your function, then run
1613'make regen_headers' to create the entries in the numerous header
1614files that perl needs to compile correctly. See L<perlguts/Internal Functions>
1615for information on the various options that you can set in embed.pl.
1616You will forget to do this a few (or many) times and you will get
1617warnings during the compilation phase. Make sure that you mention
1618this when you post your patch to P5P; the pumpking needs to know this.
1619
1620When you write your new code, please be conscious of existing code
1621conventions used in the perl source files. See <perlstyle> for
1622details. Although most of the guidelines discussed seem to focus on
1623Perl code, rather than c, they all apply (except when they don't ;).
1624See also I<Porting/patching.pod> file in the Perl source distribution
1625for lots of details about both formatting and submitting patches of
1626your changes.
1627
1628Lastly, TEST TEST TEST TEST TEST any code before posting to p5p.
1629Test on as many platforms as you can find. Test as many perl
1630Configure options as you can (e.g. MULTIPLICITY). If you have
1631profiling or memory tools, see L<EXTERNAL TOOLS FOR DEBUGGING PERL>
1632below for how to use them to futher test your code. Remember that
1633most of the people on P5P are doing this on their own time and
1634don't have the time to debug your code.
f7e1e956
MS
1635
1636=head2 Writing a test
1637
1638Every module and built-in function has an associated test file (or
1639should...). If you add or change functionality, you have to write a
1640test. If you fix a bug, you have to write a test so that bug never
1641comes back. If you alter the docs, it would be nice to test what the
1642new documentation says.
1643
1644In short, if you submit a patch you probably also have to patch the
1645tests.
1646
1647For modules, the test file is right next to the module itself.
1648F<lib/strict.t> tests F<lib/strict.pm>. This is a recent innovation,
1649so there are some snags (and it would be wonderful for you to brush
1650them out), but it basically works that way. Everything else lives in
1651F<t/>.
1652
1653=over 3
1654
1655=item F<t/base/>
1656
1657Testing of the absolute basic functionality of Perl. Things like
1658C<if>, basic file reads and writes, simple regexes, etc. These are
1659run first in the test suite and if any of them fail, something is
1660I<really> broken.
1661
1662=item F<t/cmd/>
1663
1664These test the basic control structures, C<if/else>, C<while>,
1665subroutines, etc...
1666
1667=item F<t/comp/>
1668
1669Tests basic issues of how Perl parses and compiles itself.
1670
1671=item F<t/io/>
1672
1673Tests for built-in IO functions, including command line arguments.
1674
1675=item F<t/lib/>
1676
1677The old home for the module tests, you shouldn't put anything new in
1678here. There are still some bits and pieces hanging around in here
1679that need to be moved. Perhaps you could move them? Thanks!
1680
1681=item F<t/op/>
1682
1683Tests for perl's built in functions that don't fit into any of the
1684other directories.
1685
1686=item F<t/pod/>
1687
1688Tests for POD directives. There are still some tests for the Pod
1689modules hanging around in here that need to be moved out into F<lib/>.
1690
1691=item F<t/run/>
1692
1693Testing features of how perl actually runs, including exit codes and
1694handling of PERL* environment variables.
1695
1696=back
1697
1698The core uses the same testing style as the rest of Perl, a simple
1699"ok/not ok" run through Test::Harness, but there are a few special
1700considerations.
1701
1702For most libraries and extensions, you'll want to use the Test::More
1703library rather than rolling your own test functions. If a module test
1704doesn't use Test::More, consider rewriting it so it does. For the
23acf682 1705rest it's best to use a simple C<print "ok $test_num\n"> style to avoid
f7e1e956
MS
1706broken core functionality from causing the whole test to collapse.
1707
1708When you say "make test" Perl uses the F<t/TEST> program to run the
1709test suite. All tests are run from the F<t/> directory, B<not> the
1710directory which contains the test. This causes some problems with the
1711tests in F<lib/>, so here's some opportunity for some patching.
1712
1713You must be triply conscious of cross-platform concerns. This usually
1714boils down to using File::Spec and avoiding things like C<fork()> and
1715C<system()> unless absolutely necessary.
