o don't try to create .class files for generic types
o still generate all JVM-specific wrappers for generic types even though they
won't be written out, because when specializing all the defid's have to
match exactly
o add synthetic routine implementations after generating the specializations,
so that the synthetic routines for those specializations are also generated
(we don't specialize generic versions of the synthetic generic routines
because it's not easy or even always possible to create valid generic
versions of synthetic routines)
o Note: these are Pascal-style generics, not Java-style generics. The generic
types nor their specializations are usable from Java code (specializations
may become usable in the future)
git-svn-id: branches/jvmbackend@19047 -
* only wrap the init/finalization code after the main program has been
completely processed (pass1, codegen), because this may influence
the required wrapping in case of the JVM target
* replace periods with slashes in the package name when constructing
the classes corresponding to the units (to execute their initialization
code) -- didn't notice this was wrong previously because due to the
incomplete wrapping, they were never being initialized until now
git-svn-id: branches/jvmbackend@18645 -
o since the JVM target has no stack/framepointer that can be passed
on to nested routines, all local variables and parameters accessed
from nested routines are grouped into a local record whose address
is passed to nested routines. The same technique is also required
for LLVM in the future
git-svn-id: branches/jvmbackend@18588 -
current class, since constructors are not automatically inherited in
Java
o tprocdef.getcopy() implementation, which returns an (unfinished) copy
of a tprocdef. Finalise by calling symcreat.finish_copied_procdef()
o made it possible to specify an existing procdef as argument to
read_proc(), in which case it won't try to parse a procedure declaration,
but only a body and associate it with the passed procdef. This is
required for the inherited constructor support, since we cannot generate
a textual representation of inherited constructors that is guaranteed to
parse in the context of the current unit (e.g., if they use types from
a unit that is not in the uses clause of the current unit)
o folded tprocsym.find_procdef_bypara_no_rettype() into
Tprocsym.Find_procdef_bypara, by interpreting specifying nil as
retdef as not having to check the return def (required to compare
parent constructors with child constructors to see whether they
match, since the returndef will always be the current class type)
git-svn-id: branches/jvmbackend@18488 -
parsing a routine if the localst is in fact a localsymtable, otherwise
they can be generated twice for for the staticsymtable (the localst
of the init/final code of a unit is the staticsymtable)
git-svn-id: branches/jvmbackend@18476 -
assignment-nodes. For global typed constants and typed constants/
local variable initialisers in regular functions/procedurs, the
assignments are performed in the unit initialisation code. For
those in object/record definitions and their methods, it's done
in the class constructor. Since we may not yet have parsed all
method implementations when the class constructor is parsed, part
of these may be initialised in a helper routine called from the
class constructor. The ones known when the class constructor is
parsed are inited there, because the ones marked as "final" and
declared as static class fields must be initialised in the class
constructor for Java
o new set systems_typed_constants_node_init in systems unit that
indicates that a target uses node trees to initialise typed consts
instead of an initialised data section
o mark typed constants in {$j-} mode as "final" for JVM
o mangle the name of staticvarsyms inside localtables a bit to avoid
name clashes (only with procedure names for now, no parameters yet
so can still cause problems with overloaded routines)
o after a routine has been parsed, it is now processed by
cnodeutils.wrap_proc_body(), which can add extra nodes before code
generation (used for injected the typed constant node trees)
git-svn-id: branches/jvmbackend@18475 -
hlcgobj so it can be overridden with target-specific code (e.g.,
the JVM target has to add initialisation code to allocate regular
arrays and records in constructors and unit/program initialisation)
git-svn-id: branches/jvmbackend@18452 -
artificially generated stuff rather than directly working with defs/syms
problems
o scanner state saving/restoring, and avoiding problems in case of
errors in the injected strings
o in case of the actual application (adding overriding constructors):
the parameters may be of types not visible in the current unit to
newly written code -> can't just use the scanner...
git-svn-id: branches/jvmbackend@18427 -
disabled it under all circumstances on platforms that use garbage
collection for managed types since it's not required there
git-svn-id: branches/jvmbackend@18418 -
nutils.pas into virtual class methods of a new tnodeutils class defined
in ngenutil (global factory: cnodeutils), so they can be overridden by
architecture-specific implementations (required by the JVM backend)
git-svn-id: branches/jvmbackend@18364 -
since the definition-specific adorning of JVM mangled names is Jasmin-
specific, and such code has no place in symdef
* moved code to adorn JVM mangled names for Jasmin definitions to agjasmin
git-svn-id: branches/jvmbackend@18346 -
o handle them like for regular classes (return a class instance,
although this is technically not true since they don't return
anything; will be changed in the future)
o because of the previous point, make sure that we handle the
"function result" properly and don't pop too many values from
the evaluation stack when calling one constructor from another
o added "extra_pre_call_code" method used by njvmcal to insert
the "new" opcode to create the new class instance before
calling a constructor
o when a constructor does not call any other constructor (inherited
or otherwise), automatically insert a call to the inherited
parameterless constructor as required by the jvm standard)
TODO: check that *if* an inherited or other constructor is called
from another constructor, that it does so as the first statement/
call
git-svn-id: branches/jvmbackend@18328 -
+ added arrayreftype field to treference for the jvm, because
array loads have to be performed using special instructions,
so this information has to be passed on to the code generator.
