fpc/compiler
2018-11-11 17:32:21 +00:00
..
aarch64 - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
arm + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
avr * max_operands needs only to be 2 on avr 2018-11-03 10:39:58 +00:00
generic
html
i386 * redundant code removed, work is meanwhile done by aasmcpu.optimize_ref 2018-11-11 17:32:21 +00:00
i8086 - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
jvm - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
llvm * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
m68k - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
mips - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
msg * Changed verbosity of ""???" not yet supported inside inline procedure/function" to note to be consistent with note "Call to subroutine "?" marked as inline is not inlined". 2018-11-07 12:57:23 +00:00
powerpc - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
powerpc64 - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
ppcgen * GAS readers and writers can handle section flags 2018-10-20 18:23:24 +00:00
riscv * fixed compilation with -O3 (one false positive, one real error) 2018-11-01 20:39:38 +00:00
riscv32 - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
riscv64 - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
sparc * split create_paraloc_info_intern for sparc32 and sparc64 2017-08-20 15:21:51 +00:00
sparc64 * keep track of the temp position separately from the offset in references, 2018-04-22 17:03:16 +00:00
sparcgen - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
systems * define renamed into tls_threadvars 2018-11-07 22:03:04 +00:00
utils + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
x86 + support for tlsm_general on i386-linux 2018-11-11 17:32:20 +00:00
x86_64 - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
aasmbase.pas + support exporting labels from asm blocks in intel syntax asm blocks via the 2017-10-30 16:44:13 +00:00
aasmcnst.pas * Removed unused local vars. 2018-11-02 18:44:29 +00:00
aasmdata.pas * fixed some issues to get correct values of section_count, works better than before but it is still not perfect for unknown reasons 2018-09-12 20:33:18 +00:00
aasmdef.pas
aasmsym.pas
aasmtai.pas + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
aggas.pas + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
aopt.pas * removed unused units 2017-05-09 19:53:14 +00:00
aoptbase.pas + define cpudelayslot: set during compiler compilation for CPUs having branch instructions with delay slot (MIPS, SPARC) 2017-08-20 17:20:38 +00:00
aoptda.pas * removed unused units 2017-05-09 19:53:14 +00:00
aoptobj.pas - get rid of MaxOps, it is redundant with max_operands 2018-11-02 21:32:29 +00:00
aoptutils.pas Fix 3 operand version of MatchOpType 2018-11-02 23:18:54 +00:00
assemble.pas + new fields for more precise control of alignment 2018-11-01 20:49:17 +00:00
blockutl.pas * removed unused units 2017-05-09 19:53:14 +00:00
browcol.pas
catch.pas
ccharset.pas
cclasses.pas * Rearranged the code to place implementation of inline methods before their usage within this unit. 2018-11-05 13:59:59 +00:00
cfidwarf.pas
cfileutl.pas * Removed inline for Unix2AmigaPath since it is not necessary and can't be inlined anyway due to ansistring conversion. 2018-11-05 14:37:01 +00:00
cg64f32.pas
cgbase.pas + i386-linux support for tls-based threadvars 2018-11-07 22:03:02 +00:00
cghlcpu.pas
cgobj.pas + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
cgutils.pas Added RiscV32/64 target, from a cleaned up version of branches/laksen/riscv/trunk. 2018-07-20 08:21:15 +00:00
cmsgs.pas + implemented support for codepage aware compiler messages. It can be enabled 2017-06-08 16:11:33 +00:00
comphook.pas
compiler.pas * print used unit scopes 2018-05-07 19:48:26 +00:00