1716
1717
902b9dbf
MF
1718=head1 EXTERNAL TOOLS FOR DEBUGGING PERL
1719
1720Sometimes it helps to use external tools while debugging and
1721testing Perl. This section tries to guide you through using
1722some common testing and debugging tools with Perl. This is
1723meant as a guide to interfacing these tools with Perl, not
1724as any kind of guide to the use of the tools themselves.
1725
1726=head2 Rational Software's Purify
1727
1728Purify is a commercial tool that is helpful in identifying
1729memory overruns, wild pointers, memory leaks and other such
1730badness. Perl must be compiled in a specific way for
1731optimal testing with Purify. Purify is available under
1732Windows NT, Solaris, HP-UX, SGI, and Siemens Unix.
1733
1734The only currently known leaks happen when there are
1735compile-time errors within eval or require. (Fixing these
1736is non-trivial, unfortunately, but they must be fixed
1737eventually.)
1738
1739=head2 Purify on Unix
1740
1741On Unix, Purify creates a new Perl binary. To get the most
1742benefit out of Purify, you should create the perl to Purify
1743using:
1744
1745 sh Configure -Accflags=-DPURIFY -Doptimize='-g' \
1746 -Uusemymalloc -Dusemultiplicity
1747
1748where these arguments mean:
1749
1750=over 4
1751
1752=item -Accflags=-DPURIFY
1753
1754Disables Perl's arena memory allocation functions, as well as
1755forcing use of memory allocation functions derived from the
1756system malloc.
1757
1758=item -Doptimize='-g'
1759
1760Adds debugging information so that you see the exact source
1761statements where the problem occurs. Without this flag, all
1762you will see is the source filename of where the error occurred.
1763
1764=item -Uusemymalloc
1765
1766Disable Perl's malloc so that Purify can more closely monitor
1767allocations and leaks. Using Perl's malloc will make Purify
1768report most leaks in the "potential" leaks category.
1769
1770=item -Dusemultiplicity
1771
1772Enabling the multiplicity option allows perl to clean up
1773thoroughly when the interpreter shuts down, which reduces the
1774number of bogus leak reports from Purify.
1775
1776=back
1777
1778Once you've compiled a perl suitable for Purify'ing, then you
1779can just:
1780
1781 make pureperl
1782
1783which creates a binary named 'pureperl' that has been Purify'ed.
1784This binary is used in place of the standard 'perl' binary
1785when you want to debug Perl memory problems.
1786
1787As an example, to show any memory leaks produced during the
1788standard Perl testset you would create and run the Purify'ed
1789perl as:
1790
1791 make pureperl
1792 cd t
1793 ../pureperl -I../lib harness
1794
1795which would run Perl on test.pl and report any memory problems.
1796
1797Purify outputs messages in "Viewer" windows by default. If
1798you don't have a windowing environment or if you simply
1799want the Purify output to unobtrusively go to a log file
1800instead of to the interactive window, use these following
1801options to output to the log file "perl.log":
1802
1803 setenv PURIFYOPTIONS "-chain-length=25 -windows=no \
1804 -log-file=perl.log -append-logfile=yes"
1805
1806If you plan to use the "Viewer" windows, then you only need this option:
1807
1808 setenv PURIFYOPTIONS "-chain-length=25"
1809
1810=head2 Purify on NT
1811
1812Purify on Windows NT instruments the Perl binary 'perl.exe'
1813on the fly. There are several options in the makefile you
1814should change to get the most use out of Purify:
1815
1816=over 4
1817
1818=item DEFINES
1819
1820You should add -DPURIFY to the DEFINES line so the DEFINES
1821line looks something like:
1822
1823 DEFINES = -DWIN32 -D_CONSOLE -DNO_STRICT $(CRYPT_FLAG) -DPURIFY=1
1824
1825to disable Perl's arena memory allocation functions, as
1826well as to force use of memory allocation functions derived
1827from the system malloc.
1828
1829=item USE_MULTI = define
1830
1831Enabling the multiplicity option allows perl to clean up
1832thoroughly when the interpreter shuts down, which reduces the
1833number of bogus leak reports from Purify.
1834
1835=item #PERL_MALLOC = define
1836
1837Disable Perl's malloc so that Purify can more closely monitor
1838allocations and leaks. Using Perl's malloc will make Purify
1839report most leaks in the "potential" leaks category.
1840
1841=item CFG = Debug
1842
1843Adds debugging information so that you see the exact source
1844statements where the problem occurs. Without this flag, all
1845you will see is the source filename of where the error occurred.
