summaryrefslogtreecommitdiff
path: root/Zend/zend_execute.c
Commit message (Collapse)AuthorAgeFilesLines
...
* Specializer optimizationDmitry Stogov2005-06-131-195/+77
|
* Fixed bug #33212 ([GCC 4]: 'zend_error_noreturn' aliased to external symbol ↵Dmitry Stogov2005-06-131-9/+0
| | | | | | | 'zend_error'). The fix is not tested on Solaris and DARWIN!
* Fix so that extensions like xdebug, can overload opcodes in all execution ↵Dmitry Stogov2005-06-101-0/+15
| | | | modes including goto/switch
* fixed memory leak in bug #28972 ([] operator overflow treatment is ↵Dmitry Stogov2005-06-071-1/+1
| | | | incorrect), not the bug itself.
* Fixed memory allocation bugs related to magic object handlers (__get(), __set(),Dmitry Stogov2005-06-031-50/+34
| | | | | ...)
* Added array type hinting. (This patch requires full re-make)Dmitry Stogov2005-05-261-1/+17
|
* Fixed bug #30641 (Compile error: error: symbol "zend_error" is used but not ↵Dmitry Stogov2005-05-041-1/+1
| | | | defined)
* Fixed bug #32852 (Crash with singleton and __destruct when ↵Dmitry Stogov2005-04-291-11/+15
| | | | | | | | zend.ze1_compatibility_mode = On) Fixed bug #31828 (Crash with zend.ze1_compatibility_mode=On) Fixed bug #32080 (segfault when assigning object to itself with zend.ze1_compatibility_mode=On)
* Fix certain operations to stop relying on presence of read_property andAndrei Zmievski2005-04-151-1/+1
| | | | | | write_property handlers. They may be NULL'ed out by certain objects pretending to be pure arrays, for example. Do checks first.
* Consolidate: call _get_zval_ptr_var() for IS_VAR case inAndrei Zmievski2005-03-231-66/+36
| | | | | _get_zval_ptr().
* - Two new API calls for Derick (retreive CV name and value) by DmitryAndi Gutmans2005-01-221-0/+5
|
* Fixed bug #28444 (Cannot access undefined property for object with ↵Dmitry Stogov2005-01-181-3/+12
| | | | overloaded property access).
* Restore behavior of $str["str"]. (Now $str["str"] is equivalent to $str[0] ↵Dmitry Stogov2005-01-141-22/+1
| | | | again)
* Additional fix for fix of bug #29883Dmitry Stogov2005-01-131-1/+1
|
* Fixed bug #31098 (isset false positive)Dmitry Stogov2005-01-121-1/+22
|
* "Undefined variable: %s" noticies were fixed to use one spaceDmitry Stogov2004-12-271-8/+8
|
* "Undefined variable: %s" noticies were fixed to be compatible with PHP_5_0Dmitry Stogov2004-12-241-4/+4
|
* - Checks for Darwin'ish systems that uses Mach-O, which apparently doesn'tMoriyoshi Koizumi2004-11-031-1/+1
| | | | | support weak symbol aliasing at this time.
* - Commit new VMAndi Gutmans2004-10-221-262/+47
| | | | | | - Old one is tagged as PRE_NEW_VM_GEN_PATCH - Still doing work so more commits to come. Don't complain (yet) :)
* - Fix crash (MFB PHP5_0)Andi Gutmans2004-10-221-1/+1
|
* Fixed unset() bug that was introduced with CV optimization patchDmitry Stogov2004-10-051-0/+4
|
* - Commit the variable fetch optimization.Andi Gutmans2004-10-041-15/+159
| | | | | | | - Extensions which delete global variables need to use new special function - delete_global_variable() (I'm about to rename it) to remove them. - Will post to internals@ or via commit messages if there's anything else.
* - Commit new VM architecture. This one allows people (aka Derick) toAndi Gutmans2004-09-231-3770/+459
| | | | | | - ask the engine to use function handler mode. Will update the README - about that.
