3 Pumpkin - Notes on handling the Perl Patch Pumpkin
7 There is no simple synopsis, yet.
11 This document attempts to begin to describe some of the considerations
12 involved in patching, porting, and maintaining perl.
14 This document is still under construction, and still subject to
15 significant changes. Still, I hope parts of it will be useful,
16 so I'm releasing it even though it's not done.
18 For the most part, it's a collection of anecdotal information that
19 already assumes some familiarity with the Perl sources. I really need
20 an introductory section that describes the organization of the sources
21 and all the various auxiliary files that are part of the distribution.
23 =head1 Where Do I Get Perl Sources and Related Material?
25 The Comprehensive Perl Archive Network (or CPAN) is the place to go.
26 There are many mirrors, but the easiest thing to use is probably
27 http://www.perl.com/CPAN/README.html , which automatically points you to a
28 mirror site "close" to you.
30 =head2 Perl5-porters mailing list
32 The mailing list perl5-porters@perl.org
33 is the main group working with the development of perl. If you're
34 interested in all the latest developments, you should definitely
35 subscribe. The list is high volume, but generally has a
36 fairly low noise level.
38 Subscribe by sending the message (in the body of your letter)
40 subscribe perl5-porters
42 to perl5-porters-request@perl.org .
44 Archives of the list are held at:
46 http://www.rosat.mpe-garching.mpg.de/mailing-lists/perl-porters/
48 =head1 How are Perl Releases Numbered?
50 Beginning with v5.6.0, even versions will stand for maintenance releases
51 and odd versions for development releases, i.e., v5.6.x for maintenance
52 releases, and v5.7.x for development releases. Before v5.6.0, subversions
53 _01 through _49 were reserved for bug-fix maintenance releases, and
54 subversions _50 through _99 for unstable development versions.
56 For example, in v5.6.1, the revision number is 5, the version is 6,
57 and 1 is the subversion.
59 For compatibility with the older numbering scheme the composite floating
60 point version number continues to be available as the magic variable $],
61 and amounts to C<$revision + $version/1000 + $subversion/100000>. This
62 can still be used in comparisons.
64 print "You've got an old perl\n" if $] < 5.005_03;
66 In addition, the version is also available as a string in $^V.
68 print "You've got a new perl\n" if $^V and $^V ge v5.6.0;
70 You can also require particular version (or later) with:
74 or using the new syntax available only from v5.6 onward:
78 At some point in the future, we may need to decide what to call the
79 next big revision. In the .package file used by metaconfig to
80 generate Configure, there are two variables that might be relevant:
81 $baserev=5 and $package=perl5.
83 Perl releases produced by the members of perl5-porters are usually
84 available on CPAN in the F<src/5.0/maint> and F<src/5.0/devel>
87 =head2 Maintenance and Development Subversions
89 The first rule of maintenance work is "First, do no harm."
91 Trial releases of bug-fix maintenance releases are announced on
92 perl5-porters. Trial releases use the new subversion number (to avoid
93 testers installing it over the previous release) and include a 'local
94 patch' entry in patchlevel.h. The distribution file contains the
95 string C<MAINT_TRIAL> to make clear that the file is not meant for
98 In general, the names of official distribution files for the public
99 always match the regular expression:
101 ^perl\d+\.(\d+)\.\d+(-MAINT_TRIAL_\d+)\.tar\.gz$
103 C<$1> in the pattern is always an even number for maintenance
104 versions, and odd for developer releases.
106 In the past it has been observed that pumkings tend to invent new
107 naming conventions on the fly. If you are a pumpking, before you
108 invent a new name for any of the three types of perl distributions,
109 please inform the guys from the CPAN who are doing indexing and
110 provide the trees of symlinks and the like. They will have to know
111 I<in advance> what you decide.
113 =head2 Why is it called the patch pumpkin?
115 Chip Salzenberg gets credit for that, with a nod to his cow orker,
116 David Croy. We had passed around various names (baton, token, hot
117 potato) but none caught on. Then, Chip asked:
121 Who has the patch pumpkin?
123 To explain: David Croy once told me once that at a previous job,
124 there was one tape drive and multiple systems that used it for backups.
125 But instead of some high-tech exclusion software, they used a low-tech
126 method to prevent multiple simultaneous backups: a stuffed pumpkin.
127 No one was allowed to make backups unless they had the "backup pumpkin".
133 =head1 Philosophical Issues in Patching and Porting Perl
135 There are no absolute rules, but there are some general guidelines I
136 have tried to follow as I apply patches to the perl sources.
137 (This section is still under construction.)
139 =head2 Solve problems as generally as possible
141 Never implement a specific restricted solution to a problem when you
142 can solve the same problem in a more general, flexible way.
144 For example, for dynamic loading to work on some SVR4 systems, we had
145 to build a shared libperl.so library. In order to build "FAT" binaries
146 on NeXT 4.0 systems, we had to build a special libperl library. Rather
147 than continuing to build a contorted nest of special cases, I
148 generalized the process of building libperl so that NeXT and SVR4 users
149 could still get their work done, but others could build a shared
150 libperl if they wanted to as well.
152 Contain your changes carefully. Assume nothing about other operating
153 systems, not even closely related ones. Your changes must not affect
156 Spy shamelessly on how similar patching or porting issues have been
159 If feasible, try to keep filenames 8.3-compliant to humor those poor
160 souls that get joy from running Perl under such dire limitations.
162 =head2 Seek consensus on major changes
164 If you are making big changes, don't do it in secret. Discuss the
165 ideas in advance on perl5-porters.
167 =head2 Keep the documentation up-to-date
169 If your changes may affect how users use perl, then check to be sure
170 that the documentation is in sync with your changes. Be sure to
171 check all the files F<pod/*.pod> and also the F<INSTALL> document.
173 Consider writing the appropriate documentation first and then
174 implementing your change to correspond to the documentation.
