X-Git-Url: https://perl5.git.perl.org/perl5.git/blobdiff_plain/5b6a3331e95aaaac258e709109a5361f19459b02..c440a570f986f52b764752007c070e8549b2bf7e:/pod/perldebguts.pod diff --git a/pod/perldebguts.pod b/pod/perldebguts.pod index 6bc2c37..2b5561d 100644 --- a/pod/perldebguts.pod +++ b/pod/perldebguts.pod @@ -4,7 +4,7 @@ perldebguts - Guts of Perl debugging =head1 DESCRIPTION -This is not the perldebug(1) manpage, which tells you how to use +This is not L, which tells you how to use the debugger. This manpage describes low-level details concerning the debugger's internals, which range from difficult to impossible to understand for anyone who isn't incredibly intimate with Perl's guts. @@ -38,7 +38,6 @@ Each array C<@{"_<$filename"}> holds the lines of $filename for a file compiled by Perl. The same is also true for Ced strings that contain subroutines, or which are currently being executed. The $filename for Ced strings looks like C<(eval 34)>. -Code assertions in regexes look like C<(re_eval 19)>. Values in this array are magical in numeric context: they compare equal to zero only if the line is not breakable. @@ -53,14 +52,14 @@ C<"$break_condition\0$action">. The same holds for evaluated strings that contain subroutines, or which are currently being executed. The $filename for Ced strings -looks like C<(eval 34)> or C<(re_eval 19)>. +looks like C<(eval 34)>. =item * Each scalar C<${"_<$filename"}> contains C<"_<$filename">. This is also the case for evaluated strings that contain subroutines, or which are currently being executed. The $filename for Ced -strings looks like C<(eval 34)> or C<(re_eval 19)>. +strings looks like C<(eval 34)>. =item * @@ -81,7 +80,7 @@ also exists. A hash C<%DB::sub> is maintained, whose keys are subroutine names and whose values have the form C. C has the form C<(eval 34)> for subroutines defined inside -Cs, or C<(re_eval 19)> for those within regex code assertions. +Cs. =item * @@ -99,6 +98,17 @@ C<&DB::sub>(I) is made instead, with C<$DB::sub> holding the name of the called subroutine. (This doesn't happen if the subroutine was compiled in the C package.) +X<&DB::lsub>If the call is to an lvalue subroutine, and C<&DB::lsub> +is defined C<&DB::lsub>(I) is called instead, otherwise falling +back to C<&DB::sub>(I). + +=item * + +When execution of the program uses C to enter a non-XS +subroutine and the 0x80 bit is set in C<$^P>, a call to C<&DB::goto> +is made, with C<$DB::sub> holding the name of the subroutine being +entered. + =back Note that if C<&DB::sub> needs external data for it to work, no @@ -151,9 +161,10 @@ after the debugger completes its own initialization.) After the rc file is read, the debugger reads the PERLDB_OPTS environment variable and uses it to set debugger options. The contents of this variable are treated as if they were the argument -of an C debugger command (q.v. in L). +of an C debugger command (q.v. in L). + +=head3 Debugger Internal Variables -=head3 Debugger internal variables In addition to the file and subroutine-related variables mentioned above, the debugger also maintains various magical internal variables. @@ -171,7 +182,7 @@ equal to zero only if the line is not breakable. =item * -C<%DB::dbline>, is an alias for C<%{"::_, which +C<%DB::dbline> is an alias for C<%{"::_, which contains breakpoints and actions keyed by line number in the currently-selected file, either explicitly chosen with the debugger's C command, or implicitly by flow of execution. @@ -183,7 +194,7 @@ C<"$break_condition\0$action">. =back -=head3 Debugger customization functions +=head3 Debugger Customization Functions Some functions are provided to simplify customization. @@ -191,9 +202,8 @@ Some functions are provided to simplify customization. =item * -See L for description of options parsed by -C parses debugger options; see -L for a description of options recognized. +See