Commit | Line | Data |
3967c732 |
1 | # Devel::Peek - A data debugging tool for the XS programmer |
2 | # The documentation is after the __END__ |
3 | |
4 | package Devel::Peek; |
5 | |
b162af07 |
6 | $VERSION = '1.03'; |
105cd853 |
7 | $XS_VERSION = $VERSION; |
8 | $VERSION = eval $VERSION; |
3967c732 |
9 | |
10 | require Exporter; |
9426adcd |
11 | use XSLoader (); |
3967c732 |
12 | |
9426adcd |
13 | @ISA = qw(Exporter); |
d1424c31 |
14 | @EXPORT = qw(Dump mstat DeadCode DumpArray DumpWithOP DumpProg |
bd16a5f0 |
15 | fill_mstats mstats_fillhash mstats2hash runops_debug debug_flags); |
83ee9e09 |
16 | @EXPORT_OK = qw(SvREFCNT SvREFCNT_inc SvREFCNT_dec CvGV); |
3967c732 |
17 | %EXPORT_TAGS = ('ALL' => [@EXPORT, @EXPORT_OK]); |
18 | |
9426adcd |
19 | XSLoader::load 'Devel::Peek'; |
3967c732 |
20 | |
1045810a |
21 | sub import { |
22 | my $c = shift; |
23 | my $ops_rx = qr/^:opd(=[stP]*)?\b/; |
24 | my @db = grep m/$ops_rx/, @_; |
25 | @_ = grep !m/$ops_rx/, @_; |
26 | if (@db) { |
27 | die "Too many :opd options" if @db > 1; |
28 | runops_debug(1); |
29 | my $flags = ($db[0] =~ m/$ops_rx/ and $1); |
30 | $flags = 'st' unless defined $flags; |
31 | my $f = 0; |
32 | $f |= 2 if $flags =~ /s/; |
33 | $f |= 8 if $flags =~ /t/; |
34 | $f |= 64 if $flags =~ /P/; |
35 | $^D |= $f if $f; |
36 | } |
37 | unshift @_, $c; |
38 | goto &Exporter::import; |
39 | } |
40 | |
3967c732 |
41 | sub DumpWithOP ($;$) { |
42 | local($Devel::Peek::dump_ops)=1; |
43 | my $depth = @_ > 1 ? $_[1] : 4 ; |
44 | Dump($_[0],$depth); |
45 | } |
46 | |
bd16a5f0 |
47 | $D_flags = 'psltocPmfrxuLHXDSTR'; |
48 | |
49 | sub debug_flags (;$) { |
50 | my $out = ""; |
51 | for my $i (0 .. length($D_flags)-1) { |
52 | $out .= substr $D_flags, $i, 1 if $^D & (1<<$i); |
53 | } |
54 | my $arg = shift; |
55 | my $num = $arg; |
56 | if (defined $arg and $arg =~ /\D/) { |
57 | die "unknown flags in debug_flags()" if $arg =~ /[^-$D_flags]/; |
58 | my ($on,$off) = split /-/, "$arg-"; |
59 | $num = $^D; |
60 | $num |= (1<<index($D_flags, $_)) for split //, $on; |
61 | $num &= ~(1<<index($D_flags, $_)) for split //, $off; |
62 | } |
63 | $^D = $num if defined $arg; |
64 | $out |
65 | } |
66 | |
3967c732 |
67 | 1; |
68 | __END__ |
69 | |
70 | =head1 NAME |
71 | |
72 | Devel::Peek - A data debugging tool for the XS programmer |
73 | |
74 | =head1 SYNOPSIS |
75 | |
76 | use Devel::Peek; |
77 | Dump( $a ); |
78 | Dump( $a, 5 ); |
79 | DumpArray( 5, $a, $b, ... ); |
80 | mstat "Point 5"; |
81 | |
1045810a |
82 | use Devel::Peek ':opd=st'; |
83 | |
3967c732 |
84 | =head1 DESCRIPTION |
85 | |
86 | Devel::Peek contains functions which allows raw Perl datatypes to be |
87 | manipulated from a Perl script. This is used by those who do XS programming |
88 | to check that the data they are sending from C to Perl looks as they think |
89 | it should look. The trick, then, is to know what the raw datatype is |
90 | supposed to look like when it gets to Perl. This document offers some tips |
91 | and hints to describe good and bad raw data. |
92 | |
93 | It is very possible that this document will fall far short of being useful |
94 | to the casual reader. The reader is expected to understand the material in |
95 | the first few sections of L<perlguts>. |
96 | |
97 | Devel::Peek supplies a C<Dump()> function which can dump a raw Perl |
98 | datatype, and C<mstat("marker")> function to report on memory usage |