176 =head2 Avoid machine-specific #ifdef's
178 To the extent reasonable, try to avoid machine-specific #ifdef's in
179 the sources. Instead, use feature-specific #ifdef's. The reason is
180 that the machine-specific #ifdef's may not be valid across major
181 releases of the operating system. Further, the feature-specific tests
182 may help out folks on another platform who have the same problem.
184 =head2 Machine-specific files
190 If you have many machine-specific #defines or #includes, consider
191 creating an "osish.h" (os2ish.h, vmsish.h, and so on) and including
192 that in perl.h. If you have several machine-specific files (function
193 emulations, function stubs, build utility wrappers) you may create a
194 separate subdirectory (djgpp, win32) and put the files in there.
195 Remember to update C<MANIFEST> when you add files.
197 If your system supports dynamic loading but none of the existing
198 methods at F<ext/DynaLoader/dl_*.xs> work for you, you must write
199 a new one. Study the existing ones to see what kind of interface
204 There are two kinds of hints: hints for building Perl and hints for
205 extensions. The former live in the C<hints> subdirectory, the latter
206 in C<ext/*/hints> subdirectories.
208 The top level hints are Bourne-shell scripts that set, modify and
209 unset appropriate Configure variables, based on the Configure command
210 line options and possibly existing config.sh and Policy.sh files from
211 previous Configure runs.
213 The extension hints are written in Perl (by the time they are used
214 miniperl has been built) and control the building of their respective
215 extensions. They can be used to for example manipulate compilation
218 =item build and installation Makefiles, scripts, and so forth
220 Sometimes you will also need to tweak the Perl build and installation
221 procedure itself, like for example F<Makefile.SH> and F<installperl>.
222 Tread very carefully, even more than usual. Contain your changes
227 Many of the tests in C<t> subdirectory assume machine-specific things
228 like existence of certain functions, something about filesystem
229 semantics, certain external utilities and their error messages. Use
230 the C<$^O> and the C<Config> module (which contains the results of the
231 Configure run, in effect the C<config.sh> converted to Perl) to either
232 skip (preferably not) or customize (preferable) the tests for your
237 Certain standard modules may need updating if your operating system
238 sports for example a native filesystem naming. You may want to update
239 some or all of the modules File::Basename, File::Spec, File::Path, and
240 File::Copy to become aware of your native filesystem syntax and
245 If your operating system comes from outside UNIX you almost certainly
246 will have differences in the available operating system functionality
247 (missing system calls, different semantics, whatever). Please
248 document these at F<pod/perlport.pod>. If your operating system is
249 the first B<not> to have a system call also update the list of
250 "portability-bewares" at the beginning of F<pod/perlfunc.pod>.
252 A file called F<README.youros> at the top level that explains things
253 like how to install perl at this platform, where to get any possibly
254 required additional software, and for example what test suite errors
255 to expect, is nice too. Such files are in the process of being written
256 in pod format and will eventually be renamed F<INSTALL.youros>.
258 You may also want to write a separate F<.pod> file for your operating
259 system to tell about existing mailing lists, os-specific modules,
260 documentation, whatever. Please name these along the lines of
261 F<perl>I<youros>.pod. [unfinished: where to put this file (the pod/
262 subdirectory, of course: but more importantly, which/what index files
267 =head2 Allow for lots of testing
269 We should never release a main version without testing it as a
272 =head2 Test popular applications and modules.
274 We should never release a main version without testing whether or not
275 it breaks various popular modules and applications. A partial list of
276 such things would include majordomo, metaconfig, apache, Tk, CGI,
277 libnet, and libwww, to name just a few. Of course it's quite possible
278 that some of those things will be just plain broken and need to be fixed,
279 but, in general, we ought to try to avoid breaking widely-installed
282 =head2 Automated generation of derivative files
284 The F<embed.h>, F<keywords.h>, F<opcode.h>, and F<perltoc.pod> files
285 are all automatically generated by perl scripts. In general, don't
286 patch these directly; patch the data files instead.
288 F<Configure> and F<config_h.SH> are also automatically generated by
289 B<metaconfig>. In general, you should patch the metaconfig units
290 instead of patching these files directly. However, very minor changes
291 to F<Configure> may be made in between major sync-ups with the
292 metaconfig units, which tends to be complicated operations. But be
293 careful, this can quickly spiral out of control. Running metaconfig
296 Also F<Makefile> is automatically produced from F<Makefile.SH>.
297 In general, look out for all F<*.SH> files.
299 Finally, the sample files in the F<Porting/> subdirectory are
300 generated automatically by the script F<U/mksample> included
301 with the metaconfig units. See L<"run metaconfig"> below for
302 information on obtaining the metaconfig units.
304 =head1 How to Make a Distribution
306 There really ought to be a 'make dist' target, but there isn't.
307 The 'dist' suite of tools also contains a number of tools that I haven't
308 learned how to use yet. Some of them may make this all a bit easier.
310 Here are the steps I go through to prepare a patch & distribution.
312 Lots of it could doubtless be automated but isn't. The Porting/makerel
313 (make release) perl script does now help automate some parts of it.
315 =head2 Announce your intentions
317 First, you should volunteer out loud to take the patch pumpkin. It's
318 generally counter-productive to have multiple people working in secret
321 At the same time, announce what you plan to do with the patch pumpkin,
322 to allow folks a chance to object or suggest alternatives, or do it for
323 you. Naturally, the patch pumpkin holder ought to incorporate various
324 bug fixes and documentation improvements that are posted while he or
325 she has the pumpkin, but there might also be larger issues at stake.
327 One of the precepts of the subversion idea is that we shouldn't give
328 the patch pumpkin to anyone unless we have some idea what he or she
329 is going to do with it.