1846
1847=back
1848
1849As an example, to show any memory leaks produced during the
1850standard Perl testset you would create and run Purify as:
1851
1852 cd win32
1853 make
1854 cd ../t
1855 purify ../perl -I../lib harness
1856
1857which would instrument Perl in memory, run Perl on test.pl,
1858then finally report any memory problems.
1859
09187cb1
JH
1860=head2 Compaq's/Digital's Third Degree
1861
1862Third Degree is a tool for memory leak detection and memory access checks.
1863It is one of the many tools in the ATOM toolkit. The toolkit is only
1864available on Tru64 (formerly known as Digital UNIX formerly known as
1865DEC OSF/1).
1866
1867When building Perl, you must first run Configure with -Doptimize=-g
1868and -Uusemymalloc flags, after that you can use the make targets
51a35ef1
JH
1869"perl.third" and "test.third". (What is required is that Perl must be
1870compiled using the C<-g> flag, you may need to re-Configure.)
09187cb1 1871
64cea5fd 1872The short story is that with "atom" you can instrument the Perl
83f0ef60 1873executable to create a new executable called F<perl.third>. When the
4ae3d70a 1874instrumented executable is run, it creates a log of dubious memory
83f0ef60 1875traffic in file called F<perl.3log>. See the manual pages of atom and
4ae3d70a
JH
1876third for more information. The most extensive Third Degree
1877documentation is available in the Compaq "Tru64 UNIX Programmer's
1878Guide", chapter "Debugging Programs with Third Degree".
64cea5fd 1879
83f0ef60 1880The "test.third" leaves a lot of files named F<perl.3log.*> in the t/
64cea5fd
JH
1881subdirectory. There is a problem with these files: Third Degree is so
1882effective that it finds problems also in the system libraries.
83f0ef60
JH
1883Therefore there are certain types of errors that you should ignore in
1884your debugging. Errors with stack traces matching
64cea5fd
JH
1885
1886 __actual_atof|__catgets|_doprnt|__exc_|__exec|_findio|__localtime|setlocale|__sia_|__strxfrm
1887
1888(all in libc.so) are known to be non-serious. You can also
1889ignore the combinations
1890
1891 Perl_gv_fetchfile() calling strcpy()
1892 S_doopen_pmc() calling strcmp()
1893
1894causing "rih" (reading invalid heap) errors.
1895
1896There are also leaks that for given certain definition of a leak,
1897aren't. See L</PERL_DESTRUCT_LEVEL> for more information.
1898
1899=head2 PERL_DESTRUCT_LEVEL
1900
1901If you want to run any of the tests yourself manually using the
1902pureperl or perl.third executables, please note that by default
1903perl B<does not> explicitly cleanup all the memory it has allocated
1904(such as global memory arenas) but instead lets the exit() of
1905the whole program "take care" of such allocations, also known
1906as "global destruction of objects".
1907
1908There is a way to tell perl to do complete cleanup: set the
1909environment variable PERL_DESTRUCT_LEVEL to a non-zero value.
1910The t/TEST wrapper does set this to 2, and this is what you
1911need to do too, if you don't want to see the "global leaks":
1912
1913 PERL_DESTRUCT_LEVEL=2 ./perl.third t/foo/bar.t
09187cb1 1914
51a35ef1
JH
1915=head2 Profiling
1916
1917Depending on your platform there are various of profiling Perl.
1918
1919There are two commonly used techniques of profiling executables:
10f58044 1920I<statistical time-sampling> and I<basic-block counting>.
51a35ef1
JH
1921
1922The first method takes periodically samples of the CPU program
1923counter, and since the program counter can be correlated with the code
1924generated for functions, we get a statistical view of in which
1925functions the program is spending its time. The caveats are that very
1926small/fast functions have lower probability of showing up in the
1927profile, and that periodically interrupting the program (this is
1928usually done rather frequently, in the scale of milliseconds) imposes
1929an additional overhead that may skew the results. The first problem
1930can be alleviated by running the code for longer (in general this is a
1931good idea for profiling), the second problem is usually kept in guard
1932by the profiling tools themselves.
1933
10f58044 1934The second method divides up the generated code into I<basic blocks>.