L for a description of options parsed by +C. =item * @@ -227,7 +237,7 @@ information. For example, contrast this expression trace: Loading DB routines from perl5db.pl patch level 0.94 Emacs support available. - Enter h or `h h' for help. + Enter h or 'h h' for help. main::(-e:1): 0 DB<1> sub foo { 14 } @@ -257,8 +267,8 @@ with this one, once the Cption C has been set: By way of demonstration, we present below a laborious listing resulting from setting your C environment variable to the value C, and running I from the command line. -Examples use various values of C are shown to give you a feel -for the difference between settings. Long those it may be, this +Examples using various values of C are shown to give you a feel +for the difference between settings. Long though it may be, this is not a complete listing, but only excerpts. =over 4 @@ -301,7 +311,7 @@ is not a complete listing, but only excerpts. exited Config::FETCH entering Config::FETCH -=item 4 +=item 3 in $=main::BEGIN() from /dev/null:0 in $=Config::BEGIN() from lib/Config.pm:2 @@ -319,7 +329,7 @@ is not a complete listing, but only excerpts. in $=Config::FETCH(ref(Config), 'osname') from lib/Config.pm:574 in $=Config::FETCH(ref(Config), 'osvers') from lib/Config.pm:574 -=item 6 +=item 4 in $=main::BEGIN() from /dev/null:0 in $=Config::BEGIN() from lib/Config.pm:2 @@ -343,7 +353,7 @@ is not a complete listing, but only excerpts. out $=Config::FETCH(ref(Config), 'PERL_VERSION') from lib/Config.pm:574 in $=Config::FETCH(ref(Config), 'PERL_SUBVERSION') from lib/Config.pm:574 -=item 14 +=item 5 in $=main::BEGIN() from /dev/null:0 in $=Config::BEGIN() from lib/Config.pm:2 @@ -364,7 +374,7 @@ is not a complete listing, but only excerpts. in $=Config::FETCH('Config=HASH(0x1aa444)', 'baserev') from lib/Config.pm:574 out $=Config::FETCH('Config=HASH(0x1aa444)', 'baserev') from lib/Config.pm:574 -=item 30 +=item 6 in $=CODE(0x15eca4)() from /dev/null:0 in $=CODE(0x182528)() from lib/Config.pm:2 @@ -397,7 +407,7 @@ When a package is compiled, a line like this is printed with proper indentation. -=head1 Debugging regular expressions +=head1 Debugging Regular Expressions There are two ways to enable debugging output for regular expressions. @@ -405,13 +415,14 @@ If your perl is compiled with C<-DDEBUGGING>, you may use the B<-Dr> flag on the command line. Otherwise, one can C, which has effects at -compile time and run time. It is not lexically scoped. +compile time and run time. Since Perl 5.9.5, this pragma is lexically +scoped. -=head2 Compile-time output +=head2 Compile-time Output The debugging output at compile time looks like this: - Compiling REx `[bc]d(ef*g)+h[ij]k$' + Compiling REx '[bc]d(ef*g)+h[ij]k$' size 45 Got 364 bytes for offset annotations. first at 1 rarest char g at 0 @@ -432,8 +443,8 @@ The debugging output at compile time looks like this: 42: EXACT (44) 44: EOL(45) 45: END(0) - anchored `de' at 1 floating `gh' at 3..2147483647 (checking floating) - stclass `ANYOF[bc]' minlen 7 + anchored 'de' at 1 floating 'gh' at 3..2147483647 (checking floating) + stclass 'ANYOF[bc]' minlen 7 Offsets: [45] 1[4] 0[0] 0[0] 0[0] 0[0] 0[0] 0[0] 0[0] 0[0] 0[0] 0[0] 5[1] 0[0] 12[1] 0[0] 6[1] 0[0] 7[1] 0[0] 9[1] 8[1] 0[0] 10[1] 0[0] @@ -449,8 +460,8 @@ label I of the first node that does a match. The - anchored `de' at 1 floating `gh' at 3..2147483647 (checking floating) - stclass `ANYOF[bc]' minlen 7 + anchored 'de' at 1 floating 'gh' at 3..2147483647 (checking floating) + stclass 'ANYOF[bc]' minlen 7 line (split into two lines above) contains optimizer information. In the example shown, the optimizer found that the match @@ -513,13 +524,13 @@ C<(??