* The whilen and repeatn are now completely unified into whilerepeatn. This
allows the state tracker to change while nodes automatically into
repeat nodes.
* Resulttypepass improvements to the notn. 'not not a' is optimized away and
'not(a>b)' is optimized into 'a<=b'.
* Resulttypepass improvements to the whilerepeatn. 'while not a' is optimized
by removing the notn and later switchting the true and falselabels. The
same is done with 'repeat until not a'.
* foreach gets extra argument for passing local data to the
iterator function
* -CR checks also class typecasts at runtime by changing them
into as
* fixed compiler to cycle with the -CR option
* fixed stabs with elf writer, finally the global variables can
be watched
* removed a lot of routines from cga unit and replaced them by
calls to cgobj
* u32bit-s32bit updates for and,or,xor nodes. When one element is
u32bit then the other is typecasted also to u32bit without giving
a rangecheck warning/error.
* fixed pascal calling method with reversing also the high tree in
the parast, detected by tcalcst3 test
+ first parameter offset in stack now portable
* rename some constants
+ move some cpu stuff to other units
- remove unused constents
* fix stacksize for some targets
* fix generic size problems which depend now on EXTEND_SIZE constant
error was found
* generic changeregsize(reg,size) for i386 register resizing
* removed some more routines from cga unit
* fixed returnvalue handling
* fixed default stacksize of linux and go32v2, 8kb was a bit small :-)
* LOC_CONSTANT added for better constant handling
* secondadd splitted in multiple routines
* location_force_reg added for loading a location to a register
of a specified size
* secondassignment parses now first the right and then the left node
(this is compatible with Kylix). This saves a lot of push/pop especially
with string operations
* adapted some routines to use the new cg methods
* register allocation is now handled by a class and is mostly processor
independent (+rgobj.pas and i386/rgcpu.pas)
* temp allocation is now handled by a class (+tgobj.pas, -i386\tgcpu.pas)
* some small improvements and fixes to the optimizer
* some register allocation fixes
* some fpuvaroffset fixes in the unary minus node
* push/popusedregisters is now called rg.save/restoreusedregisters and
(for i386) uses temps instead of push/pop's when using -Op3 (that code is
also better optimizable)
* fixed and optimized register saving/restoring for new/dispose nodes
* LOC_FPU locations now also require their "register" field to be set to
R_ST, not R_ST0 (the latter is used for LOC_CFPUREGISTER locations only)
- list field removed of the tnode class because it's not used currently
and can cause hard-to-find bugs
longer longint($ffffffff), but just $fffffff in psystem)
* small additional fix in 64bit rangecheck code generation for 32 bit
processors
* adaption of ranges required the matching talgorithm used for selecting
which overloaded procedure to call to be adapted. It should now always
select the closest match for ordinal parameters.
+ inttostr(qword) in sysstr.inc/sysstrh.inc
+ abs(int64), sqr(int64), sqr(qword) in systemh.inc/generic.inc (previous
fixes were required to be able to add them)
* is_in_limit() moved from ncal to types unit, should always be used
instead of direct comparisons of low/high values of orddefs because
qword is a special case
custom resulttype
* compilerproc typeconversions now set the resulttype from the type
conversion for the generated call node, because the resulttype of
of the compilerproc helper isn't always exact (e.g. the ones that
return shortstrings, actually return a shortstring[x], where x is
specified by the typeconversion node)
* ti386callnode.pass_2 now always uses resulttype instead of
procsym.definition.rettype (so the custom resulttype, if any, is
always used). Note that this "rettype" stuff is only for use with
compilerprocs.
- removed all code from the compiler which has been replaced by
compilerproc implementations (using {$ifdef hascompilerproc} is not
necessary in the compiler)
resulttype and first pass)
* made handling of read(ln)/write(ln) processor independent
* moved processor independent handling for str and reset/rewrite-typed
from firstpass to resulttype pass
* changed names of helpers in text.inc to be generic for use as
compilerprocs + added "iocheck" directive for most of them
* reading of ordinals is done by procedures instead of functions
because otherwise FPC_IOCHECK overwrote the result before it could
be stored elsewhere (range checking still works)
* compilerprocs can now be used in the system unit before they are
implemented
* added note to errore.msg that booleans can't be read using read/readln
name to lowercase(declaration_name). This prevents a normal user from
accessing the routine, but they can still be easily looked up within
the compiler. This is used for helper procedures and should facilitate
the writing of more processor independent code in the code generator
itself (mostly written by Peter)
+ new "createintern" constructor for tcal nodes to create a call to
helper exported using the "compilerproc" directive
+ support for high(dynamic_array) using the the above new things
+ definition of 'HASCOMPILERPROC' symbol (to be able to check in the
compiler and rtl whether the "compilerproc" directive is supported)
* -Oa argument supports more alignment settings that can be specified
per type: PROC,LOOP,VARMIN,VARMAX,CONSTMIN,CONSTMAX,RECORDMIN
RECORDMAX,LOCALMIN,LOCALMAX. It is possible to set the mimimum
required alignment and the maximum usefull alignment. The final
alignment will be choosen per variable size dependent on these
settings