51a35ef1
JH
1935Basic blocks are sections of code that are entered only in the
1936beginning and exited only at the end. For example, a conditional jump
1937starts a basic block. Basic block profiling usually works by
10f58044 1938I<instrumenting> the code by adding I<enter basic block #nnnn>
51a35ef1
JH
1939book-keeping code to the generated code. During the execution of the
1940code the basic block counters are then updated appropriately. The
1941caveat is that the added extra code can skew the results: again, the
1942profiling tools usually try to factor their own effects out of the
1943results.
1944
83f0ef60
JH
1945=head2 Gprof Profiling
1946
51a35ef1
JH
1947gprof is a profiling tool available in many UNIX platforms,
1948it uses F<statistical time-sampling>.
83f0ef60
JH
1949
1950You can build a profiled version of perl called "perl.gprof" by
51a35ef1
JH
1951invoking the make target "perl.gprof" (What is required is that Perl
1952must be compiled using the C<-pg> flag, you may need to re-Configure).
1953Running the profiled version of Perl will create an output file called
1954F<gmon.out> is created which contains the profiling data collected
1955during the execution.
83f0ef60
JH
1956
1957The gprof tool can then display the collected data in various ways.
1958Usually gprof understands the following options:
1959
1960=over 4
1961
1962=item -a
1963
1964Suppress statically defined functions from the profile.
1965
1966=item -b
1967
1968Suppress the verbose descriptions in the profile.
1969
1970=item -e routine
1971
1972Exclude the given routine and its descendants from the profile.
1973
1974=item -f routine
1975
1976Display only the given routine and its descendants in the profile.
1977
1978=item -s
1979
1980Generate a summary file called F<gmon.sum> which then may be given
1981to subsequent gprof runs to accumulate data over several runs.
1982
1983=item -z
1984
1985Display routines that have zero usage.
1986
1987=back
1988
1989For more detailed explanation of the available commands and output
1990formats, see your own local documentation of gprof.
1991
51a35ef1
JH
1992=head2 GCC gcov Profiling
1993
10f58044 1994Starting from GCC 3.0 I<basic block profiling> is officially available
51a35ef1
JH
1995for the GNU CC.
1996
1997You can build a profiled version of perl called F<perl.gcov> by
1998invoking the make target "perl.gcov" (what is required that Perl must
1999be compiled using gcc with the flags C<-fprofile-arcs
2000-ftest-coverage>, you may need to re-Configure).
2001
2002Running the profiled version of Perl will cause profile output to be
2003generated. For each source file an accompanying ".da" file will be
2004created.
2005
2006To display the results you use the "gcov" utility (which should
2007be installed if you have gcc 3.0 or newer installed). F<gcov> is
2008run on source code files, like this
2009
2010 gcov sv.c
2011
2012which will cause F<sv.c.gcov> to be created. The F<.gcov> files
2013contain the source code annotated with relative frequencies of
2014execution indicated by "#" markers.
2015
2016Useful options of F<gcov> include C<-b> which will summarise the
2017basic block, branch, and function call coverage, and C<-c> which
2018instead of relative frequencies will use the actual counts. For
2019more information on the use of F<gcov> and basic block profiling
2020with gcc, see the latest GNU CC manual, as of GCC 3.0 see
2021
2022 http://gcc.gnu.org/onlinedocs/gcc-3.0/gcc.html
2023
2024and its section titled "8. gcov: a Test Coverage Program"
2025
2026 http://gcc.gnu.org/onlinedocs/gcc-3.0/gcc_8.html#SEC132
2027
4ae3d70a
JH
2028=head2 Pixie Profiling
2029
51a35ef1
JH
2030Pixie is a profiling tool available on IRIX and Tru64 (aka Digital
2031UNIX aka DEC OSF/1) platforms. Pixie does its profiling using
10f58044 2032I<basic-block counting>.
4ae3d70a 2033
83f0ef60 2034You can build a profiled version of perl called F<perl.pixie> by
51a35ef1
JH
2035invoking the make target "perl.pixie" (what is required is that Perl
2036must be compiled using the C<-g> flag, you may need to re-Configure).
2037
2038In Tru64 a file called F<perl.Addrs> will also be silently created,
2039this file contains the addresses of the basic blocks. Running the
2040profiled version of Perl will create a new file called "perl.Counts"
2041which contains the counts for the basic block for that particular
2042program execution.