* Reverted the NetWare Specific Stack limit related patches as asked by AndiAnantha Kesari H Y2004-09-231-5/+0
|
* Stack limit will be checked while executing the scriptAnantha Kesari H Y2004-09-221-1/+5
|
* Fixed bug #29566 (foreach/string handling strangeness (crash)).Dmitry Stogov2004-09-221-1/+1
|
* Fixed bug in fix for bug #29707Dmitry Stogov2004-09-221-1/+0
|
* - Fix for bug #29707Andi Gutmans2004-09-211-24/+17
|
* Antony Dovgal's error message improvement - #27290Stanislav Malyshev2004-09-131-1/+11
|
* - Roll back VM commitAndi Gutmans2004-09-091-356/+3805
|
* - Some architectural changes:Andi Gutmans2004-09-081-3805/+356
| | | | | | | | | | | | | a) We specialize opcodes according to op_type fields. Each opcode has to be marked with which op_type's it uses. b) We support different execution methods. Function handlers, switch() and goto dispatching. goto seems to be the fastest but it really depends on the compiler and how well it optimizes. I suggest playing around with optimization flags. - Warning: Things might break so keep us posted on how things are going. (Dmitry, Andi)
* fix crash #29893Stanislav Malyshev2004-08-301-0/+3
|
* String offset starts with 0, fix isset($str[$len])Marcus Boerger2004-08-291-2/+2
|
* - CleanupAndi Gutmans2004-08-191-1/+1
|
* - Stop using garbage. Please let me know if you find any bugs resultingAndi Gutmans2004-08-191-370/+745
| | | | | - of this patch (very likely). (Dmitry, Andi)
* - Don't use magic numbersAndi Gutmans2004-08-121-4/+3
|
* - Significantly improve performance of foreach($arr as $data). (Marcus)Andi Gutmans2004-08-121-30/+53
|
* More meaningfull error messageMarcus Boerger2004-08-111-1/+1
|
* - Fix warningsMarcus Boerger2004-08-101-1/+1
|
* A gentler (performance wise) allocation of buffer for temp variables.Ilia Alshanetsky2004-08-021-2/+11
|
* - More ptr_stack optimizations and cleanupsAndi Gutmans2004-07-301-2/+2
|
* - Improve performance by inlining zend_ptr_stack_n_push(). var_args canAndi Gutmans2004-07-301-5/+5
| | | | | usually not be inlined by compilers.
* Revert goto opcodeSara Golemon2004-07-291-27/+0
|
* &tmp and label are the same thing, don't free it till we're done with it.Sara Golemon2004-07-291-1/+1
|
* Plug some memory leaks and promote unknown label to E_ERROR.Sara Golemon2004-07-291-1/+3
| | | | | | If someone tries to jump to a non-existant label execution really shouldn't try to carry on.
* Add goto operator by popular request.Sara Golemon2004-07-291-0/+25
|
* - Fix bug #29368 : The destructor is called when an exception is thrown from ↵Marcus Boerger2004-07-251-0/+3
| | | | the constructor
* Fix 0 Byte leak after alloca to emalloc changeMarcus Boerger2004-07-231-2/+2
|
* - Nuke empty_string. It is a reminanent from the time where RETURN_FALSE()Andi Gutmans2004-07-191-1/+1
| | | | | | | | | | | | | | | used to return "" and not bool(false). It's not worth keeping it because STR_FREE() and zval_dtor() always have to check for it and it slows down the general case. In addition, it seems that empty_string has been abused quite a lot, and was used not only for setting zval's but generally in PHP code instead of "", which wasn't the intention. Last but not least, nuking empty_string should improve stability as I doubt every place correctly checked if they are not mistakenly erealloc()'ing it or calling efree() on it. NOTE: Some code is probably broken. Each extension maintainer should check and see that my changes are OK. Also, I haven't had time to touch PECL yet. Will try and do it tomorrow.
* Bugfix #28464 catch() does not catch exceptions by interfacesMarcus Boerger2004-07-161-11/+6
|
View Lorry Controller Status