99 | (if perl is compiled with corresponding option). The function |
100 | DeadCode() provides statistics on the data "frozen" into inactive |
101 | C<CV>. Devel::Peek also supplies C<SvREFCNT()>, C<SvREFCNT_inc()>, and |
102 | C<SvREFCNT_dec()> which can query, increment, and decrement reference |
103 | counts on SVs. This document will take a passive, and safe, approach |
104 | to data debugging and for that it will describe only the C<Dump()> |
d1424c31 |
105 | function. |
3967c732 |
106 | |
107 | Function C<DumpArray()> allows dumping of multiple values (useful when you |
076c2fc0 |
108 | need to analyze returns of functions). |
3967c732 |
109 | |
110 | The global variable $Devel::Peek::pv_limit can be set to limit the |
111 | number of character printed in various string values. Setting it to 0 |
112 | means no limit. |
113 | |
1045810a |
114 | If C<use Devel::Peek> directive has a C<:opd=FLAGS> argument, |
115 | this switches on debugging of opcode dispatch. C<FLAGS> should be a |
116 | combination of C<s>, C<t>, and C<P> (see B<-D> flags in L<perlrun>). |
117 | C<:opd> is a shortcut for C<:opd=st>. |
118 | |
bd16a5f0 |
119 | =head2 Runtime debugging |
120 | |
121 | C<CvGV($cv)> return one of the globs associated to a subroutine reference $cv. |
122 | |
123 | debug_flags() returns a string representation of C<$^D> (similar to |
124 | what is allowed for B<-D> flag). When called with a numeric argument, |
125 | sets $^D to the corresponding value. When called with an argument of |
126 | the form C<"flags-flags">, set on/off bits of C<$^D> corresponding to |
127 | letters before/after C<->. (The returned value is for C<$^D> before |
128 | the modification.) |
129 | |
130 | runops_debug() returns true if the current I<opcode dispatcher> is the |
131 | debugging one. When called with an argument, switches to debugging or |
132 | non-debugging dispatcher depending on the argument (active for |
133 | newly-entered subs/etc only). (The returned value is for the dispatcher before the modification.) |
134 | |
d1424c31 |
135 | =head2 Memory footprint debugging |
136 | |
137 | When perl is compiled with support for memory footprint debugging |
138 | (default with Perl's malloc()), Devel::Peek provides an access to this API. |
139 | |
140 | Use mstat() function to emit a memory state statistic to the terminal. |
141 | For more information on the format of output of mstat() see |
f3487f28 |
142 | L<perldebguts/Using C<$ENV{PERL_DEBUG_MSTATS}>>. |
d1424c31 |
143 | |
144 | Three additional functions allow access to this statistic from Perl. |
145 | First, use C<mstats_fillhash(%hash)> to get the information contained |
146 | in the output of mstat() into %hash. The field of this hash are |
147 | |
148 | minbucket nbuckets sbrk_good sbrk_slack sbrked_remains sbrks start_slack |
149 | topbucket topbucket_ev topbucket_odd total total_chain total_sbrk totfree |
150 | |
151 | Two additional fields C<free>, C<used> contain array references which |
152 | provide per-bucket count of free and used chunks. Two other fields |
153 | C<mem_size>, C<available_size> contain array references which provide |
154 | the information about the allocated size and usable size of chunks in |
f3487f28 |
155 | each bucket. Again, see L<perldebguts/Using C<$ENV{PERL_DEBUG_MSTATS}>> |
d1424c31 |
156 | for details. |
157 | |
158 | Keep in mind that only the first several "odd-numbered" buckets are |