{ code })>. =item C -If the pattern may match only at a handful of places, (with C -being C, C, or C. See the table below. +If the pattern may match only at a handful of places, with C +being C, C, or C. See the table below. =back If a substring is known to match at end-of-line only, it may be -followed by C<$>, as in C. +followed by C<$>, as in C. The optimizer-specific information is used to avoid entering (a slow) regex engine on strings that will not definitely match. If the C flag @@ -531,119 +542,254 @@ form of the regex. Each line has format C< >I: I I (I) -=head2 Types of nodes - -Here are the possible types, with short descriptions: - - # TYPE arg-description [num-args] [longjump-len] DESCRIPTION - - # Exit points - END no End of program. - SUCCEED no Return from a subroutine, basically. - - # Anchors: - BOL no Match "" at beginning of line. - MBOL no Same, assuming multiline. - SBOL no Same, assuming singleline. - EOS no Match "" at end of string. - EOL no Match "" at end of line. - MEOL no Same, assuming multiline. - SEOL no Same, assuming singleline. - BOUND no Match "" at any word boundary - BOUNDL no Match "" at any word boundary - NBOUND no Match "" at any word non-boundary - NBOUNDL no Match "" at any word non-boundary - GPOS no Matches where last m//g left off. - - # [Special] alternatives - ANY no Match any one character (except newline). - SANY no Match any one character. - ANYOF sv Match character in (or not in) this class. - ALNUM no Match any alphanumeric character - ALNUML no Match any alphanumeric char in locale - NALNUM no Match any non-alphanumeric character - NALNUML no Match any non-alphanumeric char in locale - SPACE no Match any whitespace character - SPACEL no Match any whitespace char in locale - NSPACE no Match any non-whitespace character - NSPACEL no Match any non-whitespace char in locale - DIGIT no Match any numeric character - NDIGIT no Match any non-numeric character - - # BRANCH The set of branches constituting a single choice are hooked - # together with their "next" pointers, since precedence prevents - # anything being concatenated to any individual branch. The - # "next" pointer of the last BRANCH in a choice points to the - # thing following the whole choice. This is also where the - # final "next" pointer of each individual branch points; each - # branch starts with the operand node of a BRANCH node. - # - BRANCH node Match this alternative, or the next... - - # BACK Normal "next" pointers all implicitly point forward; BACK - # exists to make loop structures possible. - # not used - BACK no Match "", "next" ptr points backward. - - # Literals - EXACT sv Match this string (preceded by length). - EXACTF sv Match this string, folded (prec. by length). - EXACTFL sv Match this string, folded in locale (w/len). - - # Do nothing - NOTHING no Match empty string. - # A variant of above which delimits a group, thus stops optimizations - TAIL no Match empty string. Can jump here from outside. - - # STAR,PLUS '?', and complex '*' and '+', are implemented as circular - # BRANCH structures using BACK. Simple cases (one character - # per match) are implemented with STAR and PLUS for speed - # and to minimize recursive plunges. - # - STAR node Match this (simple) thing 0 or more times. - PLUS node Match this (simple) thing 1 or more times. - - CURLY sv 2 Match this simple thing {n,m} times. - CURLYN no 2 Match next-after-this simple thing - # {n,m} times, set parens. - CURLYM no 2 Match this medium-complex thing {n,m} times. - CURLYX sv 2 Match this complex thing {n,m} times. - - # This terminator creates a loop structure for CURLYX - WHILEM no Do curly processing and see if rest matches. - - # OPEN,CLOSE,GROUPP ...are numbered at compile time. - OPEN num 1 Mark this point in input as start of #n. - CLOSE num 1 Analogous to OPEN. - - REF num 1 Match some already matched string - REFF num 1 Match already matched string, folded - REFFL num 1 Match already matched string, folded in loc. - - # grouping assertions - IFMATCH off 1 2 Succeeds if the following matches. - UNLESSM off 1 2 Fails if the