331 =head2 refresh pod/perltoc.pod
333 Presumably, you have done a full C<make> in your working source
334 directory. Before you C<make spotless> (if you do), and if you have
335 changed any documentation in any module or pod file, change to the
336 F<pod> directory and run C<make toc>.
338 =head2 run installhtml to check the validity of the pod files
340 =head2 update patchlevel.h
342 Don't be shy about using the subversion number, even for a relatively
343 modest patch. We've never even come close to using all 99 subversions,
344 and it's better to have a distinctive number for your patch. If you
345 need feedback on your patch, go ahead and issue it and promise to
346 incorporate that feedback quickly (e.g. within 1 week) and send out a
349 =head2 run metaconfig
351 If you need to make changes to Configure or config_h.SH, it may be best to
352 change the appropriate metaconfig units instead, and regenerate Configure.
356 will regenerate Configure and config_h.SH. Much more information
357 on obtaining and running metaconfig is in the F<U/README> file
358 that comes with Perl's metaconfig units. Perl's metaconfig units
359 should be available on CPAN. A set of units that will work with
360 perl5.005 is in the file F<mc_units-5.005_00-01.tar.gz> under
361 http://www.perl.com/CPAN/authors/id/ANDYD/ . The mc_units tar file
362 should be unpacked in your main perl source directory. Note: those
363 units were for use with 5.005. There may have been changes since then.
364 Check for later versions or contact perl5-porters@perl.org to obtain a
365 pointer to the current version.
367 Alternatively, do consider if the F<*ish.h> files might be a better
368 place for your changes.
372 Make sure the MANIFEST is up-to-date. You can use dist's B<manicheck>
373 program for this. You can also use
375 perl -w -MExtUtils::Manifest=fullcheck -e fullcheck
377 Both commands will also list extra files in the directory that are not
380 The MANIFEST is normally sorted.
382 If you are using metaconfig to regenerate Configure, then you should note
383 that metaconfig actually uses MANIFEST.new, so you want to be sure
384 MANIFEST.new is up-to-date too. I haven't found the MANIFEST/MANIFEST.new
385 distinction particularly useful, but that's probably because I still haven't
386 learned how to use the full suite of tools in the dist distribution.
388 =head2 Check permissions
390 All the tests in the t/ directory ought to be executable. The
391 main makefile used to do a 'chmod t/*/*.t', but that resulted in
392 a self-modifying distribution--something some users would strongly
393 prefer to avoid. The F<t/TEST> script will check for this
394 and do the chmod if needed, but the tests still ought to be
397 In all, the following files should probably be executable:
412 vms/ext/Stdio/test.pl
416 Other things ought to be readable, at least :-).
418 Probably, the permissions for the files could be encoded in MANIFEST
419 somehow, but I'm reluctant to change MANIFEST itself because that
420 could break old scripts that use MANIFEST.
422 I seem to recall that some SVR3 systems kept some sort of file that listed
423 permissions for system files; something like that might be appropriate.
427 This will build a config.sh and config.h. You can skip this if you haven't
428 changed Configure or config_h.SH at all. I use the following command
430 sh Configure -Dprefix=/opt/perl -Doptimize=-O -Dusethreads \
432 -Dcf_email='yourname@yourhost.yourplace.com' \
433 -Dperladmin='yourname@yourhost.yourplace.com' \
434 -Dmydomain='.yourplace.com' \
435 -Dmyhostname='yourhost' \
438 =head2 Update Porting/config.sh and Porting/config_H
441 This section needs revision. We're currently working on easing
442 the task of keeping the vms, win32, and plan9 config.sh info
443 up-to-date. The plan is to use keep up-to-date 'canned' config.sh
444 files in the appropriate subdirectories and then generate 'canned'
445 config.h files for vms, win32, etc. from the generic config.sh file.
446 This is to ease maintenance. When Configure gets updated, the parts
447 sometimes get scrambled around, and the changes in config_H can
448 sometimes be very hard to follow. config.sh, on the other hand, can
449 safely be sorted, so it's easy to track (typically very small) changes
450 to config.sh and then propoagate them to a canned 'config.h' by any
451 number of means, including a perl script in win32/ or carrying
452 config.sh and config_h.SH to a Unix system and running sh
453 config_h.SH.) Vms uses configure.com to generate its own config.sh
454 and config.h. If you want to add a new variable to config.sh check
455 with vms folk how to add it to configure.com too.
458 The Porting/config.sh and Porting/config_H files are provided to
459 help those folks who can't run Configure. It is important to keep
460 them up-to-date. If you have changed config_h.SH, those changes must
461 be reflected in config_H as well. (The name config_H was chosen to
462 distinguish the file from config.h even on case-insensitive file systems.)
463 Simply edit the existing config_H file; keep the first few explanatory
464 lines and then copy your new config.h below.
466 It may also be necessary to update win32/config.?c, and
467 plan9/config.plan9, though you should be quite careful in doing so if
468 you are not familiar with those systems. You might want to issue your
469 patch with a promise to quickly issue a follow-up that handles those
472 =head2 make run_byacc
474 If you have byacc-1.8.2 (available from CPAN), and if there have been
475 changes to F<perly.y>, you can regenerate the F<perly.c> file. The
476 run_byacc makefile target does this by running byacc and then applying
477 some patches so that byacc dynamically allocates space, rather than
478 having fixed limits. This patch is handled by the F<perly.fixer>
479 script. Depending on the nature of the changes to F<perly.y>, you may
480 or may not have to hand-edit the patch to apply correctly. If you do,
481 you should include the edited patch in the new distribution. If you
482 have byacc-1.9, the patch won't apply cleanly. Changes to the printf
483 output statements mean the patch won't apply cleanly. Long ago I
484 started to fix F<perly.fixer> to detect this, but I never completed the
487 If C<perly.c> or C<perly.h> changes, make sure you run C<perl vms/vms_yfix.pl>
488 to update the corresponding VMS files. This could be taken care of by
489 the regen_all target in the Unix Makefile. See also
490 L<VMS-specific updates>.
492 Some additional notes from Larry on this:
494 Don't forget to regenerate perly_c.diff.
498 patch perly.c <perly_c.diff
499 # manually apply any failed hunks
500 diff -c2 perly.c.orig perly.c >perly_c.diff
502 One chunk of lines that often fails begins with
506 and ends one line before
508 #define YYERRCODE 256
510 This only happens when you add or remove a token type. I suppose this
511 could be automated, but it doesn't happen very often nowadays.