159 | used, so the information on size of the "odd-numbered" buckets which are |
160 | not used is probably meaningless. |
161 | |
162 | The information in |
163 | |
164 | mem_size available_size minbucket nbuckets |
165 | |
166 | is the property of a particular build of perl, and does not depend on |
167 | the current process. If you do not provide the optional argument to |
168 | the functions mstats_fillhash(), fill_mstats(), mstats2hash(), then |
169 | the information in fields C<mem_size>, C<available_size> is not |
170 | updated. |
171 | |
172 | C<fill_mstats($buf)> is a much cheaper call (both speedwise and |
173 | memory-wise) which collects the statistic into $buf in |
174 | machine-readable form. At a later moment you may need to call |
175 | C<mstats2hash($buf, %hash)> to use this information to fill %hash. |
176 | |
177 | All three APIs C<fill_mstats($buf)>, C<mstats_fillhash(%hash)>, and |
178 | C<mstats2hash($buf, %hash)> are designed to allocate no memory if used |
179 | I<the second time> on the same $buf and/or %hash. |
180 | |
181 | So, if you want to collect memory info in a cycle, you may call |
182 | |
183 | $#buf = 999; |
184 | fill_mstats($_) for @buf; |
185 | mstats_fillhash(%report, 1); # Static info too |
186 | |
187 | foreach (@buf) { |
188 | # Do something... |
189 | fill_mstats $_; # Collect statistic |
190 | } |
191 | foreach (@buf) { |
192 | mstats2hash($_, %report); # Preserve static info |
193 | # Do something with %report |
194 | } |
195 | |
3967c732 |
196 | =head1 EXAMPLES |
197 | |
198 | The following examples don't attempt to show everything as that would be a |
199 | monumental task, and, frankly, we don't want this manpage to be an internals |
200 | document for Perl. The examples do demonstrate some basics of the raw Perl |
201 | datatypes, and should suffice to get most determined people on their way. |
202 | There are no guidewires or safety nets, nor blazed trails, so be prepared to |
203 | travel alone from this point and on and, if at all possible, don't fall into |
204 | the quicksand (it's bad for business). |
205 | |
206 | Oh, one final bit of advice: take L<perlguts> with you. When you return we |
207 | expect to see it well-thumbed. |
208 | |
209 | =head2 A simple scalar string |
210 | |
211 | Let's begin by looking a simple scalar which is holding a string. |
212 | |
a423dfdd |
213 | use Devel::Peek; |
3967c732 |
214 | $a = "hello"; |
215 | Dump $a; |
216 | |
217 | The output: |
218 | |
219 | SV = PVIV(0xbc288) |
220 | REFCNT = 1 |
221 | FLAGS = (POK,pPOK) |
222 | IV = 0 |
223 | PV = 0xb2048 "hello"\0 |
224 | CUR = 5 |
225 | LEN = 6 |
226 | |
227 | This says C<$a> is an SV, a scalar. The scalar is a PVIV, a string. |
228 | Its reference count is 1. It has the C<POK> flag set, meaning its |
229 | current PV field is valid. Because POK is set we look at the PV item |
230 | to see what is in the scalar. The \0 at the end indicate that this |
231 | PV is properly NUL-terminated. |
232 | If the FLAGS had been IOK we would look |
233 | at the IV item. CUR indicates the number of characters in the PV. |
234 | LEN indicates the number of bytes requested for the PV (one more than |
235 | CUR, in this case, because LEN includes an extra byte for the |
236 | end-of-string marker). |
237 | |
238 | =head2 A simple scalar number |
239 | |