This information will have to be added in t(cg)vecnode.
* moved concatenating the generated code for a procedure to the
al_procedures list to thlcg.record_generated_code_for_procdef(),
which is overridden by jvm/hlcgcpu to attach that code to the
procdef it belongs with. The reason is that a Java class file
can only contain code for one class, so we have to write out
the assembler grouped per class instead of in the order the
routines appear in the source code
* also committed forgotten changes in psub after moving the
register allocator initialisation to hlcgobj
git-svn-id: branches/jvmbackend@18308 -
whether the mangled name is for defining a symbol, or for referencing
it later (e.g. for a call or load of its address). The reason is that
on the JVM both cases are different.
+ jvmdef unit to encode types according to the JVM rules
+ tprocdef.jvmmangledname() to encode a procdef's JVM mangled name
(the common part of defining/referencing it; tprocdef.mangledname
afterwards adorns it as required)
git-svn-id: branches/jvmbackend@18288 -
* made ttgobj.create virtual, added a "tgobjclass: class of ttgobj = ttgobj"
variable and use that one to instantiate new temp allocators
* created ttgjvm descendant that only allows allocations of 4 or 8 bytes
(rounding allocations < 4 bytes up to 4 bytes), and that divides the
offsets/sizes by 4 so we get stack slot numbers
git-svn-id: branches/jvmbackend@18282 -
in Windows, Emx and OS2 targets.
This fixes test failures: test/library/tlib1b.pp and lib2b.pp
link unit TLinker.AddImportSymbol: Added symmangledname parameter.
ogbase unit TImportSymbol.Create: Add AMangledName parameter.
fmodule unit TModule.AddExternalImport: Add symmangledname parameter.
fppu unit: Also put mangled name string for imported symbol.
ppu unit: Increase PPUVersion
utils/ppudump.pp: Adapt to PPU change above.
systems/T_OS units: Use ImportSymbol.MangledName property to create
import libraries or .idata sections.
git-svn-id: trunk@17804 -
* Moved nestedprocs from tcgprocinfo to tprocinfo, in order to be able to access the entire nested procedure hierarchy without depending on psub.pas or code generator.
* Creating an instance of tprocinfo automatically inserts it into list of parent's nested procedures.
* nestedprocs list is created on demand. Public read-only access is provided by has_nestedprocs and get_first_nestedproc functions.
+ Method destroy_tree is provided for destroying the entire hierarchy of procinfo's. It can be called on any procinfo object in the tree.
+ Also added methods save_jump_labels and restore_jump_labels for asmlabel maintenance, which is currently being repeatedly done all over the codegenerator.
git-svn-id: trunk@17197 -
- remove thelpersearch again => adjustments to searchsym_in_class and calls to it
- rename sto_has_classhelper to sto_has_helper
* symbase.pas: make push and pop in tsymtablestack virtual
* symdef.pas:
- add a new class tdefawaresymtablestack which overrides push and pop of tsymtablestack and adjusts the new extendeddefs field of the current tmodule
- tobjectdef.create: sto_has_classhelper => sto_has_helper
* fmodule.pas:
- add new hash object list to tmodule (key: mangled type name) which holds object list instances that contain all helpers currently active for a given type (= key of the hash list)
- tmodule.create: the hash list owns the sublists (but those don't own the defs)
- tmodule.destroy: free the hash list
* pdecobjpas:
- rename parse_extended_class to parse_extended_type
- parsing of constructors:
# for all helper types: no class constructors allowed
# for record helpers: as long as constructors for records themselves are disabled they are for record helpers as well
- object_dec: manually add the helper to the extendeddefs if the overall owner of the current def is a static symtable (implementation section or program/library main file), because the symtable won't be popped and pushed then
* parser.pas: instantiate the new stack class
* psub.pas: backup the extendeddefs in specialize_objectdefs as well
* ptype.pas:
- generate_specialization: backup the extendeddefs as well
- record_dec: _RECORD is now consumed in read_named_type itself
- read_named_type: parse "record helper" if advanced record syntax is enabled
* symtable.pas:
- correct searchsym_in_class declaration => adjustments in this unit as well
- add the possibility to pass a context def to search_last_objectpascal_helper
- rename search_objectpascal_class_helper to search_objectpascal_helper
- rename search_class_helper to search_objc_helper
- searchsym_in_class:
# search for helpers in every level of the tree of the class
# the contextclassh can also be a subclass of the extendeddef
- searchsym_in_record: search for helper methods as well
- searchsym_in_helper:
# search for symbols in class and record helpers is the same except for the search in the class' ancestors
# search the extendeddef directly and rely on searchsym_in_class only for the class' ancestors as we need to find the helpers there as well
- search_last_objectpascal_helper: use the extendeddefs list of current_module instead of scanning the symbol stack itself
* pexpr.pas: adjustments because of renaming of sto_has_classhelper
* pinline.pas: adjustment because of removing of thelpersearch
* nflw.pas:
- renamed classhelper to helperdef
- adjusted search_last_objectpascal_helper call
* msg*:
- adjusted error message for constructors in records (this currently applies to record helpers as well)
- renamed parser_e_not_allowed_in_classhelper to parser_e_not_allowed_in_helper => adjustments in code
- added parser_e_no_class_constructors_in_helpers
