sergei
|
b8a76c5da2
- Removed gen_proc_symbol procedure, no longer used, replaced by thlcgobj.gen_proc_symbol method.
|
12 gadi atpakaļ |
sergei
|
c3350d13f9
* MIPS: floating point parameters on stack should be loaded to/from FPU registers directly, without using temp.
|
12 gadi atpakaļ |
Jonas Maebe
|
2dd75e707e
* renamed thlcgobj.tcgsize2orddef to defutil.cgsize_orddef
|
12 gadi atpakaļ |
Jonas Maebe
|
5051453806
+ support for LOC_(C)MMREGISTER in hlcg
|
12 gadi atpakaļ |
nickysn
|
47fe8e03b1
* refactored ncgutil.gen_load_cgpara_loc to use cgpara.locations_count
|
12 gadi atpakaļ |
nickysn
|
ecb5a4866d
* refactored the int64 result passing in ax:bx:cx:dx to use 4 paralocs, instead of the GetNextReg hack
|
12 gadi atpakaļ |
florian
|
d4c7afbfe8
+ generate thumb_func directives for thumb as well
|
12 gadi atpakaļ |
florian
|
0e41df598e
* merge i8086 branch by Nikolay Nikolov
|
12 gadi atpakaļ |
nickysn
|
a5c5b05362
* initial support for system procs that use calling conventions that push left to right on i8086 or i386
|
12 gadi atpakaļ |
sergei
|
d54d38b4ab
* Moved all code responsible for writing VMTs and interface wrappers (nobj.TVMTWriter, ncgutil.gen_intf_wrappers, ptype.write_persistent_type_info) into a new unit ncgvmt.pas.
|
12 gadi atpakaļ |
nickysn
|
3870f76cda
* ncgutil: set pushexceptaddr_frametype_cgsize and setjmp_result_cgsize according to cpu16bitaddr, instead of cpu16bitalu
|
12 gadi atpakaļ |
florian
|
075abd6220
+ support of stackframesize for arm thumb
|
12 gadi atpakaļ |
florian
|
43b6139b66
+ setlocalloc resets currentregloc
|
12 gadi atpakaļ |
florian
|
364f170765
* cosmetic formatting changes
|
12 gadi atpakaļ |
nickysn
|
1e28790e99
* fpc_PushExceptAddr's parameter Ft changed from LongInt to SmallInt on 16-bit CPUs for better efficiency
|
12 gadi atpakaļ |
nickysn
|
f38114b34b
* the result of FPC_SETJMP changed to 16-bit int on 16-bit CPUs
|
12 gadi atpakaļ |
nickysn
|
8d2b7df6d6
* fixed functions returning longint on i8086
|
12 gadi atpakaļ |
sergei
|
5e46732bc8
+ One more missing call to unget_para (MIPS-specific)
|
12 gadi atpakaļ |
sergei
|
54aefb1ba0
* gen_load_cgpara_loc, MIPS-specific: added missing calls to unget_para, and disabled part that loads upper and lower halves of Double values from integer registers. It produces invalid code, but that's hidden by MIPS code generator copying all parameters to stack (so this code is never executed).
|
12 gadi atpakaļ |
paul
|
51825b6f2e
compiler: change ret_in_param to accept tabstractprocdef instead of tproccalloption to allow check more options (required for record constructor implementation)
|
12 gadi atpakaļ |
Jonas Maebe
|
69c29a415f
* pass the procdef to getintparaloc instead of only the proccalloption, so
|
12 gadi atpakaļ |
florian
|
4f30ac0247
* put records with 16 bytes size into two register on 64 bit targets if possible
|
12 gadi atpakaļ |
florian
|
c781f21a46
* records with two times the size of a register can be kept in registers
|
12 gadi atpakaļ |
florian
|
04543b179f
o merge of the branch laksen/arm-embedded of Jeppe Johansen:
|
13 gadi atpakaļ |
svenbarth
|
8e07ddb2bc
* made internal errors for M68K unique
|
13 gadi atpakaļ |
pierre
|
be5839e44a
Add warning/error message about possible problems with nostackframe modifier
|
13 gadi atpakaļ |
Jeppe Johansen
|
8b17a358e4
Remove all traces of the interrupt vector table generation mechanism
|
13 gadi atpakaļ |
pierre
|
4b7a6ecc14
move currentregloc setting to ncgutil to avoid sysym unit ependency in aasmtai unit
|
13 gadi atpakaļ |
pierre
|
fcaff0489c
* psub.pas : translate tregister for registerhi also
|
13 gadi atpakaļ |
Jonas Maebe
|
944c1172ee
* implementation of fix from r22376 without using introspection
|
13 gadi atpakaļ |