240 | If the scalar contains a number the raw SV will be leaner. |
241 | |
a423dfdd |
242 | use Devel::Peek; |
3967c732 |
243 | $a = 42; |
244 | Dump $a; |
245 | |
246 | The output: |
247 | |
248 | SV = IV(0xbc818) |
249 | REFCNT = 1 |
250 | FLAGS = (IOK,pIOK) |
251 | IV = 42 |
252 | |
253 | This says C<$a> is an SV, a scalar. The scalar is an IV, a number. Its |
254 | reference count is 1. It has the C<IOK> flag set, meaning it is currently |
255 | being evaluated as a number. Because IOK is set we look at the IV item to |
256 | see what is in the scalar. |
257 | |
258 | =head2 A simple scalar with an extra reference |
259 | |
260 | If the scalar from the previous example had an extra reference: |
261 | |
a423dfdd |
262 | use Devel::Peek; |
3967c732 |
263 | $a = 42; |
264 | $b = \$a; |
265 | Dump $a; |
266 | |
267 | The output: |
268 | |
269 | SV = IV(0xbe860) |
270 | REFCNT = 2 |
271 | FLAGS = (IOK,pIOK) |
272 | IV = 42 |
273 | |
274 | Notice that this example differs from the previous example only in its |
275 | reference count. Compare this to the next example, where we dump C<$b> |
276 | instead of C<$a>. |
277 | |
278 | =head2 A reference to a simple scalar |
279 | |
280 | This shows what a reference looks like when it references a simple scalar. |
281 | |
a423dfdd |
282 | use Devel::Peek; |
3967c732 |
283 | $a = 42; |
284 | $b = \$a; |
285 | Dump $b; |
286 | |
287 | The output: |
288 | |
289 | SV = RV(0xf041c) |
290 | REFCNT = 1 |
291 | FLAGS = (ROK) |
292 | RV = 0xbab08 |
293 | SV = IV(0xbe860) |
294 | REFCNT = 2 |
295 | FLAGS = (IOK,pIOK) |
296 | IV = 42 |
297 | |
298 | Starting from the top, this says C<$b> is an SV. The scalar is an RV, a |
299 | reference. It has the C<ROK> flag set, meaning it is a reference. Because |
300 | ROK is set we have an RV item rather than an IV or PV. Notice that Dump |
301 | follows the reference and shows us what C<$b> was referencing. We see the |
302 | same C<$a> that we found in the previous example. |
303 | |
304 | Note that the value of C<RV> coincides with the numbers we see when we |
305 | stringify $b. The addresses inside RV() and IV() are addresses of |
306 | C<X***> structure which holds the current state of an C<SV>. This |
307 | address may change during lifetime of an SV. |
308 | |
309 | =head2 A reference to an array |
310 | |
311 | This shows what a reference to an array looks like. |
312 | |
a423dfdd |
313 | use Devel::Peek; |
3967c732 |
314 | $a = [42]; |
315 | Dump $a; |
316 | |
317 | The output: |
318 | |
319 | SV = RV(0xf041c) |
320 | REFCNT = 1 |
321 | FLAGS = (ROK) |
322 | RV = 0xb2850 |
323 | SV = PVAV(0xbd448) |
324 | REFCNT = 1 |
325 | FLAGS = () |
326 | IV = 0 |
327 | NV = 0 |
328 | ARRAY = 0xb2048 |
329 | ALLOC = 0xb2048 |
330 | FILL = 0 |
331 | MAX = 0 |
332 | ARYLEN = 0x0 |
333 | FLAGS = (REAL) |
334 | Elt No. 0 0xb5658 |
335 | SV = IV(0xbe860) |
336 | REFCNT = 1 |
337 | FLAGS = (IOK,pIOK) |
338 | IV = 42 |
339 | |
340 | This says C<$a> is an SV and that it is an RV. That RV points to |
341 | another SV which is a PVAV, an array. The array has one element, |
342 | element zero, which is another SV. The field C<FILL> above indicates |
343 | the last element in the array, similar to C<$#$a>. |
344 | |
345 | If C<$a> pointed to an array of two elements then we would see the |
346 | following. |
347 | |
348 | use Devel::Peek 'Dump'; |
349 | $a = [42,24]; |
350 | Dump $a; |
351 | |
352 | The output: |