compinnr.pas + inline support for the x86 'in' and 'out' instructions. Currently only enabled 2018-07-03 17:01:42 +00:00
comprsrc.pas renamed script unit to cscript to avoid namespace conflict with univint package on Darwin. this makes it possible to build the textmode IDE on Darwin 2017-11-17 11:27:19 +00:00
constexp.pas
COPYING.txt
cpid.pas
crefs.pas
cresstr.pas * fixed some issues to get correct values of section_count, works better than before but it is still not perfect for unknown reasons 2018-09-12 20:33:18 +00:00
cscript.pas * quote assembly filenames in Unix script, fixes sh errors with smartlinked small filenames like unitname(1), etc 2018-01-06 10:17:18 +00:00
cstreams.pas * broken compilation after r39347 fixed 2018-06-30 12:30:25 +00:00
cutils.pas + Added overloaded version of stringdup() for ansistring to allow stringdup to be inlined. 2018-11-05 14:00:51 +00:00
cwindirs.pp
dbgbase.pas * revert r37968: it does more harm than good 2018-01-15 16:41:22 +00:00
dbgcodeview.pas + added new debug output option -godwarfomflinnum for generating line number 2018-05-17 14:38:50 +00:00
dbgdwarf.pas Use correct field for sl_absolutetype or sl_typeconv ppropaccesslistitem type (revealed by compilation with -CriotR) 2018-10-31 14:43:18 +00:00
dbgstabs.pas * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
dbgstabx.pas
defcmp.pas * make "formal const/var" parameters the least preferred conversion 2018-10-21 21:02:17 +00:00
defutil.pas * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
dirparse.pas
elfbase.pas * The MaybeSwapXXXHeader functions moved from ogelf to the elfbase unit, so they 2018-06-14 13:01:08 +00:00
entfile.pas More -CriotR fixes: 2018-11-01 21:58:54 +00:00
export.pas * removed unused units 2017-05-09 19:53:14 +00:00
expunix.pas * removed unused units 2017-05-09 19:53:14 +00:00
finput.pas
fmodule.pas * removed unused units 2017-05-09 19:53:14 +00:00
fpccrc.pas
fpcdefs.inc * Enabled safecall support for aarch64 to be on par with other cpus. 2018-10-08 08:37:18 +00:00
fpcp.pas * removed unused units 2017-05-09 19:53:14 +00:00
fpkg.pas
fppu.pas * keep track if the unit was found by using a default namespace as prefix 2018-05-05 17:18:39 +00:00
gendef.pas Add obsolete prefix to cpu_ia64, system_ia64_win64 and system_ia64_embedded, ia64 specfic code has been removed in rev 30836 dated 2015-05-09 2017-08-04 08:54:34 +00:00
globals.pas + tsettings.tlsmodel 2018-11-07 22:03:01 +00:00
globstat.pas
globtype.pas + tsettings.tlsmodel 2018-11-07 22:03:01 +00:00
hlcg2ll.pas + implementation of the vectorcall calling convention by J. Gareth Moreton 2018-02-11 17:50:37 +00:00
hlcgobj.pas * removed temppos field again from parameter locations: they're not allocated 2018-04-27 19:18:55 +00:00
htypechk.pas * Further improvement for r40180: 2018-11-04 15:37:52 +00:00
impdef.pas
import.pas * removed unused units 2017-05-09 19:53:14 +00:00
ldscript.pas
link.pas * do not use a wildcard while deleting smartlinked objects after linking stage in a script, conflicts with script filename quoting on Unix at least 2018-01-06 10:19:21 +00:00
macho.pas
machoutils.pas * Addition to r40225: Forgot to remove inline in the implementation. 2018-11-05 14:46:09 +00:00
Makefile Add explicit rtlclean/rtl targets in fullcycle rule if DOWPOCYCLE is set 2018-10-31 15:50:26 +00:00
Makefile.fpc Add explicit rtlclean/rtl targets in fullcycle rule if DOWPOCYCLE is set 2018-10-31 15:50:26 +00:00
MPWMake
msgidx.inc * Changed verbosity of ""???" not yet supported inside inline procedure/function" to note to be consistent with note "Call to subroutine "?" marked as inline is not inlined". 2018-11-07 12:57:23 +00:00
msgtxt.inc * Changed verbosity of ""???" not yet supported inside inline procedure/function" to note to be consistent with note "Call to subroutine "?" marked as inline is not inlined". 2018-11-07 12:57:23 +00:00
