summaryrefslogtreecommitdiff
path: root/sv.c
Commit message (Collapse)AuthorAgeFilesLines
* move PL_expect and PL_copline into the PL_parser structureDave Mitchell2007-05-041-5/+5
| | | p4raw-id: //depot/perl@31134
* Initialise PL_delayedisa on thread clone.Nicholas Clark2007-05-031-0/+1
| | | p4raw-id: //depot/perl@31125
* Re: mro status, etcBrandon Black2007-04-301-5/+34
| | | | | | From: "Brandon Black" <blblack@gmail.com> Message-ID: <84621a60704291527y1b39be37l221ef66e4c828f66@mail.gmail.com> p4raw-id: //depot/perl@31107
* Fix a bug in method caching. Better version (broader) of change #29336.Brandon Black2007-04-261-3/+3
| | | | | | | | Subject: Re: mro status, etc From: "Brandon Black" <blblack@gmail.com> Message-ID: <84621a60704260852y5a3dd2a5jeb633bb46cc7a8c0@mail.gmail.com> p4raw-link: @29336 on //depot/perl: ae6d515f69537dd5e2631b15104c2c90d022fd19 p4raw-id: //depot/perl@31091
* Make PL_uudmap a constant global. Remove the run time initialisationNicholas Clark2007-04-241-1/+0
| | | | | | | code, and instead use it to generate a header at perl build time. By removing uudmap from the interpreter structure we save 256 bytes per child thread. p4raw-id: //depot/perl@31059
* move PL_linestr from the interpreter struct to the parser structDave Mitchell2007-04-241-14/+16
| | | p4raw-id: //depot/perl@31058
* Where possible, use SvIV instead of SvIVX, SvNV instead of SvNVX,Nicholas Clark2007-04-211-6/+6
| | | | | | | | | SvUV instead of SvUVX, and SvPV* variants instead of SvPVX*. Document that the non-x variants are preferable whenever the expression has no side effects. (Compilers perform common subexression elimination). Likewise SvREFCNT_inc simple variants are valid for all cases apart from expressions with side effects. p4raw-id: //depot/perl@31010
* Re: new C3 MRO patchBrandon Black2007-04-191-2/+7
| | | | | | From: "Brandon Black" <blblack@gmail.com> Message-ID: <84621a60704171114k29b0460el5b08ce5185d55ed5@mail.gmail.com> p4raw-id: //depot/perl@30980
* sv.c: %d ne size_tJarkko Hietaniemi2007-04-161-2/+2
| | | | | Message-Id: <200704161131.l3GBVg3n104476@kosh.hut.fi> p4raw-id: //depot/perl@30966
* follow PL_watchaddr when cloning a threadDave Mitchell2007-04-151-2/+10
| | | p4raw-id: //depot/perl@30963
* with DEBUG_LEAKING_SCALARS, dump multiply-freed scalarsDave Mitchell2007-04-151-0/+4
| | | p4raw-id: //depot/perl@30962
* Move PL_comppad nulling from do_clean_all to sv_clearDave Mitchell2007-04-141-4/+4
| | | p4raw-id: //depot/perl@30952
* Fix two errors in the OP debugging code. Now all test errors relateNicholas Clark2007-04-121-0/+4
| | | | | to newCONSTSUB(). p4raw-id: //depot/perl@30936
* Ensure PL_last_in_gv doesn't end up pointing to garbage by NULLing itNicholas Clark2007-04-071-0/+5
| | | | | if the GV it is pointing to gets freed. p4raw-id: //depot/perl@30868
* Remove op_pmnext from PMOPs, and instead store the list for reset asNicholas Clark2007-04-061-8/+8
| | | | | | an array hanging from the mg_ptr of the symbol table magic. (Previously the linked list head was in the mg_obj member) p4raw-id: //depot/perl@30853
* Re: pmdynflags and thread safetyYves Orton2007-04-041-1/+5
| | | | | Message-ID: <9b18b3110704031646p7ac8dbearf9e41397a5f884d8@mail.gmail.com> p4raw-id: //depot/perl@30841
* Eliminate the use of PL_bufend outside of toke.c:Dave Mitchell2007-04-011-6/+1
| | | | | | give Perl_scan_vstring() an explicit 'end' arg rather than using PL_bufend, and replace it with a local var in Perl_find_script() p4raw-id: //depot/perl@30820
