 sergei
|
3dba3a0219
* Float constants must be pooled using their the actual type, not the mapped type of assembler instruction. Otherwise e.g. extended may be mixed up with cextended, because both are mapped to ait_real_80bit.
|
14 years ago |
|
sergei
|
1470d91ba1
* Optimized pooling of floating point and large set constants using hash tables.
|
14 years ago |
 Jonas Maebe
|
025ec34e4d
+ "CExtended" type that is the same as "extended", but conforming to the
|
15 years ago |
 paul
|
b084d70ae4
compiler: cleanup - remove unused variables
|
15 years ago |
 florian
|
b5e7b3e1e7
o patch by Sergei Gorelkin which basically resolves #14308 (still misses some tests):
|
15 years ago |
|
Jonas Maebe
|
606b7f4f80
* fixed (harmless) range error in cross-endian crosscompiling code for
|
16 years ago |
|
Jonas Maebe
|
c2c68ddb8f
- removed FPC_NEW_BIGENDIAN_SETS define and if(n)defs
|
16 years ago |
|
Jonas Maebe
|
7d459cf12a
* the compiler now explicitly keeps track of the minimally guaranteed
|
16 years ago |
|
florian
|
b178b08ba7
Merged revisions 11665-11738 via svnmerge from
|
17 years ago |
|
Jonas Maebe
|
ffeba2fd10
* use LOC_REGISTER instead of LOC_CREGISTER for constant string pointers
|
17 years ago |
|
Jonas Maebe
|
3f39c95163
* fixed asd_references for ansi/widestrings on Darwin
|
17 years ago |
|
florian
|
9955d5b061
o patch from Sergej Gorelkin to improvement code generation for string literals
|
17 years ago |
 yury
|
fcceb9cfa1
* Removed/ifdefed/commented unused local variables.
|
17 years ago |
|
florian
|
0a4314206f
* range check error checking for floats refactored
|
17 years ago |
|
Jonas Maebe
|
632570fc74
* fixed compilation with 2.2.0 on big endian systems after r10432
|
17 years ago |
 peter
|
8f239d04b6
* cleanup and simplify the set type handling
|
17 years ago |
|
Jonas Maebe
|
f36e5411af
* split cpu64bit compiler define into
|
17 years ago |
|
Jonas Maebe
|
5a0a6d0d9e
* fixed and enabled smartlinking on Darwin by adding more .reference
|
18 years ago |
|
Jonas Maebe
|
70c2414daa
* split off sec_rodata_norel from sec_rodata, and only put constant data
|
18 years ago |
|
Jonas Maebe
|
0040eecf9f
+ support for extra packing of sets whose lower element number
|
18 years ago |
|
yury
|
1782305ae2
* Completely fixed float constants in softfloat mode when compiler was compiled in hardfloat mode on arm-linux host.
|
18 years ago |
|
yury
|
b86574cf36
* Fixed float constants in softfloat mode when compiler was compiled in hardfloat mode on arm-linux host.
|
18 years ago |
 daniel
|
9b50dc572b
* Change rest of compiler to system unit swapendian routines.
|
18 years ago |
|
daniel
|
9adb202a92
* Rework the constexprint to allow operations from low(int64) to high(qword).
|
18 years ago |
|
Jonas Maebe
|
628ddfa160
* fixed (harmless) range error
|
18 years ago |
|
Jonas Maebe
|
5cf2511283
- removed/disabled all normalset code (except for bootstrapping), since
|
18 years ago |
|
Jonas Maebe
|
a0b57eddb5
* new internal set format for big endian systems. Advantages:
|
18 years ago |
|
florian
|
efb025310d
* tried to restore old set behaviour for big endian targets
|
18 years ago |
|
florian
|
9a0f769b2f
* cleaned up set conversion
|
18 years ago |
|
florian
|
fbc197f27a
* fixed writing of double constants on WinCE
|
18 years ago |