following matches. - SUSPEND off 1 1 "Independent" sub-regex. - IFTHEN off 1 1 Switch, should be preceded by switcher . - GROUPP num 1 Whether the group matched. - - # Support for long regex - LONGJMP off 1 1 Jump far away. - BRANCHJ off 1 1 BRANCH with long offset. - - # The heavy worker - EVAL evl 1 Execute some Perl code. - - # Modifiers - MINMOD no Next operator is not greedy. - LOGICAL no Next opcode should set the flag only. - - # This is not used yet - RENUM off 1 1 Group with independently numbered parens. - - # This is not really a node, but an optimized away piece of a "long" node. - # To simplify debugging output, we mark it as if it were a node - OPTIMIZED off Placeholder for dump. +=head2 Types of Nodes + +Here are the current possible types, with short descriptions: + +=for comment +This table is generated by regen/regcomp.pl. Any changes made here +will be lost. + +=for regcomp.pl begin + + # TYPE arg-description [num-args] [longjump-len] DESCRIPTION + + # Exit points + + END no End of program. + SUCCEED no Return from a subroutine, basically. + + # Line Start Anchors: + SBOL no Match "" at beginning of line: /^/, /\A/ + MBOL no Same, assuming multiline: /^/m + + # Line End Anchors: + SEOL no Match "" at end of line: /$/ + MEOL no Same, assuming multiline: /$/m + EOS no Match "" at end of string: /\z/ + + # Match Start Anchors: + GPOS no Matches where last m//g left off. + + # Word Boundary Opcodes: + BOUND no Like BOUNDA for non-utf8, otherwise match "" + between any Unicode \w\W or \W\w + BOUNDL no Like BOUND/BOUNDU, but \w and \W are defined + by current locale + BOUNDU no Match "" at any boundary of a given type + using Unicode rules + BOUNDA no Match "" at any boundary between \w\W or + \W\w, where \w is [_a-zA-Z0-9] + NBOUND no Like NBOUNDA for non-utf8, otherwise match + "" between any Unicode \w\w or \W\W + NBOUNDL no Like NBOUND/NBOUNDU, but \w and \W are + defined by current locale + NBOUNDU no Match "" at any non-boundary of a given type + using using Unicode rules + NBOUNDA no Match "" betweeen any \w\w or \W\W, where \w + is [_a-zA-Z0-9] + + # [Special] alternatives: + REG_ANY no Match any one character (except newline). + SANY no Match any one character. + CANY no Match any one byte. + ANYOF sv 1 Match character in (or not in) this class, + single char match only + ANYOFL sv 1 Like ANYOF, but /l is in effect + + # POSIX Character Classes: + POSIXD none Some [[:class:]] under /d; the FLAGS field + gives which one + POSIXL none Some [[:class:]] under /l; the FLAGS field + gives which one + POSIXU none Some [[:class:]] under /u; the FLAGS field + gives which one + POSIXA none Some [[:class:]] under /a; the FLAGS field + gives which one + NPOSIXD none complement of POSIXD, [[:^class:]] + NPOSIXL none complement of POSIXL, [[:^class:]] + NPOSIXU none complement of POSIXU, [[:^class:]] + NPOSIXA none complement of POSIXA, [[:^class:]] + + CLUMP no Match any extended grapheme cluster sequence + + # Alternation + + # BRANCH The set of branches constituting a single choice are + # hooked together with their "next" pointers, since + # precedence prevents anything being concatenated to + # any individual branch. The "next" pointer of the last + # BRANCH in a choice points to the thing following the + # whole choice. This is also where the final "next" + # pointer of each individual branch points; each branch + # starts with the operand node of a BRANCH node. + # + BRANCH node Match this alternative, or the next... + + # Literals + + EXACT str Match this string (preceded by length). + EXACTL str Like EXACT, but /l is in effect. + EXACTF str Match this non-UTF-8 string (not guaranteed + to be folded) using /id rules (w/len). + EXACTFL str Match this string (not guaranteed to be + folded) using /il rules (w/len). + EXACTFU str Match this string (folded iff in UTF-8, + length in folding doesn't change