* In struct regexp replace the two arrays of I32s accessed via startpNicholas Clark2007-03-261-4/+3
| | | | | | | | | | and endp with a single array of struct regexp_paren_pair, which has 2 I32 members. PL_regstartp and PL_regendp are replaced with a pointer to regexp_paren_pair. The regexp swap structure now only has one member, so abolish it and store the pointer to the swap array directly. Hopefully keeping the corresponding start and end adjacent in memory will help with cache coherency. p4raw-id: //depot/perl@30769
* say() should parse exactly like print()Rafael Garcia-Suarez2007-03-261-0/+1
| | | p4raw-id: //depot/perl@30760
* Don't try to set the NV on a typeglob-in-action. Should cure bugNicholas Clark2007-03-211-1/+2
| | | | | 41920. p4raw-id: //depot/perl@30659
* AVs and HVs never had the NV slot. (Not that this error mattered, asNicholas Clark2007-03-211-2/+2
| | | | | | you can't upgrade from them, and the only code path that uses the flag in sv_upgrade isn't reached when upgrading to AVs or HVs) p4raw-id: //depot/perl@30658
* Re: [PATCH] (Re: [PATCH] unicode/utf8 pod)Juerd Waalboer2007-03-071-1/+1
| | | | | Message-ID: <20070304150019.GN4723@c4.convolution.nl> p4raw-id: //depot/perl@30493
* As SvUPGRADE() is a macro wrapping a call to sv_upgrade() insideNicholas Clark2007-03-051-3/+1
| | | | | | | a check on the existing SvTYPE(), there's no need to wrap it inside another explcit check of SvTYPE(). This won't make any difference to the output of an optimising compiler, but it makes the source clearer. p4raw-id: //depot/perl@30481
* Avoid *some* g++ errors. (But not all yet)Nicholas Clark2007-02-191-1/+1
| | | p4raw-id: //depot/perl@30365
* blead (honestly :-) g++ with -DPERL_GLOBAL_STRUCT_PRIVATE needs tweakingJarkko Hietaniemi2007-02-191-2/+2
| | | | | | | | Message-Id: <20070219174107.63EEB43A67@anubis.hut.fi> Plus a regen picked up changes in pod/perlapi.pod related to change #30347. p4raw-id: //depot/perl@30362
* Add a new API function newSV_type, to replace the idiom:Nicholas Clark2007-02-181-4/+20
| | | | | | sv = newSV(0); sv_upgrade(sv, type); p4raw-id: //depot/perl@30347
* Use newer 'threads' constructsJerry D. Hedden2007-02-151-1/+1
| | | | | | From: "Jerry D. Hedden" <jdhedden@yahoo.com> Message-ID: <844555.64815.qm@web30202.mail.mud.yahoo.com> p4raw-id: //depot/perl@30310
* Purge all references to 5005 threads variables in the code.Nicholas Clark2007-02-031-22/+0
| | | | | (Retain the aides memoire in makedef.pl) p4raw-id: //depot/perl@30105
* g++ wants castJarkko Hietaniemi2007-01-281-1/+1
| | | | | Message-Id: <20070128150444.AC62043A57@anubis.hut.fi> p4raw-id: //depot/perl@30050
* Silence a few pedantic warningsRafael Garcia-Suarez2007-01-261-4/+4
| | | p4raw-id: //depot/perl@30017
* The last parameter to gv_stashpv/gv_stashpvn/gv_stashsv is a bitmaskNicholas Clark2007-01-251-1/+1
| | | | | of flags, not a boolean, so correct the documenation and callers. p4raw-id: //depot/perl@29977
* Fix breakage for -DDEBUGGING accidentally introduced with change 29882.Nicholas Clark2007-01-191-1/+1
| | | p4raw-id: //depot/perl@29883
* Add a parameter to Perl_get_arena() to pass in the SV type, and recordNicholas Clark2007-01-191-21/+18
| | | | | | | this in the arena description. Change all sizes to unsigned values. Make Perl_sv_free_arenas() loop downwards to free memory, simplifying the logic. Remove my erroneous comment added in change 29881. p4raw-id: //depot/perl@29882
* I think that it's clearer if aroot is only struct arena_set *Nicholas Clark2007-01-191-7/+12