nadd.pas * Removed unused local vars. 2018-11-02 18:44:29 +00:00
nbas.pas * forgot to commit extension to tspecializenode 2018-09-21 15:17:52 +00:00
ncal.pas + Added the pio_inline_not_possible flag to timplprocoption. It is set if a function can't be inline (contains an assembler block, etc). When calling such functions, no note "Call to subroutine "?" marked as inline is not inlined" is emitted. 2018-11-07 12:20:30 +00:00
ncgadd.pas * removed unused units 2017-05-09 19:53:14 +00:00
ncgbas.pas * fixed some issues to get correct values of section_count, works better than before but it is still not perfect for unknown reasons 2018-09-12 20:33:18 +00:00
ncgcal.pas o fix several issues with floating point exceptions 2018-08-19 15:26:00 +00:00
ncgcnv.pas * keep track of the temp position separately from the offset in references, 2018-04-22 17:03:16 +00:00
ncgcon.pas * removed unused units 2017-05-09 19:53:14 +00:00
ncgflw.pas + new fields for more precise control of alignment 2018-11-01 20:49:17 +00:00
ncghlmat.pas
ncginl.pas ncginl: another approach to fix the same issue #39184 supposed to address. keep left.resultdef, but try to make the target def always ordinal of an equal size, so it ends up in a data register on m68k 2018-06-07 06:43:02 +00:00
ncgld.pas + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
ncgmat.pas + support mmx shifting 2018-02-27 21:40:12 +00:00
ncgmem.pas * use sizeuinttype, instead of ptruinttype for the indexdef conversion in 2018-07-19 15:27:14 +00:00
ncgnstfl.pas
ncgnstld.pas * moved nf_typedaddr to addrnodeflags (anf_typedaddr) 2018-04-03 16:41:01 +00:00
ncgnstmm.pas * moved nf_typedaddr to addrnodeflags (anf_typedaddr) 2018-04-03 16:41:01 +00:00
ncgobjc.pas * removed unused units 2017-05-09 19:53:14 +00:00
ncgopt.pas * removed unused units 2017-05-09 19:53:14 +00:00
ncgrtti.pas * Moved inline methods to the beginning in order to allow them to be actually inlined in this unit. 2018-11-04 19:50:09 +00:00
ncgset.pas * broken compilation after r39347 fixed 2018-06-30 12:30:25 +00:00
ncgutil.pas * aarch64: Fixed loading of a function result which is returned in 4 32-bit MM registers to 2 64-bit int registers. Bug #30329. 2018-10-17 11:41:34 +00:00
ncgvmt.pas * move getrecorddef and gettabledef from TVMTWritter to symdef as they might be used from the RTTI as well 2018-08-29 19:21:23 +00:00
ncnv.pas * handle widechar constants directly in the scanner, instead of in the 2018-10-21 17:34:00 +00:00
ncon.pas * broken compilation after r39347 fixed 2018-06-30 12:30:25 +00:00
nflw.pas * ttryfinallynode now inherits from tbinarynode instead of tloopnode, as its 2018-11-01 13:19:33 +00:00
ngenutil.pas * fixed execution order of implicit finalization and class destructors: first 2018-04-08 15:41:00 +00:00
ngtcon.pas More -CriotR fixes: 2018-11-01 21:58:54 +00:00
ninl.pas Implement mark_write override for tinilinenode 2018-11-08 20:18:30 +00:00
nld.pas + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
nmat.pas * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
nmem.pas * accept dereferences in generics definitions, resolves #33700 properly 2018-07-13 21:00:33 +00:00
nobj.pas * removed unused units 2017-05-09 19:53:14 +00:00
nobjc.pas * broken compilation after r39347 fixed 2018-06-30 12:30:25 +00:00
node.pas * allow nil to be assigned to generic types, resolves #34037 2018-10-14 07:38:13 +00:00
nopt.pas * moved nf_typedaddr to addrnodeflags (anf_typedaddr) 2018-04-03 16:41:01 +00:00
nset.pas * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
nutils.pas * ttryfinallynode now inherits from tbinarynode instead of tloopnode, as its 2018-11-01 13:19:33 +00:00
objcasm.pas
objcdef.pas * removed unused units 2017-05-09 19:53:14 +00:00
objcgutl.pas * broken compilation after r39347 fixed 2018-06-30 12:30:25 +00:00