if not in + UTF-8) using /iu rules (w/len). + EXACTFA str Match this string (not guaranteed to be + folded) using /iaa rules (w/len). + + EXACTFU_SS str Match this string (folded iff in UTF-8, + length in folding may change even if not in + UTF-8) using /iu rules (w/len). + EXACTFLU8 str Rare cirucmstances: like EXACTFU, but is + under /l, UTF-8, folded, and everything in + it is above 255. + EXACTFA_NO_TRIE str Match this string (which is not trie-able; + not guaranteed to be folded) using /iaa + rules (w/len). + + # Do nothing types + + NOTHING no Match empty string. + # A variant of above which delimits a group, thus stops optimizations + TAIL no Match empty string. Can jump here from + outside. + + # Loops + + # STAR,PLUS '?', and complex '*' and '+', are implemented as + # circular BRANCH structures. Simple cases + # (one character per match) are implemented with STAR + # and PLUS for speed and to minimize recursive plunges. + # + STAR node Match this (simple) thing 0 or more times. + PLUS node Match this (simple) thing 1 or more times. + + CURLY sv 2 Match this simple thing {n,m} times. + CURLYN no 2 Capture next-after-this simple thing + CURLYM no 2 Capture this medium-complex thing {n,m} + times. + CURLYX sv 2 Match this complex thing {n,m} times. + + # This terminator creates a loop structure for CURLYX + WHILEM no Do curly processing and see if rest matches. + + # Buffer related + + # OPEN,CLOSE,GROUPP ...are numbered at compile time. + OPEN num 1 Mark this point in input as start of #n. + CLOSE num 1 Analogous to OPEN. + + REF num 1 Match some already matched string + REFF num 1 Match already matched string, folded using + native charset rules for non-utf8 + REFFL num 1 Match already matched string, folded in loc. + REFFU num 1 Match already matched string, folded using + unicode rules for non-utf8 + REFFA num 1 Match already matched string, folded using + unicode rules for non-utf8, no mixing ASCII, + non-ASCII + + # Named references. Code in regcomp.c assumes that these all are after + # the numbered references + NREF no-sv 1 Match some already matched string + NREFF no-sv 1 Match already matched string, folded using + native charset rules for non-utf8 + NREFFL no-sv 1 Match already matched string, folded in loc. + NREFFU num 1 Match already matched string, folded using + unicode rules for non-utf8 + NREFFA num 1 Match already matched string, folded using + unicode rules for non-utf8, no mixing ASCII, + non-ASCII + + # Support for long RE + LONGJMP off 1 1 Jump far away. + BRANCHJ off 1 1 BRANCH with long offset. + + # Special Case Regops + IFMATCH off 1 2 Succeeds if the following matches. + UNLESSM off 1 2 Fails if the following matches. + SUSPEND off 1 1 "Independent" sub-RE. + IFTHEN off 1 1 Switch, should be preceded by switcher. + GROUPP num 1 Whether the group matched. + + # The heavy worker + + EVAL evl/flags Execute some Perl code. + 2L + + # Modifiers + + MINMOD no Next operator is not greedy. + LOGICAL no Next opcode should set the flag only. + + # This is not used yet + RENUM off 1 1 Group with independently numbered parens. + + # Trie Related + + # Behave the same as A|LIST|OF|WORDS would. The '..C' variants + # have inline charclass data (ascii only), the 'C' store it in the + # structure. + + TRIE trie 1 Match many EXACT(F[ALU]?)