515 =head2 make regen_all
517 This target takes care of the PERLYVMS, regen_headers, and regen_pods
520 =head2 make regen_headers
522 The F<embed.h>, F<keywords.h>, and F<opcode.h> files are all automatically
523 generated by perl scripts. Since the user isn't guaranteed to have a
524 working perl, we can't require the user to generate them. Hence you have
525 to, if you're making a distribution.
527 I used to include rules like the following in the makefile:
529 # The following three header files are generated automatically
530 # The correct versions should be already supplied with the perl kit,
531 # in case you don't have perl or 'sh' available.
532 # The - is to ignore error return codes in case you have the source
533 # installed read-only or you don't have perl yet.
534 keywords.h: keywords.pl
535 @echo "Don't worry if this fails."
539 However, I got B<lots> of mail consisting of people worrying because the
540 command failed. I eventually decided that I would save myself time
541 and effort by manually running C<make regen_headers> myself rather
542 than answering all the questions and complaints about the failing
545 =head2 make regen_pods
547 Will run `make regen_pods` in the pod directory for indexing.
549 =head2 global.sym, interp.sym and perlio.sym
551 Make sure these files are up-to-date. Read the comments in these
552 files and in perl_exp.SH to see what to do.
554 =head2 Binary compatibility
556 If you do change F<global.sym> or F<interp.sym>, think carefully about
557 what you are doing. To the extent reasonable, we'd like to maintain
558 source and binary compatibility with older releases of perl. That way,
559 extensions built under one version of perl will continue to work with
560 new versions of perl.
562 Of course, some incompatible changes may well be necessary. I'm just
563 suggesting that we not make any such changes without thinking carefully
564 about them first. If possible, we should provide
565 backwards-compatibility stubs. There's a lot of XS code out there.
566 Let's not force people to keep changing it.
570 Be sure to update the F<Changes> file. Try to include both an overall
571 summary as well as detailed descriptions of the changes. Your
572 audience will include other developers and users, so describe
573 user-visible changes (if any) in terms they will understand, not in
574 code like "initialize foo variable in bar function".
576 There are differing opinions on whether the detailed descriptions
577 ought to go in the Changes file or whether they ought to be available
578 separately in the patch file (or both). There is no disagreement that
579 detailed descriptions ought to be easily available somewhere.
583 The F<Todo> file contains a roughly-catgorized unordered list of
584 aspects of Perl that could use enhancement, features that could be
585 added, areas that could be cleaned up, and so on. During your term as
586 pumpkin-holder, you will probably address some of these issues, and
587 perhaps identify others which, while you decide not to address them
588 this time around, may be tackled in the future. Update the file
589 reflect the situation as it stands when you hand over the pumpkin.
591 You might like, early in your pumpkin-holding career, to see if you
592 can find champions for partiticular issues on the to-do list: an issue
593 owned is an issue more likely to be resolved.
595 There are also some more porting-specific L<Todo> items later in this
598 =head2 OS/2-specific updates
600 In the os2 directory is F<diff.configure>, a set of OS/2-specific
601 diffs against B<Configure>. If you make changes to Configure, you may
602 want to consider regenerating this diff file to save trouble for the
605 You can also consider the OS/2 diffs as reminders of portability
606 things that need to be fixed in Configure.
608 =head2 VMS-specific updates
610 If you have changed F<perly.y> or F<perly.c>, then you most probably want
611 to update F<vms/perly_{h,c}.vms> by running C<perl vms/vms_yfix.pl>, or
612 by running `make regen_all` which will run that script for you.
614 The Perl revision number appears as "perl5" in configure.com.
615 It is courteous to update that if necessary.
617 =head2 Making the new distribution
619 Suppose, for example, that you want to make version 5.004_08. Then you can
620 do something like the following
622 mkdir ../perl5.004_08
623 awk '{print $1}' MANIFEST | cpio -pdm ../perl5.004_08
625 tar cf perl5.004_08.tar perl5.004_08
626 gzip --best perl5.004_08.tar
628 These steps, with extra checks, are automated by the Porting/makerel
631 =head2 Making a new patch
633 I find the F<makepatch> utility quite handy for making patches.
634 You can obtain it from any CPAN archive under
635 http://www.perl.com/CPAN/authors/Johan_Vromans/ . There are a couple
636 of differences between my version and the standard one. I have mine do
639 # Print a reassuring "End of Patch" note so people won't
640 # wonder if their mailer truncated patches.
641 print "\n\nEnd of Patch.\n";
643 at the end. That's because I used to get questions from people asking
644 if their mail was truncated.
646 It also writes Index: lines which include the new directory prefix
647 (change Index: print, approx line 294 or 310 depending on the version,
648 to read: print PATCH ("Index: $newdir$new\n");). That helps patches
649 work with more POSIX conformant patch programs.
651 Here's how I generate a new patch. I'll use the hypothetical
652 5.004_07 to 5.004_08 patch as an example.
654 # unpack perl5.004_07/
655 gzip -d -c perl5.004_07.tar.gz | tar -xof -
656 # unpack perl5.004_08/
657 gzip -d -c perl5.004_08.tar.gz | tar -xof -
658 makepatch perl5.004_07 perl5.004_08 > perl5.004_08.pat
660 Makepatch will automatically generate appropriate B<rm> commands to remove
661 deleted files. Unfortunately, it will not correctly set permissions
662 for newly created files, so you may have to do so manually. For example,
663 patch 5.003_04 created a new test F<t/op/gv.t> which needs to be executable,
664 so at the top of the patch, I inserted the following lines:
670 Now, of course, my patch is now wrong because makepatch didn't know I
671 was going to do that command, and it patched against /dev/null.