objcutil.pas * removed unused units 2017-05-09 19:53:14 +00:00
ogbase.pas * Removed unused local vars. 2018-11-02 18:44:29 +00:00
ogcoff.pas * changed type used for section alignment from byte/shortint to longint, so the 2017-10-25 15:47:29 +00:00
ogelf.pas * The MaybeSwapXXXHeader functions moved from ogelf to the elfbase unit, so they 2018-06-14 13:01:08 +00:00
oglx.pas
ogmacho.pas * changed type used for section alignment from byte/shortint to longint, so the 2017-10-25 15:47:29 +00:00
ogmap.pas * removed unused units 2017-05-09 19:53:14 +00:00
ognlm.pas * changed type used for section alignment from byte/shortint to longint, so the 2017-10-25 15:47:29 +00:00
ogomf.pas * more compilation fixes 2018-06-30 12:48:54 +00:00
omfbase.pas More -CriotR fixes: 2018-11-01 21:58:54 +00:00
optbase.pas Complement commit 40104, by changing type of executionweight in toptinfo record and adapt pass_2 code 2018-10-31 15:47:53 +00:00
optconstprop.pas * use an enum instead of integer constants to represent inline numbers 2017-05-10 14:41:43 +00:00
optcse.pas * correctly take care of short boolean evaluation when rearranging nodes during CSE 2018-09-25 20:31:03 +00:00
optdead.pas * removed unused units 2017-05-09 19:53:14 +00:00
optdeadstore.pas * removed unused units 2017-05-09 19:53:14 +00:00
optdfa.pas * wrapcomplexinlinepara takes care of function results which need different temp. handling, resolves #34438 2018-11-03 17:31:10 +00:00
options.pas + tsettings.tlsmodel 2018-11-07 22:03:01 +00:00
optloadmodifystore.pas + enabled the load-modify-store optimization for the double argument version 2017-07-21 13:17:18 +00:00
optloop.pas * do not unroll loops if the counter variable is accessed outside the current scope 2018-06-03 15:12:48 +00:00
opttail.pas
optutils.pas Use longint type instead of AWord for Initial parameter in CalcExecutionWeigths (to avoid range error for avr compiler) 2018-10-31 10:21:51 +00:00
optvirt.pas * removed unused units 2017-05-09 19:53:14 +00:00
owar.pas
owbase.pas
owomflib.pas More -CriotR fixes: 2018-11-01 21:58:54 +00:00
parabase.pas * removed temppos field again from parameter locations: they're not allocated 2018-04-27 19:18:55 +00:00
paramgr.pas * broken compilation after r39347 fixed 2018-06-30 12:30:25 +00:00
parser.pas renamed script unit to cscript to avoid namespace conflict with univint package on Darwin. this makes it possible to build the textmode IDE on Darwin 2017-11-17 11:27:19 +00:00
pass_1.pas * removed unused units 2017-05-09 19:53:14 +00:00
pass_2.pas Complement commit 40104, by changing type of executionweight in toptinfo record and adapt pass_2 code 2018-10-31 15:47:53 +00:00
pbase.pas * a symbol might be specified using a partial namespace path as long as the missing part is a namespace symbol, 2018-05-05 17:18:45 +00:00
pcp.pas
pdecl.pas * when creating a unique type alias for an object, class or interface, 2018-01-07 12:25:16 +00:00
pdecobj.pas + add support for threadvars inside classes 2018-06-23 13:49:12 +00:00
pdecsub.pas * Removed unused local vars. 2018-11-02 18:44:29 +00:00
pdecvar.pas * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
pexports.pas
pexpr.pas * handle widechar constants directly in the scanner, instead of in the 2018-10-21 17:34:00 +00:00
pgentype.pas
pgenutil.pas * if the generic dummy symbol is a procsym (with procdefs added) then this is the symbol that should be resolved to 2018-06-15 13:49:21 +00:00
pinline.pas * fix for Mantis #34333: improve error output for incorrect calls to Copy() 2018-09-26 21:33:51 +00:00
pkgutil.pas * check that the PPU version of a unit loaded for package adjustment matches exactly (we don't accept different versions when loading PPUs normally, so why should we with packages?) 2018-01-14 21:16:37 +00:00
