 Sven/Sarah Barth
|
ecbd0fc0da
* move aliases of fpc_copy and fpc_copy_by_move_semantics to aliases and adjust their names to match other aliaes
|
8 months ago |
 Rika Ichinose
|
798d793cdc
Specialized fpc_copy_with_move_semantics for destructive copies.
|
8 months ago |
 Frederic Kehrein
|
ca92c49f8c
* Avoid calling copy operator when moving data from temporary objects
|
9 months ago |
|
Rika Ichinose
|
b8fdd7148b
Simplify RTTIRecordOp.
|
11 months ago |
|
Rika Ichinose
|
cd2c8a6539
Faster RTTIManagementAndSize -_-.
|
11 months ago |
|
Rika Ichinose
|
6de766e40a
Redesign RTTISizeAndOp to return more useful information.
|
11 months ago |
|
Rika Ichinose
|
d7352e7b66
Remove most of the VER3_0 conditionals.
|
11 months ago |
 Michael VAN CANNEYT
|
d2d3fe6bc3
* Char -> AnsiChar
|
2 years ago |
 Károly Balogh
|
45b5e9215c
rtl: remove defines for v2.x compiler from the common part of the RTL. It's no longer suppported to build with that
|
4 years ago |
 svenbarth
|
97c8464b11
* fix for Mantis #36738: when copying a record using its copy operator we assume that we've copied the whole record; this way managed records inside non-managed records are handled correctly
|
5 years ago |
 florian
|
931d4dcfee
* ensure the rtl and the packages for embedded compile with features exceptions and classes disabled
|
5 years ago |
|
svenbarth
|
00a418f589
* also correctly align the pointer for the record elements inside FPC_COPY
|
6 years ago |
|
svenbarth
|
0c46460548
* m68k has slightly different alignment requirements, so that aligntoptr might result in incorrect addresses for a record's element entries, thus we introduce a separate AlignTypeData function that handles this correctly
|
6 years ago |
 pierre
|
d6d3e171ce
Try to fix alignment issues in RTTI code for 64-bit targets which require proper alignment
|
6 years ago |
|
florian
|
a60ab40237
* handle records with managed fields correctly after attribute merge fixes
|
6 years ago |
|
florian
|
217ae6e4bb
* more fixes to rtti after attribute branch merging
|
6 years ago |
|
svenbarth
|
a076677336
* instead of a flags field generate a table with all Initialize operators of all fields (nested or not, parent class or not) for class types; objects and records don't use this (yet)
|
7 years ago |
|
svenbarth
|
8536abce8a
* as the management operator feature is not platform specific using a specific define for it aside from a
|
7 years ago |
|
svenbarth
|
4d0ab82ef7
* optimize class initialization by calling RTTIRecord() only when required for a non-trivial initialization
|
7 years ago |
|
svenbarth
|
393d1c2256
* move RTTI related types from rtti.inc to rttidecl.inc
|
7 years ago |
|
svenbarth
|
bbf578b193
* move definition of USE_PACKED from rtti.inc to rttidecl.inc
|
7 years ago |
|
svenbarth
|
8b5461367b
* move TTypeKind from TypInfo unit to System unit as it's necessary for the future GetTypeKind() intrinsic
|
8 years ago |
 maciej-izak
|
4154b4ca8a
* Call management operators also for records with zero-size. For that purpose was changed function RTTISize. New name for RTTISize is RTTISizeAndOp. RTTISizeAndOp can return size of required type and information about existing management operators for that type in current context (to determine context is used enum TRTTIRecOpType which is passed also as parameter in RTTISizeAndOp). RTTISizeAndOp is used in fpc_initialize_array, fpc_finalize_array, fpc_addref_array and CopyArray.
|
8 years ago |
|
maciej-izak
|
9a9eb1b840
* Use RTTIRecordRttiInfoToInitInfo (when FPC_HAS_MANAGEMENT_OPERATORS is not defined) for records/objects in fpc_Copy as well as for fpc_Initialize, fpc_finalize, fpc_Addref
|
8 years ago |
|
maciej-izak
|
8a68a7aea8
* Declare field RecordOp for TRecordInfoInit only when FPC_HAS_MANAGEMENT_OPERATORS is defined, same for additional types for management operators.
|
8 years ago |
|
maciej-izak
|
d42428bc3c
* Initial support for management operators in RTL (rtl/inc/rtti.inc):
|
8 years ago |
|
maciej-izak
|
af8e0efe57
* Rtti info about Management Operators (MO) in init table for records. Note: commit contains minimal changes for RTL (for proper "make") no functional changes. More RTL changes for fully functionally MO in next commits.
|
8 years ago |
|
florian
|
0b82f71e8d
* take care of the fact that there are 32 bit abis (e.g. ARMHF) which align qword on 8 byte boundaries, resolves issue #31132
|
8 years ago |
|
svenbarth
|
121a857af8
* fix for Mantis #31118: applied patch by Maciej Izak to fix usage of wrong variable
|
8 years ago |
|
svenbarth
|
7089677a22
* change RecInitTable into a direct reference to avoid an unnecessary indirection as both the full and init RTTI of a record always reside inside the same unit
|
8 years ago |