673 So, what I do is sort out all such shell commands that need to be in the
674 patch (including possible mv-ing of files, if needed) and put that in the
675 shell commands at the top of the patch. Next, I delete all the patch parts
676 of perl5.004_08.pat, leaving just the shell commands. Then, I do the
680 sh ../perl5.004_08.pat
682 makepatch perl5.004_07 perl5.004_08 >> perl5.004_08.pat
684 (Note the append to preserve my shell commands.)
685 Now, my patch will line up with what the end users are going to do.
687 =head2 Testing your patch
689 It seems obvious, but be sure to test your patch. That is, verify that
690 it produces exactly the same thing as your full distribution.
693 gzip -d -c perl5.004_07.tar.gz | tar -xf -
695 sh ../perl5.004_08.pat
696 patch -p1 -N < ../perl5.004_08.pat
698 gdiff -r perl5.004_07 perl5.004_08
700 where B<gdiff> is GNU diff. Other diff's may also do recursive checking.
704 Again, it's obvious, but you should test your new version as widely as you
705 can. You can be sure you'll hear about it quickly if your version doesn't
706 work on both ANSI and pre-ANSI compilers, and on common systems such as
707 SunOS 4.1.[34], Solaris, and Linux.
709 If your changes include conditional code, try to test the different
710 branches as thoroughly as you can. For example, if your system
711 supports dynamic loading, you can also test static loading with
715 You can also hand-tweak your config.h to try out different #ifdef
724 To test the correct use of printf-style arguments, C<Configure> with
725 S<-Dccflags='-DCHECK_FORMAT -Wformat'> and run C<make>. The compiler
726 will produce warning of incorrect use of format arguments. CHECK_FORMAT
727 changes perl-defined formats to common formats, so DO NOT USE the executable
728 produced by this process.
730 A more accurate approach is the following commands:
732 sh Configure -des -Dccflags=-Wformat ...
733 make miniperl # without -DCHECK_FORMAT
734 perl -i.orig -pwe 's/-Wformat/-DCHECK_FORMAT $&/' config.sh
736 make >& make.log # build from correct miniperl
738 make miniperl >& mini.log # build miniperl with -DCHECK_FORMAT
739 perl -nwe 'print if /^\S+:/ and not /^make\b/' mini.log make.log
742 (-Wformat support by Robin Barker.)
746 =head1 Running Purify
748 Purify is a commercial tool that is helpful in identifying memory
749 overruns, wild pointers, memory leaks and other such badness. Perl
750 must be compiled in a specific way for optimal testing with Purify.
752 Use the following commands to test perl with Purify:
754 sh Configure -des -Doptimize=-g -Uusemymalloc -Dusemultiplicity \
756 setenv PURIFYOPTIONS "-chain-length=25"
759 ln -s ../pureperl perl
760 setenv PERL_DESTRUCT_LEVEL 5
763 Disabling Perl's malloc allows Purify to monitor allocations and leaks
764 more closely; using Perl's malloc will make Purify report most leaks
765 in the "potential" leaks category. Enabling the multiplicity option
766 allows perl to clean up thoroughly when the interpreter shuts down, which
767 reduces the number of bogus leak reports from Purify. The -DPURIFY
768 enables any Purify-specific debugging code in the sources.
770 Purify outputs messages in "Viewer" windows by default. If you don't have
771 a windowing environment or if you simply want the Purify output to
772 unobtrusively go to a log file instead of to the interactive window,
773 use the following options instead:
775 setenv PURIFYOPTIONS "-chain-length=25 -windows=no -log-file=perl.log \
778 The only currently known leaks happen when there are compile-time errors
779 within eval or require. (Fixing these is non-trivial, unfortunately, but
780 they must be fixed eventually.)
782 =head1 Common Gotcha's
788 The '#elif' preprocessor directive is not understood on all systems.
789 Specifically, I know that Pyramids don't understand it. Thus instead of the
800 You have to do the more Byzantine
812 Incidentally, whitespace between the leading '#' and the preprocessor
813 command is not guaranteed, but is very portable and you may use it freely.
814 I think it makes things a bit more readable, especially once things get
815 rather deeply nested. I also think that things should almost never get
816 too deeply nested, so it ought to be a moot point :-)
818 =item Probably Prefer POSIX
820 It's often the case that you'll need to choose whether to do
821 something the BSD-ish way or the POSIX-ish way. It's usually not
822 a big problem when the two systems use different names for similar
823 functions, such as memcmp() and bcmp(). The perl.h header file
824 handles these by appropriate #defines, selecting the POSIX mem*()
825 functions if available, but falling back on the b*() functions, if
828 More serious is the case where some brilliant person decided to
829 use the same function name but give it a different meaning or
830 calling sequence :-). getpgrp() and setpgrp() come to mind.
831 These are a real problem on systems that aim for conformance to
832 one standard (e.g. POSIX), but still try to support the other way
833 of doing things (e.g. BSD). My general advice (still not really
834 implemented in the source) is to do something like the following.
835 Suppose there are two alternative versions, fooPOSIX() and
839 /* use fooPOSIX(); */
842 /* try to emulate fooPOSIX() with fooBSD();
843 perhaps with the following: */
844 # define fooPOSIX fooBSD
846 # /* Uh, oh. We have to supply our own. */
847 # define fooPOSIX Perl_fooPOSIX
851 =item Think positively
853 If you need to add an #ifdef test, it is usually easier to follow if you
854 think positively, e.g.