pmodules.pas * two comment spelling fixes. Mantis #34092 & #34097 2018-08-08 08:39:20 +00:00
pp.lpi
pp.pas Added RiscV32/64 target, from a cleaned up version of branches/laksen/riscv/trunk. 2018-07-20 08:21:15 +00:00
pparautl.pas + introduce a new method tparamanager.asm_result_var(), which determines whether 2018-02-15 17:41:07 +00:00
ppc68k.lpi
ppc8086.lpi * use an enum instead of integer constants to represent inline numbers 2017-05-10 14:41:43 +00:00
ppcaarch64.lpi
ppcarm.lpi * version update 2018-01-31 18:52:18 +00:00
ppcavr.lpi + version update 2018-01-31 18:46:23 +00:00
ppcjvm.lpi
ppcmips.lpi
ppcmipsel.lpi
ppcppc64.lpi
ppcppc.lpi * version update 2018-10-28 16:21:06 +00:00
ppcriscv32.lpi * RISC-V 32 compilation fixed 2018-07-26 19:18:47 +00:00
ppcriscv64.lpi * fix assembling with official binutils 2018-07-22 13:10:24 +00:00
ppcsparc64.lpi + basics for sparc64 support, we continue to use the name "sparc" for sparc32, both sparc are identified by sparcgen (dirs, defines etc.) 2017-05-30 21:17:17 +00:00
ppcsparc.lpi + basics for sparc64 support, we continue to use the name "sparc" for sparc32, both sparc are identified by sparcgen (dirs, defines etc.) 2017-05-30 21:17:17 +00:00
ppcx64.lpi * made lazarus project names more systematic 2017-12-28 14:34:04 +00:00
ppcx64llvm.lpi * made lazarus project names more systematic 2017-12-28 14:34:04 +00:00
ppheap.pas
ppu.pas * ttryfinallynode now inherits from tbinarynode instead of tloopnode, as its 2018-11-01 13:19:33 +00:00
procinfo.pas + initial work for tls-based threadvar support on arm-linux 2018-11-07 22:02:58 +00:00
pstatmnt.pas * also pass the pattern with which the potential unit/namespace symbol had been found to try_consume_unitsym*() 2018-05-05 17:18:31 +00:00
psub.pas * non-linux arm targets fixed 2018-11-08 19:57:30 +00:00
psystem.pas * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
ptconst.pas * removed unused units 2017-05-09 19:53:14 +00:00
ptype.pas * converted Boolean8 to an internal type, and mapped Boolean to the 2018-10-16 21:14:18 +00:00
raatt.pas * GAS readers and writers can handle section flags 2018-10-20 18:23:24 +00:00
rabase.pas
rasm.pas * forgotten comment 2018-02-06 21:38:51 +00:00
rautils.pas Write real atomic operations, and add memory barrier operations. 2018-07-29 16:43:09 +00:00
README.txt
rescmn.pas
rgbase.pas
rgobj.pas * Removed unused local vars. 2018-11-02 18:44:29 +00:00
scandir.pas + basic infrastructure to generate code for floating point exception 2018-08-19 10:54:45 +00:00
scanner.pas + tsettings.tlsmodel 2018-11-07 22:03:01 +00:00
switches.pas * allow $MinEnumSize, $PackSet and $PackRecords to be used with $Push and $Pop 2018-06-12 19:40:45 +00:00
symbase.pas + add a comment to the destructor of tsymtable so that others (and maybe me a well) will hopefully see that the destructor of that class is not completely trivial! 2018-08-31 19:16:48 +00:00
symconst.pas + Added the pio_inline_not_possible flag to timplprocoption. It is set if a function can't be inline (contains an assembler block, etc). When calling such functions, no note "Call to subroutine "?" marked as inline is not inlined" is emitted. 2018-11-07 12:20:30 +00:00
symcreat.pas * moved nf_typedaddr to addrnodeflags (anf_typedaddr) 2018-04-03 16:41:01 +00:00
symdef.pas * Removed inline for methods which are too big. 2018-11-05 14:33:39 +00:00
symsym.pas * do not unroll loops if the counter variable is accessed outside the current scope 2018-06-03 15:12:48 +00:00
symtable.pas * fix big endian targets by using an explicit assignment with typecast instead of an absolute variable 2018-09-10 21:12:29 +00:00
symtype.pas + add method has_non_trivial_init_child() to tdef which provides access to the symtable's 2018-06-20 19:00:11 +00:00
symutil.pas * removed unused units 2017-05-09 19:53:14 +00:00
syscinfo.pas Add arm-palmos target to list supporting po_syscall 2017-09-15 20:25:59 +00:00
