Histórico de commits

Autor SHA1 Mensagem Data
  maciej-izak 99f7c010a8 fcl-passrc: similar to r36619. New option SemicolonAtEOL for better control of TPasImplCommand (previously, semicolon was implicitly added for command for generated pascal code by paswrite.pp module for TPasImplCommand) 8 anos atrás
  svenbarth ad65ff5600 * insert symbol only once 8 anos atrás
  maciej-izak aaf7f5137e fcl-passrc: new option SemicolonAtEOL for better control of TPasImplCommands (previously, semicolon was implicitly added for each line for generated pascal code by paswrite.pp module for TPasImplCommands) 8 anos atrás
  Károly Balogh 2d8313165d m68k: support FFU exception settings, also reworked FPU initalization 8 anos atrás
  svenbarth a301bf75ea * ngenutil.tnodeutils.sym_maybe_initialize: don't use "is" operator, but corresponding is_* functions 8 anos atrás
  maciej-izak 6006c6a8bd fcl-passrc: paswrite: allow write TPasImplCommand and TPasImplCommands directly in TPasWriter.WriteElement. Useful way to build pascal code by using predfined code blocks. 8 anos atrás
  maciej-izak 584ef096a0 fcl-passrc: paswrite: write interfaces GUID 8 anos atrás
  Károly Balogh 4001435454 fixed typo in previous commit 8 anos atrás
  Károly Balogh a12111e1bd m68k: define safecall_is_cdecl on m68k too in tcalvar6 and tcalst6 8 anos atrás
  Károly Balogh 358f8eb85d m68k: do not internalerror on localsize < 0. this condition is handled elsewhere in the compiler properly, so we just let it through 8 anos atrás
  Károly Balogh ad89a972e5 m68k: also have tf_safecall_clearstack flag defined for Linux 8 anos atrás
  maciej-izak 3fd4e52345 fcl-passrc: paswrite: handled calling conventions for procedure declaration 8 anos atrás
  Károly Balogh 0370d52f20 m68k: support longword to double conversions with the FPU, without a helper, better code for some other cases 8 anos atrás
  maciej-izak bcacbe1cba fcl-passrc: paswrite: proper write for properties (previously WriteProperty was not executed) 8 anos atrás
  maciej-izak b35522d947 Commit again r36573 reverted by r36601 -,- 8 anos atrás
  michael 54a8a73ca7 * Append end to correct location in class 8 anos atrás
  Károly Balogh 02ed753fab m68k: yet another attempt to fix small struct alignments on stack 8 anos atrás
  Károly Balogh 8a8753eb79 m68k: reenabled some safety checks and removed obsolete TODO comment 8 anos atrás
  Károly Balogh d5e1b391f9 m68k: when calling g_concatcopy for para copy, still only copy cgpara.intsize amount of bytes 8 anos atrás
  Jeppe Johansen 09a8cafcd7 Restricted MlaCmp>Mlas optimization to only work in ARM mode. 8 anos atrás
  Mattias Gaertner 8935b3c05c fcl-passrc: const eval: pred(), succ(), ord() 8 anos atrás
  Mattias Gaertner 84b08b61a8 fcl-js: jssrcmap: added property SourceTranslatedFiles to allow custom filenames written to JSON 8 anos atrás
  Mattias Gaertner a66eaa4769 fcl-js: jswriter: allow overriding Writing, less grows(faster and less allocations) 8 anos atrás
  Károly Balogh 22ae3cd186 m68k: revert the r36568 and the followup fix attempts altogether, as they broke Amiga support entirely. there will be a cleanup before another fixing attempt 8 anos atrás
  Jeppe Johansen f3889a191b Generate bx lr exit instruction in Thumb-2 instead of mov pc,lr as bx lr will trigger an exception return but mov doesn't. 8 anos atrás
  Károly Balogh a14b2fd80a m68k: modify some alignment calculations in the register calling convention code, hopefully it fixes some Amiga regressions introduced in r36568 8 anos atrás
  Károly Balogh 890f4ac2bc m68k: use 2 byte alignments for C structs on Linux 8 anos atrás
  Károly Balogh ea635bfda5 typinfo: push/pop doesn't save packrecords. also, always use AlignTypeData on m68k, which might have less than pointer sized (word, 2 byte) alignments in the typeinfo, because on most systems C compilers also use 2 byte alignments, so we need that for compatibility 8 anos atrás
  Károly Balogh 41f72a0e6d m68k: some initial support for C ABIs which use an address register to return structs by address 8 anos atrás
  Károly Balogh da11451934 m68k: revert the 2 byte C record alignment fix on Linux. sadly it completely breaks RTTI, which takes longer to fix, so meanwhile, at least cause no regressions 8 anos atrás