353 | |
354 | SV = RV(0xf041c) |
355 | REFCNT = 1 |
356 | FLAGS = (ROK) |
357 | RV = 0xb2850 |
358 | SV = PVAV(0xbd448) |
359 | REFCNT = 1 |
360 | FLAGS = () |
361 | IV = 0 |
362 | NV = 0 |
363 | ARRAY = 0xb2048 |
364 | ALLOC = 0xb2048 |
365 | FILL = 0 |
366 | MAX = 0 |
367 | ARYLEN = 0x0 |
368 | FLAGS = (REAL) |
369 | Elt No. 0 0xb5658 |
370 | SV = IV(0xbe860) |
371 | REFCNT = 1 |
372 | FLAGS = (IOK,pIOK) |
373 | IV = 42 |
374 | Elt No. 1 0xb5680 |
375 | SV = IV(0xbe818) |
376 | REFCNT = 1 |
377 | FLAGS = (IOK,pIOK) |
378 | IV = 24 |
379 | |
380 | Note that C<Dump> will not report I<all> the elements in the array, |
381 | only several first (depending on how deep it already went into the |
382 | report tree). |
383 | |
384 | =head2 A reference to a hash |
385 | |
386 | The following shows the raw form of a reference to a hash. |
387 | |
a423dfdd |
388 | use Devel::Peek; |
3967c732 |
389 | $a = {hello=>42}; |
390 | Dump $a; |
391 | |
392 | The output: |
393 | |
d5889220 |
394 | SV = RV(0x8177858) at 0x816a618 |
395 | REFCNT = 1 |
396 | FLAGS = (ROK) |
397 | RV = 0x814fc10 |
398 | SV = PVHV(0x8167768) at 0x814fc10 |
399 | REFCNT = 1 |
400 | FLAGS = (SHAREKEYS) |
401 | IV = 1 |
402 | NV = 0 |
403 | ARRAY = 0x816c5b8 (0:7, 1:1) |
404 | hash quality = 100.0% |
405 | KEYS = 1 |
406 | FILL = 1 |
407 | MAX = 7 |
408 | RITER = -1 |
409 | EITER = 0x0 |
410 | Elt "hello" HASH = 0xc8fd181b |
411 | SV = IV(0x816c030) at 0x814fcf4 |
412 | REFCNT = 1 |
413 | FLAGS = (IOK,pIOK) |
414 | IV = 42 |
3967c732 |
415 | |
416 | This shows C<$a> is a reference pointing to an SV. That SV is a PVHV, a |
417 | hash. Fields RITER and EITER are used by C<L<each>>. |
418 | |
d5889220 |
419 | The "quality" of a hash is defined as the total number of comparisons needed |
420 | to access every element once, relative to the expected number needed for a |
421 | random hash. The value can go over 100%. |
422 | |
423 | The total number of comparisons is equal to the sum of the squares of the |
424 | number of entries in each bucket. For a random hash of C<<n>> keys into |
425 | C<<k>> buckets, the expected value is: |
426 | |
427 | n + n(n-1)/2k |
428 | |
3967c732 |
429 | =head2 Dumping a large array or hash |
430 | |
431 | The C<Dump()> function, by default, dumps up to 4 elements from a |
432 | toplevel array or hash. This number can be increased by supplying a |
433 | second argument to the function. |
434 | |
a423dfdd |
435 | use Devel::Peek; |
3967c732 |
436 | $a = [10,11,12,13,14]; |
437 | Dump $a; |
438 | |
439 | Notice that C<Dump()> prints only elements 10 through 13 in the above code. |
440 | The following code will print all of the elements. |
441 | |
442 | use Devel::Peek 'Dump'; |
443 | $a = [10,11,12,13,14]; |
444 | Dump $a, 5; |
445 | |
446 | =head2 A reference to an SV which holds a C pointer |
447 | |
448 | This is what you really need to know as an XS programmer, of course. When |
449 | an XSUB returns a pointer to a C structure that pointer is stored in an SV |
450 | and a reference to that SV is placed on the XSUB stack. So the output from |
451 | an XSUB which uses something like the T_PTROBJ map might look something like |
452 | this: |
453 | |
454 | SV = RV(0xf381c) |
455 | REFCNT = 1 |
456 | FLAGS = (ROK) |
457 | RV = 0xb8ad8 |
458 | SV = PVMG(0xbb3c8) |
459 | REFCNT = 1 |
460 | FLAGS = (OBJECT,IOK,pIOK) |
461 | IV = 729160 |
462 | NV = 0 |
463 | PV = 0 |