856 #ifdef HAS_NEATO_FEATURE
857 /* use neato feature */
859 /* use some fallback mechanism */
862 rather than the more impenetrable
864 #ifndef MISSING_NEATO_FEATURE
865 /* Not missing it, so we must have it, so use it */
867 /* Are missing it, so fall back on something else. */
870 Of course for this toy example, there's not much difference. But when
871 the #ifdef's start spanning a couple of screen fulls, and the #else's
872 are marked something like
874 #else /* !MISSING_NEATO_FEATURE */
876 I find it easy to get lost.
878 =item Providing Missing Functions -- Problem
880 Not all systems have all the neat functions you might want or need, so
881 you might decide to be helpful and provide an emulation. This is
882 sound in theory and very kind of you, but please be careful about what
883 you name the function. Let me use the C<pause()> function as an
886 Perl5.003 has the following in F<perl.h>
889 #define pause() sleep((32767<<16)+32767)
892 Configure sets HAS_PAUSE if the system has the pause() function, so
893 this #define only kicks in if the pause() function is missing.
896 Unfortunately, some systems apparently have a prototype for pause()
897 in F<unistd.h>, but don't actually have the function in the library.
898 (Or maybe they do have it in a library we're not using.)
900 Thus, the compiler sees something like
902 extern int pause(void);
904 #define pause() sleep((32767<<16)+32767)
906 and dies with an error message. (Some compilers don't mind this;
907 others apparently do.)
909 To work around this, 5.003_03 and later have the following in perl.h:
911 /* Some unistd.h's give a prototype for pause() even though
912 HAS_PAUSE ends up undefined. This causes the #define
913 below to be rejected by the compiler. Sigh.
918 # define Pause() sleep((32767<<16)+32767)
923 The curious reader may wonder why I didn't do the following in
929 sleep((32767<<16)+32767);
933 That is, since the function is missing, just provide it.
934 Then things would probably be been alright, it would seem.
936 Well, almost. It could be made to work. The problem arises from the
937 conflicting needs of dynamic loading and namespace protection.
939 For dynamic loading to work on AIX (and VMS) we need to provide a list
940 of symbols to be exported. This is done by the script F<perl_exp.SH>,
941 which reads F<global.sym> and F<interp.sym>. Thus, the C<pause>
942 symbol would have to be added to F<global.sym> So far, so good.
944 On the other hand, one of the goals of Perl5 is to make it easy to
945 either extend or embed perl and link it with other libraries. This
946 means we have to be careful to keep the visible namespace "clean".
947 That is, we don't want perl's global variables to conflict with
948 those in the other application library. Although this work is still
949 in progress, the way it is currently done is via the F<embed.h> file.
950 This file is built from the F<global.sym> and F<interp.sym> files,
951 since those files already list the globally visible symbols. If we
952 had added C<pause> to global.sym, then F<embed.h> would contain the
955 #define pause Perl_pause
957 and calls to C<pause> in the perl sources would now point to
958 C<Perl_pause>. Now, when B<ld> is run to build the F<perl> executable,
959 it will go looking for C<perl_pause>, which probably won't exist in any
960 of the standard libraries. Thus the build of perl will fail.
962 Those systems where C<HAS_PAUSE> is not defined would be ok, however,
963 since they would get a C<Perl_pause> function in util.c. The rest of
964 the world would be in trouble.
966 And yes, this scenario has happened. On SCO, the function C<chsize>
967 is available. (I think it's in F<-lx>, the Xenix compatibility
968 library.) Since the perl4 days (and possibly before), Perl has
969 included a C<chsize> function that gets called something akin to
972 I32 chsize(fd, length)
978 #define chsize Perl_chsize
980 to F<embed.h>, the compile started failing on SCO systems.
982 The "fix" is to give the function a different name. The one
983 implemented in 5.003_05 isn't optimal, but here's what was done:
986 # ifdef my_chsize /* Probably #defined to Perl_my_chsize in embed.h */
989 # define my_chsize chsize
992 My explanatory comment in patch 5.003_05 said:
994 Undef and then re-define my_chsize from Perl_my_chsize to
995 just plain chsize if this system HAS_CHSIZE. This probably only
996 applies to SCO. This shows the perils of having internal
997 functions with the same name as external library functions :-).
999 Now, we can safely put C<my_chsize> in F<global.sym>, export it, and
1000 hide it with F<embed.h>.
1002 To be consistent with what I did for C<pause>, I probably should have
1003 called the new function C<Chsize>, rather than C<my_chsize>.
1004 However, the perl sources are quite inconsistent on this (Consider
1005 New, Mymalloc, and Myremalloc, to name just a few.)
1007 There is a problem with this fix, however, in that C<Perl_chsize>
1008 was available as a F<libperl.a> library function in 5.003, but it
1009 isn't available any more (as of 5.003_07). This means that we've
1010 broken binary compatibility. This is not good.
1012 =item Providing missing functions -- some ideas
1014 We currently don't have a standard way of handling such missing
1015 function names. Right now, I'm effectively thinking aloud about a
1016 solution. Some day, I'll try to formally propose a solution.
1018 Part of the problem is that we want to have some functions listed as
1019 exported but not have their names mangled by embed.h or possibly
1020 conflict with names in standard system headers. We actually already
1021 have such a list at the end of F<perl_exp.SH> (though that list is
1024 # extra globals not included above.