4ae3d70a 2043
51a35ef1 2044To display the results you use the F<prof> utility. The exact
4ae3d70a
JH
2045incantation depends on your operating system, "prof perl.Counts" in
2046IRIX, and "prof -pixie -all -L. perl" in Tru64.
2047
6c41479b
JH
2048In IRIX the following prof options are available:
2049
2050=over 4
2051
2052=item -h
2053
2054Reports the most heavily used lines in descending order of use.
6e36760b 2055Useful for finding the hotspot lines.
6c41479b
JH
2056
2057=item -l
2058
2059Groups lines by procedure, with procedures sorted in descending order of use.
2060Within a procedure, lines are listed in source order.
6e36760b 2061Useful for finding the hotspots of procedures.
6c41479b
JH
2062
2063=back
2064
2065In Tru64 the following options are available:
2066
2067=over 4
2068
3958b146 2069=item -p[rocedures]
6c41479b 2070
3958b146 2071Procedures sorted in descending order by the number of cycles executed
6e36760b 2072in each procedure. Useful for finding the hotspot procedures.
6c41479b
JH
2073(This is the default option.)
2074
24000d2f 2075=item -h[eavy]
6c41479b 2076
6e36760b
JH
2077Lines sorted in descending order by the number of cycles executed in
2078each line. Useful for finding the hotspot lines.
6c41479b 2079
24000d2f 2080=item -i[nvocations]
6c41479b 2081
6e36760b
JH
2082The called procedures are sorted in descending order by number of calls
2083made to the procedures. Useful for finding the most used procedures.
6c41479b 2084
24000d2f 2085=item -l[ines]
6c41479b
JH
2086
2087Grouped by procedure, sorted by cycles executed per procedure.
6e36760b 2088Useful for finding the hotspots of procedures.
6c41479b
JH
2089
2090=item -testcoverage
2091
2092The compiler emitted code for these lines, but the code was unexecuted.
2093
24000d2f 2094=item -z[ero]
6c41479b
JH
2095
2096Unexecuted procedures.
2097
aa500c9e 2098=back
6c41479b
JH
2099
2100For further information, see your system's manual pages for pixie and prof.
4ae3d70a 2101
a422fd2d
SC
2102=head2 CONCLUSION
2103
2104We've had a brief look around the Perl source, an overview of the stages
2105F<perl> goes through when it's running your code, and how to use a
902b9dbf
MF
2106debugger to poke at the Perl guts. We took a very simple problem and
2107demonstrated how to solve it fully - with documentation, regression
2108tests, and finally a patch for submission to p5p. Finally, we talked
2109about how to use external tools to debug and test Perl.
a422fd2d
SC
2110
2111I'd now suggest you read over those references again, and then, as soon
2112as possible, get your hands dirty. The best way to learn is by doing,
2113so:
2114
2115=over 3
2116
2117=item *
2118
2119Subscribe to perl5-porters, follow the patches and try and understand
2120them; don't be afraid to ask if there's a portion you're not clear on -
2121who knows, you may unearth a bug in the patch...
2122
2123=item *
2124
2125Keep up to date with the bleeding edge Perl distributions and get
2126familiar with the changes. Try and get an idea of what areas people are
2127working on and the changes they're making.
2128
2129=item *
2130
3e148164 2131Do read the README associated with your operating system, e.g. README.aix
a1f349fd
MB
2132on the IBM AIX OS. Don't hesitate to supply patches to that README if
2133you find anything missing or changed over a new OS release.
2134
2135=item *
2136
a422fd2d
SC
2137Find an area of Perl that seems interesting to you, and see if you can
2138work out how it works. Scan through the source, and step over it in the
2139debugger. Play, poke, investigate, fiddle! You'll probably get to
2140understand not just your chosen area but a much wider range of F<perl>'s
2141activity as well, and probably sooner than you'd think.
2142
2143=back
2144
2145=over 3
2146
2147=item I<The Road goes ever on and on, down from the door where it began.>
2148
2149=back
2150
2151If you can do these things, you've started on the long road to Perl porting.
2152Thanks for wanting to help make Perl better - and happy hacking!
2153
e8cd7eae
GS
2154=head1 AUTHOR
2155
2156This document was written by Nathan Torkington, and is maintained by
2157the perl5-porters mailing list.
2158