464 | STASH = 0xc1d10 "CookBookB::Opaque" |
465 | |
466 | This shows that we have an SV which is an RV. That RV points at another |
467 | SV. In this case that second SV is a PVMG, a blessed scalar. Because it is |
468 | blessed it has the C<OBJECT> flag set. Note that an SV which holds a C |
469 | pointer also has the C<IOK> flag set. The C<STASH> is set to the package |
470 | name which this SV was blessed into. |
471 | |
472 | The output from an XSUB which uses something like the T_PTRREF map, which |
473 | doesn't bless the object, might look something like this: |
474 | |
475 | SV = RV(0xf381c) |
476 | REFCNT = 1 |
477 | FLAGS = (ROK) |
478 | RV = 0xb8ad8 |
479 | SV = PVMG(0xbb3c8) |
480 | REFCNT = 1 |
481 | FLAGS = (IOK,pIOK) |
482 | IV = 729160 |
483 | NV = 0 |
484 | PV = 0 |
485 | |
486 | =head2 A reference to a subroutine |
487 | |
488 | Looks like this: |
489 | |
490 | SV = RV(0x798ec) |
491 | REFCNT = 1 |
492 | FLAGS = (TEMP,ROK) |
493 | RV = 0x1d453c |
494 | SV = PVCV(0x1c768c) |
495 | REFCNT = 2 |
496 | FLAGS = () |
497 | IV = 0 |
498 | NV = 0 |
499 | COMP_STASH = 0x31068 "main" |
500 | START = 0xb20e0 |
501 | ROOT = 0xbece0 |
502 | XSUB = 0x0 |
503 | XSUBANY = 0 |
504 | GVGV::GV = 0x1d44e8 "MY" :: "top_targets" |
57843af0 |
505 | FILE = "(eval 5)" |
3967c732 |
506 | DEPTH = 0 |
507 | PADLIST = 0x1c9338 |
508 | |
509 | This shows that |
510 | |
bbc7dcd2 |
511 | =over 4 |
3967c732 |
512 | |
a45bd81d |
513 | =item * |
3967c732 |
514 | |
515 | the subroutine is not an XSUB (since C<START> and C<ROOT> are |
516 | non-zero, and C<XSUB> is zero); |
517 | |
a45bd81d |
518 | =item * |
3967c732 |
519 | |
520 | that it was compiled in the package C<main>; |
521 | |
a45bd81d |
522 | =item * |
3967c732 |
523 | |
524 | under the name C<MY::top_targets>; |
525 | |
a45bd81d |
526 | =item * |
3967c732 |
527 | |
528 | inside a 5th eval in the program; |
529 | |
a45bd81d |
530 | =item * |
3967c732 |
531 | |
532 | it is not currently executed (see C<DEPTH>); |
533 | |
a45bd81d |
534 | =item * |
3967c732 |
535 | |
536 | it has no prototype (C<PROTOTYPE> field is missing). |
537 | |
a45bd81d |
538 | =back |
3967c732 |
539 | |
540 | =head1 EXPORTS |
541 | |
542 | C<Dump>, C<mstat>, C<DeadCode>, C<DumpArray>, C<DumpWithOP> and |
7c6ca602 |
543 | C<DumpProg>, C<fill_mstats>, C<mstats_fillhash>, C<mstats2hash> by |
d1424c31 |
544 | default. Additionally available C<SvREFCNT>, C<SvREFCNT_inc> and |
545 | C<SvREFCNT_dec>. |
3967c732 |
546 | |
547 | =head1 BUGS |
548 | |
549 | Readers have been known to skip important parts of L<perlguts>, causing much |
550 | frustration for all. |
551 | |
552 | =head1 AUTHOR |
553 | |
554 | Ilya Zakharevich ilya@math.ohio-state.edu |
555 | |
556 | Copyright (c) 1995-98 Ilya Zakharevich. All rights reserved. |
557 | This program is free software; you can redistribute it and/or |
558 | modify it under the same terms as Perl itself. |
559 | |
560 | Author of this software makes no claim whatsoever about suitability, |
561 | reliability, edability, editability or usability of this product, and |
562 | should not be kept liable for any damage resulting from the use of |
563 | it. If you can use it, you are in luck, if not, I should not be kept |
564 | responsible. Keep a handy copy of your backup tape at hand. |
565 | |
566 | =head1 SEE ALSO |
567 | |
568 | L<perlguts>, and L<perlguts>, again. |
569 | |
570 | =cut |