1025 cat <<END >> perl.exp
1049 This still needs much thought, but I'm inclined to think that one
1050 possible solution is to prefix all such functions with C<perl_> in the
1051 source and list them along with the other C<perl_*> functions in
1054 Thus, for C<chsize>, we'd do something like the following:
1058 # define perl_chsize chsize
1061 then in some file (e.g. F<util.c> or F<doio.c>) do
1064 I32 perl_chsize(fd, length)
1065 /* implement the function here . . . */
1068 Alternatively, we could just always use C<chsize> everywhere and move
1069 C<chsize> from F<global.sym> to the end of F<perl_exp.SH>. That would
1070 probably be fine as long as our C<chsize> function agreed with all the
1071 C<chsize> function prototypes in the various systems we'll be using.
1072 As long as the prototypes in actual use don't vary that much, this is
1073 probably a good alternative. (As a counter-example, note how Configure
1074 and perl have to go through hoops to find and use get Malloc_t and
1075 Free_t for C<malloc> and C<free>.)
1077 At the moment, this latter option is what I tend to prefer.
1079 =item All the world's a VAX
1081 Sorry, showing my age:-). Still, all the world is not BSD 4.[34],
1082 SVR4, or POSIX. Be aware that SVR3-derived systems are still quite
1083 common (do you have any idea how many systems run SCO?) If you don't
1084 have a bunch of v7 manuals handy, the metaconfig units (by default
1085 installed in F</usr/local/lib/dist/U>) are a good resource to look at
1090 =head1 Miscellaneous Topics
1094 Why does perl use a metaconfig-generated Configure script instead of an
1095 autoconf-generated configure script?
1097 Metaconfig and autoconf are two tools with very similar purposes.
1098 Metaconfig is actually the older of the two, and was originally written
1099 by Larry Wall, while autoconf is probably now used in a wider variety of
1100 packages. The autoconf info file discusses the history of autoconf and
1101 how it came to be. The curious reader is referred there for further
1104 Overall, both tools are quite good, I think, and the choice of which one
1105 to use could be argued either way. In March, 1994, when I was just
1106 starting to work on Configure support for Perl5, I considered both
1107 autoconf and metaconfig, and eventually decided to use metaconfig for the
1112 =item Compatibility with Perl4
1114 Perl4 used metaconfig, so many of the #ifdef's were already set up for
1115 metaconfig. Of course metaconfig had evolved some since Perl4's days,
1116 but not so much that it posed any serious problems.
1118 =item Metaconfig worked for me
1120 My system at the time was Interactive 2.2, a SVR3.2/386 derivative that
1121 also had some POSIX support. Metaconfig-generated Configure scripts
1122 worked fine for me on that system. On the other hand, autoconf-generated
1123 scripts usually didn't. (They did come quite close, though, in some
1124 cases.) At the time, I actually fetched a large number of GNU packages
1125 and checked. Not a single one configured and compiled correctly
1126 out-of-the-box with the system's cc compiler.
1128 =item Configure can be interactive
1130 With both autoconf and metaconfig, if the script works, everything is
1131 fine. However, one of my main problems with autoconf-generated scripts
1132 was that if it guessed wrong about something, it could be B<very> hard to
1133 go back and fix it. For example, autoconf always insisted on passing the
1134 -Xp flag to cc (to turn on POSIX behavior), even when that wasn't what I
1135 wanted or needed for that package. There was no way short of editing the
1136 configure script to turn this off. You couldn't just edit the resulting
1137 Makefile at the end because the -Xp flag influenced a number of other
1140 Metaconfig's Configure scripts, on the other hand, can be interactive.
1141 Thus if Configure is guessing things incorrectly, you can go back and fix
1142 them. This isn't as important now as it was when we were actively
1143 developing Configure support for new features such as dynamic loading,
1144 but it's still useful occasionally.
1148 At the time, autoconf-generated scripts were covered under the GNU Public
1149 License, and hence weren't suitable for inclusion with Perl, which has a
1150 different licensing policy. (Autoconf's licensing has since changed.)
1154 Metaconfig builds up Configure from a collection of discrete pieces
1155 called "units". You can override the standard behavior by supplying your
1156 own unit. With autoconf, you have to patch the standard files instead.
1157 I find the metaconfig "unit" method easier to work with. Others
1158 may find metaconfig's units clumsy to work with.
1162 =head2 Why isn't there a directory to override Perl's library?
1164 Mainly because no one's gotten around to making one. Note that
1165 "making one" involves changing perl.c, Configure, config_h.SH (and
1166 associated files, see above), and I<documenting> it all in the
1169 Apparently, most folks who want to override one of the standard library
1170 files simply do it by overwriting the standard library files.
1174 In the perl.c sources, you'll find an undocumented APPLLIB_EXP
1175 variable, sort of like PRIVLIB_EXP and ARCHLIB_EXP (which are
1176 documented in config_h.SH). Here's what APPLLIB_EXP is for, from
1177 a mail message from Larry:
1179 The main intent of APPLLIB_EXP is for folks who want to send out a
1180 version of Perl embedded in their product. They would set the symbol
1181 to be the name of the library containing the files needed to run or to
1182 support their particular application. This works at the "override"
1183 level to make sure they get their own versions of any library code that
1184 they absolutely must have configuration control over.
1186 As such, I don't see any conflict with a sysadmin using it for a
1187 override-ish sort of thing, when installing a generic Perl. It should
1188 probably have been named something to do with overriding though. Since
1189 it's undocumented we could still change it... :-)
1191 Given that it's already there, you can use it to override
1192 distribution modules. If you do
1194 sh Configure -Dccflags='-DAPPLLIB_EXP=/my/override'
1196 then perl.c will put /my/override ahead of ARCHLIB and PRIVLIB.
1198 =head2 Shared libperl.so location
1200 Why isn't the shared libperl.so installed in /usr/lib/ along
1201 with "all the other" shared libraries? Instead, it is installed
1202 in $archlib, which is typically something like
1204 /usr/local/lib/perl5/archname/5.00404
1206 and is architecture- and version-specific.
1208 The basic reason why a shared libperl.so gets put in $archlib is so that
1209 you can have more than one version of perl on the system at the same time,
1210 and have each refer to its own libperl.so.
1212 Three examples might help. All of these work now; none would work if you
1213 put libperl.so in /usr/lib.