? at once. + flags==type + TRIEC trie Same as TRIE, but with embedded charclass + charclass data + + AHOCORASICK trie 1 Aho Corasick stclass. flags==type + AHOCORASICKC trie Same as AHOCORASICK, but with embedded + charclass charclass data + + # Regex Subroutines + GOSUB num/ofs 2L recurse to paren arg1 at (signed) ofs arg2 + GOSTART no recurse to start of pattern + + # Special conditionals + NGROUPP no-sv 1 Whether the group matched. + INSUBP num 1 Whether we are in a specific recurse. + DEFINEP none 1 Never execute directly. + + # Backtracking Verbs + ENDLIKE none Used only for the type field of verbs + OPFAIL none Same as (?!) + ACCEPT parno 1 Accepts the current matched string. + + # Verbs With Arguments + VERB no-sv 1 Used only for the type field of verbs + PRUNE no-sv 1 Pattern fails at this startpoint if no- + backtracking through this + MARKPOINT no-sv 1 Push the current location for rollback by + cut. + SKIP no-sv 1 On failure skip forward (to the mark) before + retrying + COMMIT no-sv 1 Pattern fails outright if backtracking + through this + CUTGROUP no-sv 1 On failure go to the next alternation in the + group + + # Control what to keep in $&. + KEEPS no $& begins here. + + # New charclass like patterns + LNBREAK none generic newline pattern + + # SPECIAL REGOPS + + # This is not really a node, but an optimized away piece of a "long" + # node. To simplify debugging output, we mark it as if it were a node + OPTIMIZED off Placeholder for dump. + + # Special opcode with the property that no opcode in a compiled program + # will ever be of this type. Thus it can be used as a flag value that + # no other opcode has been seen. END is used similarly, in that an END + # node cant be optimized. So END implies "unoptimizable" and PSEUDO + # mean "not seen anything to optimize yet". + PSEUDO off Pseudo opcode for internal use. + +=for regcomp.pl end =for unprinted-credits Next section M-J. Dominus (mjd-perl-patch+@plover.com) 20010421 @@ -675,7 +821,7 @@ is, it corresponds to the C<+> symbol in the precompiled regex. C<0[0]> items indicate that there is no corresponding node. -=head2 Run-time output +=head2 Run-time Output First of all, when doing a match, one may get no run-time output even if debugging is enabled. This means that the regex engine was never @@ -683,7 +829,7 @@ entered and that all of the job was therefore done by the optimizer. If the regex engine was entered, the output may look like this: - Matching `[bc]d(ef*g)+h[ij]k$' against `abcdefg__gh__' + Matching '[bc]d(ef*g)+h[ij]k$' against 'abcdefg__gh__' Setting an EVAL scope, savestack=3 2 | 1: ANYOF 3 | 11: EXACT @@ -718,7 +864,7 @@ C< >I > > |I: I The I info is indented with respect to the backtracking level. Other incidental information appears interspersed within. -=head1 Debugging Perl memory usage +=head1 Debugging Perl Memory Usage Perl is a profligate wastrel when it comes to memory use. There is a saying that to estimate memory usage of Perl, assume a reasonable @@ -754,7 +900,7 @@ The B<-DL> command-line switch is obsolete since circa Perl 5.6.0 The switch was used to track Perl's memory allocations and possible memory leaks. These days the use of malloc debugging tools like F or F is suggested instead. See also -L. +L. One way to find out how much memory is being used by Perl data structures is to install the Devel::Size module from CPAN: it gives @@ -763,7 +909,7 @@ structure. Please be mindful of the difference between the size() and total_size(). If Perl has been compiled using Perl's malloc you can analyze Perl -memory usage by setting the $ENV{PERL_DEBUG_MSTATS}. +memory usage by setting $ENV{PERL_DEBUG_MSTATS}. =head2 Using C<$ENV{PERL_DEBUG_MSTATS}> @@ -811,7 +957,7 @@ would have usable size 8188, and the memory footprint would be 8192. In a Perl built for debugging, some buckets may have negative usable size. This means that these buckets cannot (and will not) be used. For larger buckets, the memory footprint may be one page greater -than a power of 2. If so, case the corresponding power of two is +than a power of 2. If so, the corresponding power of two is printed in the C field above. =item Free/Used @@ -829,7 +975,7 @@ were free: 8 16 32 64 128 256 512 1024 2048 4096 8192 4 12 24 48 80 -With non-C perl, the buckets starting from C<128> have +With a non-C perl, the buckets starting from C<128> have a 4-byte overhead, and thus an 8192-long bucket may take up to 8188-byte allocations.