* pdecsub.pas: adjusted renamed error message
* htypechk.pas: check for helpers in every step of the hierarchy
* nobj.pas: search_class_helper => search_objc_helper
* utils/ppudump.pas: adjust, because of renames
Note: the define "useoldsearch" will be only used for performance comparison on my (faster) Linux machine; that (and its associated code) will be removed afterwards
git-svn-id: branches/svenbarth/classhelpers@17151 -
* generate add.w instead of add for thumb-2 in case one of the registers
is > r8
* add register interferences for the "add" instruction so the register
allocator can detect invalid instruction forms (even for assembler code)
* fixed error in thumb2.inc detected by the previous change
git-svn-id: trunk@16633 -
* Link rtti to ancestor by writing rtti of ancestor as a field of type tkObject and offset zero. This is a cheat from formal point of view (as it replaces inheritance with aggregation), but is fine for the intended purpose of representing memory layout. Now RTL can handle entire instance of descendant object, and such objects can be statically allocated without leaks.
* Bypass finalization in inherited object destructors, as the instance is now entirely finalized in outermost destructor.
+ test
git-svn-id: trunk@16632 -
- add delphi operator tokens into token enum
- move optoken search from parse_proc_dec to parse_proc_head and add delphi operator name search
- map delphi operators to existent fpc operators and skip some delphi operators for now
- implement store operators in record symtable and search in it
git-svn-id: trunk@16624 -
------------------------------------------------------------------------
r16513 compiler: add support for visibility blocks in records and type, const declarations:
- add parse_record_members function to parse record blocks based on parse_object_members code
- disable published section in records
- rename in_class argument in some functions to in_structure because the same code can work for records now which are not classes
------------------------------------------------------------------------
r16514 compiler: check visibility of record members, allow access to record consts and types:
- add searchsym_in_record function
- change is_visible_for_object to accept tabstractrecorddef instead of tobjectdef arguments because records also have visibility sections now
- change arguments in do_member_read, do_proc_call to tabstractrecorddef from tobjectdef to accept records
- rename classh arguments to structh and change their type to tabstractrecorddef to show that they can accept records now too
- move RttiName from tobjectdef to tabstractrecorddef
------------------------------------------------------------------------
r16515 compiler: add current_structdef: tabstractrecorddef and point current_objectdef to it
------------------------------------------------------------------------
r16516 compiler: allow access other record symbols than fields, first extended records tests
------------------------------------------------------------------------
r16519 compiler:
- move objname, objrealname fields from tobjectdef to tabstractrecorddef,
- load and save them from/to ppu file,
- use tabstarctrecorddef in some more places where previously code worked for tobjectdef
- change push_nested_hierarchy, pop_nested_hierarchy to handle records too
------------------------------------------------------------------------
r16526 compiler: implement record methods and class methods:
- rename tprocdef._class to tprocdef.struct and change the type from tobjectdef to tabstractrecorddef because methods can belong not to classes only now but to records too
- replace in many places use of current_objectdef to current_structdef with typcast where is needed
- add an argument to comp_expr, expr, factor, sub_expr to notify that we are searching type only symbol to solve the problem with records,objects,classes which contains fields with the same name as previosly declared type (like:
HWND = type Handle;
rec = record
hWnd: HWND;
end;)
- disable check in factor_read_id which was made for object that only static fields can be accessed as TObjectType.FieldName outside the object because it makes SizeOf(TObjectType.FieldName) imposible and since the same method was extended to handle records it also breaks a52 package compilation
- rename tcallcandidates.collect_overloads_in_class to tcallcandidates.collect_overloads_in_struct and addapt the code to handle overloads in records too
- fix searchsym_type to search also in object ancestors if we found an object symtable
- add pd_record, pd_notrecord flags to mark procedure modifies which can or can't be used with records. Disallow the next modifiers for records: abstract, dynamic, export, external, far, far16, final, forward, internconst, internproc, interrupt, message, near, override, public, reintroduce, virtual, weakexternal,
Allow the next modifiers for records: static
------------------------------------------------------------------------
r16530 compiler: fix compilation problems caused by tprocdef._class -> tprocdef.struct rename which was found by make fullcycle
------------------------------------------------------------------------
r16531 compiler: redo searchsym_type:
- remove complex condition which was used previosly to decide whether objectsymtable or recordsymtable requires a search for type or not - they require a search because contains types (although not all object types)
- don't search only for typesym but for other symbols too - for example UnitSym.TypeSym or ConstSym..ConstSym are also valid types. Skip the next symtypes during the search because they can't be used in type declaration: fieldvarsym, paravarsym, propertysym, procsym, labelsym