1219 Suppose you want to have both threaded and non-threaded perl versions
1220 around. Configure will name both perl libraries "libperl.so" (so that
1221 you can link to them with -lperl). The perl binaries tell them apart
1222 by having looking in the appropriate $archlib directories.
1226 Suppose you have perl5.004_04 installed and you want to try to compile
1227 it again, perhaps with different options or after applying a patch.
1228 If you already have libperl.so installed in /usr/lib/, then it may be
1229 either difficult or impossible to get ld.so to find the new libperl.so
1230 that you're trying to build. If, instead, libperl.so is tucked away in
1231 $archlib, then you can always just change $archlib in the current perl
1232 you're trying to build so that ld.so won't find your old libperl.so.
1233 (The INSTALL file suggests you do this when building a debugging perl.)
1237 The shared perl library is not a "well-behaved" shared library with
1238 proper major and minor version numbers, so you can't necessarily
1239 have perl5.004_04 and perl5.004_05 installed simultaneously. Suppose
1240 perl5.004_04 were to install /usr/lib/libperl.so.4.4, and perl5.004_05
1241 were to install /usr/lib/libperl.so.4.5. Now, when you try to run
1242 perl5.004_04, ld.so might try to load libperl.so.4.5, since it has
1243 the right "major version" number. If this works at all, it almost
1244 certainly defeats the reason for keeping perl5.004_04 around. Worse,
1245 with development subversions, you certaily can't guarantee that
1246 libperl.so.4.4 and libperl.so.4.55 will be compatible.
1248 Anyway, all this leads to quite obscure failures that are sure to drive
1249 casual users crazy. Even experienced users will get confused :-). Upon
1250 reflection, I'd say leave libperl.so in $archlib.
1254 =head1 Upload Your Work to CPAN
1256 You can upload your work to CPAN if you have a CPAN id. Check out
1257 http://www.perl.com/CPAN/modules/04pause.html for information on
1258 _PAUSE_, the Perl Author's Upload Server.
1260 I typically upload both the patch file, e.g. F<perl5.004_08.pat.gz>
1261 and the full tar file, e.g. F<perl5.004_08.tar.gz>.
1263 If you want your patch to appear in the F<src/5.0/unsupported>
1264 directory on CPAN, send e-mail to the CPAN master librarian. (Check
1265 out http://www.perl.com/CPAN/CPAN.html ).
1267 =head1 Help Save the World
1269 You should definitely announce your patch on the perl5-porters list.
1270 You should also consider announcing your patch on
1271 comp.lang.perl.announce, though you should make it quite clear that a
1272 subversion is not a production release, and be prepared to deal with
1273 people who will not read your disclaimer.
1277 Here, in no particular order, are some Configure and build-related
1278 items that merit consideration. This list isn't exhaustive, it's just
1279 what I came up with off the top of my head.
1281 =head2 Good ideas waiting for round tuits
1285 =item Configure -Dsrc=/blah/blah
1287 We should be able to emulate B<configure --srcdir>. Tom Tromey
1288 tromey@creche.cygnus.com has submitted some patches to
1289 the dist-users mailing list along these lines. They have been folded
1290 back into the main distribution, but various parts of the perl
1291 Configure/build/install process still assume src='.'.
1293 =item Hint file fixes
1295 Various hint files work around Configure problems. We ought to fix
1296 Configure so that most of them aren't needed.
1298 =item Hint file information
1300 Some of the hint file information (particularly dynamic loading stuff)
1301 ought to be fed back into the main metaconfig distribution.
1305 =head2 Probably good ideas waiting for round tuits
1309 =item GNU configure --options
1311 I've received sensible suggestions for --exec_prefix and other
1312 GNU configure --options. It's not always obvious exactly what is
1313 intended, but this merits investigation.
1317 Currently, B<make clean> isn't all that useful, though
1318 B<make realclean> and B<make distclean> are. This needs a bit of
1319 thought and documentation before it gets cleaned up.
1321 =item Try gcc if cc fails
1323 Currently, we just give up.
1325 =item bypassing safe*alloc wrappers
1327 On some systems, it may be safe to call the system malloc directly
1328 without going through the util.c safe* layers. (Such systems would
1329 accept free(0), for example.) This might be a time-saver for systems
1330 that already have a good malloc. (Recent Linux libc's apparently have
1331 a nice malloc that is well-tuned for the system.)
1335 =head2 Vague possibilities
1341 Get some of the Macintosh stuff folded back into the main distribution.
1343 =item gconvert replacement
1345 Maybe include a replacement function that doesn't lose data in rare
1346 cases of coercion between string and numerical values.
1348 =item Improve makedepend
1350 The current makedepend process is clunky and annoyingly slow, but it
1351 works for most folks. Alas, it assumes that there is a filename
1352 $firstmakefile that the B<make> command will try to use before it uses
1353 F<Makefile>. Such may not be the case for all B<make> commands,
1354 particularly those on non-Unix systems.
1356 Probably some variant of the BSD F<.depend> file will be useful.
1357 We ought to check how other packages do this, if they do it at all.
1358 We could probably pre-generate the dependencies (with the exception of
1359 malloc.o, which could probably be determined at F<Makefile.SH>
1362 =item GNU Makefile standard targets
1364 GNU software generally has standardized Makefile targets. Unless we
1365 have good reason to do otherwise, I see no reason not to support them.
1369 Somehow, straighten out, document, and implement lockf(), flock(),
1370 and/or fcntl() file locking. It's a mess. See $d_fcntl_can_lock
1371 in recent config.sh files though.
1377 Original author: Andy Dougherty doughera@lafcol.lafayette.edu .
1378 Additions by Chip Salzenberg chip@perl.com and
1379 Tim Bunce Tim.Bunce@ig.co.uk .
1381 All opinions expressed herein are those of the authorZ<>(s).
1383 =head1 LAST MODIFIED
1385 $Id: pumpkin.pod,v 1.23 2000/01/13 19:45:13 doughera Released $