systems.inc + added support for x86_64-android target. 2018-10-17 16:56:27 +00:00
systems.pas + new fields for more precise control of alignment 2018-11-01 20:49:17 +00:00
tgobj.pas Downgrade EXTDEBUG warning to note about zero size temp, as it is used for empty sets 2018-11-08 20:19:54 +00:00
tokens.pas + implementation of the vectorcall calling convention by J. Gareth Moreton 2018-02-11 17:50:37 +00:00
verbose.pas + implemented support for codepage aware compiler messages. It can be enabled 2017-06-08 16:11:33 +00:00
version.pas Merged riscv_new branch 2018-09-26 19:49:08 +00:00
widestr.pas
wpo.pas
wpobase.pas
wpoinfo.pas * removed unused units 2017-05-09 19:53:14 +00:00

This directory contains the sources of the Free Pascal Compiler

If you want to compile/modify the compiler, please read first the
programmers manual.

To recompile the compiler, you can use the batch files :
 + mppc386.bat    if you want to build a cross compiler from i386 to m68k
 + mppcsparc      if you want to build a cross compiler from i386 to SPARC
 
 or
Use the make utility as following
  
      make OS_TARGET="compiler OS target" \
      CPU_TARGET="compiler CPU target" \
      FPCCPUOPT="Optimization level" \
      PP="compiler used to compile FPC" \
      COMPILER_OPTIONS="Options passed to compiler" \
      
      
If an option is omitted, then target CPU/OS will be same as current CPU/OS
 
Possibles targets are : linux go32v2 win32 os2 freebsd beos netbsd amiga haiku
atari sunos qnx netware openbsd wdosx palmos macos macosx emx
   
Possible compiler switches (* marks a currently required switch):
  -----------------------------------------------------------------
  GDB*                support of the GNU Debugger
  I386                generate a compiler for the Intel i386+
  x86_64              generate a compiler for the AMD x86-64 architecture
  M68K                generate a compiler for the M68000
  SPARC               generate a compiler for SPARC
  POWERPC             generate a compiler for the PowerPC
  VIS                 generate a compile for the VIS
  DEBUG               version with debug code is generated
  EXTDEBUG            some extra debug code is executed
  SUPPORT_MMX         only i386: releases the compiler switch
                      MMX which allows the compiler to generate
                      MMX instructions
  EXTERN_MSG          Don't compile the msgfiles in the compiler, always
                      use external messagefiles, default for TP
  NOAG386INT          no Intel Assembler output
  NOAG386NSM          no NASM output
  NOAG386BIN          leaves out the binary writer, default for TP
  NORA386DIR          No direct i386 assembler reader
  TEST_GENERIC        Test Generic version of code generator
                      (uses generic RTL calls)
  -----------------------------------------------------------------
  cpuflags            The target processor has status flags (on by default)
  cpufpemu            The target compiler will also support emitting software
                       floating point operations
  cpu64bitaddr        The targets use a 64-bit address space (pointers and
                       the default integer type are 64 bit)
  cpu64bitalu         The target cpu has 64-bit registers available (unless
                       cpu64bitaddr is also defined, pointers and default
                       integer type remain 32 bit, but the cpu can perform
                       64 bit calculations directly without needing helpers)
  -----------------------------------------------------------------

  Required switches for a i386 compiler be compiled by Free Pascal Compiler:
  GDB;I386

to build a compiler to SPARC target using a Win32/i386 you just use :
      make CPU_TARGET=SPARC