------------------------------------------------------------------------
r16541 compiler: move objectoptions to tabstractrecorddef because it will be needed for records too
------------------------------------------------------------------------
r16542 compiler: remove unneeded checks and typecasting caused by move of objectoptions into tabstractrecorddef
------------------------------------------------------------------------
r16543 compiler: move constructor, destructor parsers to interface section to allow future use by records + change current_objectdef to current_structdef for the same reason, make constructor return type = record for records
r16544 compiler: start parsing of record constructors and destructors:
- disallow record destructor
- raise internal error for constructor because it is not yet implemented
- handle class constructors and destructors for records
- move find_procdef_bytype to tabstractpointerdef
------------------------------------------------------------------------
r16545 compiler: fix static fields creation and access for records
+ extend test to check class constructor/destructor for records and static fields access
------------------------------------------------------------------------
r16546 tests: add a test which checks that records destructors are not allowed
------------------------------------------------------------------------
r16547 compiler: rename aclass->astruct in read_property_dec because records will have properties too
------------------------------------------------------------------------
r16548 compiler: implement properties in records:
- rename property_dec from pdecobj to struct_property_dec because pdecl also has property_dec and move it to interface to use by records + allow properties for records
- use struct_property_dec in record parser
- change structh type from objectdef to abstractrecorddef in read_property_dec to use by records
- disallow stored and default modifiers for records because records are not used for streaming
- fix misuse of search_sym_in_class for records in few places
------------------------------------------------------------------------
r16550 compiler: disallow regular class methods in records and allow only static class methods. delphi do so and it is logical because records have no inheritance. + tests
------------------------------------------------------------------------
r16560 compiler: a trial to implement record constructor
- map self to constructor result
- don't push vmt for records
At the moment generated assembler has errors although node tree is correct
------------------------------------------------------------------------
r16561 compiler:
- add mode switch extendedrecords to allow/disallow extended records syntax, add this mode switch to delphi mode by default
- disable/enable extended records parsing based on extendedreocrds mode switch
------------------------------------------------------------------------
r16562 compiler: fix record constructor return type when parsing procedure declaration
------------------------------------------------------------------------
r16568 compiler: partly revert r16560 and disable (at least temporary) constructors in records
------------------------------------------------------------------------
r16569 test: add records Self test
------------------------------------------------------------------------
r16570 compiler: generate debug info for record methods:
- extended write_symtable_procdefs to handle record definitions
- fix stabs info generator to use prefixed symbol names for record methods, also handle records static symbols the same way as for object types
- fix dwarf info record generation: add visibility info and write methods, also prefix record members the same way as object members
git-svn-id: trunk@16574 -
- map self to constructor result
- don't push vmt for records
At the moment generated assembler has errors although node tree is correct
git-svn-id: branches/paul/extended_records@16560 -
- rename tprocdef._class to tprocdef.struct and change the type from tobjectdef to tabstractrecorddef because methods can belong not to classes only now but to records too
- replace in many places use of current_objectdef to current_structdef with typcast where is needed
- add an argument to comp_expr, expr, factor, sub_expr to notify that we are searching type only symbol to solve the problem with records,objects,classes which contains fields with the same name as previosly declared type (like:
HWND = type Handle;
rec = record
hWnd: HWND;
end;)
- disable check in factor_read_id which was made for object that only static fields can be accessed as TObjectType.FieldName outside the object because it makes SizeOf(TObjectType.FieldName) imposible and since the same method was extended to handle records it also breaks a52 package compilation
- rename tcallcandidates.collect_overloads_in_class to tcallcandidates.collect_overloads_in_struct and addapt the code to handle overloads in records too
- fix searchsym_type to search also in object ancestors if we found an object symtable
- add pd_record, pd_notrecord flags to mark procedure modifies which can or can't be used with records. Disallow the next modifiers for records: abstract, dynamic, export, external, far, far16, final, forward, internconst, internproc, interrupt, message, near, override, public, reintroduce, virtual, weakexternal,
Allow the next modifiers for records: static
git-svn-id: branches/paul/extended_records@16526 -