| | | | | | | (rather than ** and constantly dereferneced). Move the declaration of new_root into the block it is used in. Add a comment describing where the arenas for arena_sets themselves aren't rooted anywhere. p4raw-id: //depot/perl@29881
* As Perl_get_arena() is dealing with sizes, use size_t rather than int,Nicholas Clark2007-01-191-2/+2
| | | | | | as it's both unsigned and semantically the correct width for a size. As all arenas get cleared later on, can use Newx() rather than Newxz(). p4raw-id: //depot/perl@29879
* Move C<static bool done_sanity_check;> inside the only function thatNicholas Clark2007-01-181-7/+4
| | | | | uses it. p4raw-id: //depot/perl@29878
* Move the flag checking for isGV_with_GP() into the call to visit()Nicholas Clark2007-01-181-2/+3
| | | p4raw-id: //depot/perl@29876
* Given that we now do the tests in S_visit() to avoid calling theNicholas Clark2007-01-181-2/+4
| | | | | | helper function when skippable, no need to retain those tests inside the helper functions do_clean_objs() and do_clean_named_objs(). p4raw-id: //depot/perl@29872
* As PVLVs can hold globs in some cases, I think that we should beNicholas Clark2007-01-181-10/+5
| | | | | | treating them as a strict subclass of PVGV. Hence drop through into the PVGV handling code. p4raw-id: //depot/perl@29867
* Replace SvRELEASE_IVX(dstr) with SvOOK_off(dstr) in sv_setsv_flags(),Nicholas Clark2007-01-171-1/+1
| | | | | | because it's not possible for dstr to be COW at this point, due to an earlier force_normal. p4raw-id: //depot/perl@29855
* Make PERL_OLD_COPY_ON_WRITE build again. Inline Perl_sv_release_IVX().Nicholas Clark2007-01-171-17/+18
| | | | | (Currently it fails ext/Compress/Raw/Zlib/t/07bufsize.t) p4raw-id: //depot/perl@29853
* If we use SvUPGRADE() rather than sv_upgrade() then it can make theNicholas Clark2007-01-171-3/+2
| | | | | | decision on not "up"grading PVLV to PVGV for us. However, it appears that the other actions conducted for "not-yet-a-glob" are useful here. p4raw-id: //depot/perl@29852
* when cloning PL_regex_pad, copy SVf_BREAK flag tooDave Mitchell2007-01-151-0/+2
| | | p4raw-id: //depot/perl@29837
* 4th patch from: Marcus Holland-Moritz2007-01-041-8/+8
| | | | | | | | | Subject: [PATCH] Cleanup SVf arguments (2nd try) Message-ID: <20070101201613.4120d9ef@r2d2> Introduce an SVfARG() macro for %SVf (%-p here) arguments to perl's printf p4raw-id: //depot/perl@29687
* The first three patches from:Marcus Holland-Moritz2007-01-041-9/+9
| | | | | | | | Subject: [PATCH] Cleanup SVf arguments (2nd try) Message-ID: <20070101201613.4120d9ef@r2d2> Adds (void*) casts to %-p and %p printf arguments p4raw-id: //depot/perl@29686
* Re: [PATCH] Change implementation of %+ to use a proper tied hash interface ↵Yves Orton2007-01-041-3/+0
| | | | | | | and add support for %- Message-ID: <9b18b3110612291245q792fe91cu69422d2b81bb4f0b@mail.gmail.com> p4raw-id: //depot/perl@29682
* Rename OURSTASH to SvOURSTASH and OURSTASH_set to SvOURSTASH_set.Nicholas Clark2007-01-031-2/+2
| | | p4raw-id: //depot/perl@29679
* handle cloning of parsers on the save stackDave Mitchell2007-01-031-0/+11
| | | p4raw-id: //depot/perl@29678
* Make the ptr_table functions available in an unthreaded perl, as otherNicholas Clark2007-01-031-0/+3
| | | | | code (such as Storable) can take advantage of them. p4raw-id: //depot/perl@29673
* Copying to FORMATs should work (and not fail assertions if the valueNicholas Clark2007-01-021-1/+8
| | | | | | is floating point). Copying to ARRAYs and HASHes can't work and should croak. p4raw-id: //depot/perl@29666
View Lorry Controller Status