Like the symbols and the VM functions this is data that is static from startup until shutdown and has no need to be subjected to garbage collection. All things combined this reduces the amount of GC-sensitive objects at startup from 9600 to 600.
Making this an object had little to no advantage, except being able to remove the deleter code. Now, with some of the class data already being allocated in a memory arena so that freeing it is easier, this can also be used for the drop item lists which makes it unnecessary to subject them to the GC. This also merges the memory arenas for VM functions and flat pointers because both get deleted at the same time so they can share the same one.
Now all actors have the same metaclass and therefore it will always be the same size which will finally allow some needed changes to the type system which couldn't be done because it was occasionally necessary to replace tentatively created classes due to size mismatches.
There's simply never enough of them and they are used far too infrequently to justify the hassle of tagging along two TMaps per class.
For what they provide, single global lists that handle all player classes at once are fully sufficient.
The goal is to get rid of PClassPlayerPawn and PClassInventory so that the old assumption that all actor class descriptors have the same size can be restored
This is important to remove some code that seriously blocks optimization of the type table because that can only be done if types do not need to be replaced.
Symbols are very easy to manage once they are in a symbol table and there's lots of them so this reduces the amount of work the GC needs to do quite considerably.
After cleaning out compile-time-only symbols there will still be more than 2000 left, one for each function and one for each member variable of a class or struct.
This means more than 2000 object that won't need to tracked constantly by the garbage collector.
Note that loose fields which do occur during code generation will be GC'd just as before.
- removed all pointer declarations to types from the symbols. All types must be placed into the type table which means that they can be considered static.
As it stood, just compiling the internal ZScript code created more than 9000 DObjects, none of which really need to be subjected to garbage collection, aside from allowing lazy deallocation.
This puts an incredible drag on the garbage collector which often needs several minutes to finish processing before actual deletion can start.
The VM functions with roughly 1800 of these objects were by far the easiest to refactor so they are now. They also use a memory arena now which significantly reduces their memory footprint.
- use a memory arena to store flat pointers so that the messed up cleanup can be avoided by deallocating this in bulk.
- added a new SO opcode to the VM to execute a write barrier. This is necessary for all objects that are not linked into one global table, i.e. everything except thinkers and class types.
- always use the cheaper LOS opcode for reading pointers to classes and defaults because these cannot be destroyed during normal operation.
- removed the pointless validation from String.Mid. If the values are read as unsigned the internal validation of FString::Mid will automatically ensure proper results.
- removed some code repetition by inherit all variable types which reference a PField for a variable offset from a base class so that PField replacements can be done with one set of code.
error: use of undeclared identifier 'op'
error: no matching function for call to 'ListEnd'
error: no matching function for call to 'ListGetInt'
error: no matching function for call to 'ListGetDouble'
...
This isn't done for register based variables so for consistency it should not be done for stack based variables, too.
This difference in handling made it impossible to check the target of a hitscan attack if it was destroyed by getting damaged.
This function will truncate everything that is larger than LONG_MAX or smaller than LONG_MIN to fit into a long variable, but longs are 32 bit on Windows and 64 bit elsewhere, so to ensure consistency and the ability to parse larger values better use strtoll which does not truncate 32 bit values.
The general rule is as follows: A class name as a string will always be looked up fully, even if the class name gets shadows by another variable because strings are not identifiers.
It is only class names as identifiers that must obey the rule that if it is not known yet or hidden by something else that it may not be found to ensure that the older variable does not take over the name if it gets reused.
If a later module reused an existing name for a different class or struct type, this new name would completely shadow the old one, even in the base files.
Changed it so that each compilation unit (i.e. each ZScript and DECORATE lump) get their own symbol table and can only see the symbol tables that got defined in lower numbered resource files so that later definitions do not pollute the available list of symbols when running the compiler backend and code generator - which happens after everything has been parsed.
Another effect of this is that a mod that reuses the name of an internal global constant will only see its own constant, again reducing the risk of potential errors in case the internal definitions add some new values.
Global constants are still discouraged from being used because what this does not and can not handle is the case that a mod defines a global constant with the same name as a class variable. In such a case the class variable will always take precedence for code inside that class.
Note that the internal struct String had to be renamed for this because the stricter checks did not let the type String pass on the left side of a '.' anymore.
- made PEnum inherit from PInt and not from PNamedType.
The old inheritance broke nearly every check for integer compatibility in the compiler, so this hopefully leads to a working enum implementation.
- load internal shaders only from file 0. This does not contain aborts, like most of the other checks,but it will now refuse to load any core shader file from anything but gzdoom.pk3.
- fixed the return type checks in CallStateChain. These made some bogus assumptions about what return prototypes to support and would have skipped any multi-return function whose first argument was actually usable.
This should for now conclude actor class scriptification. The remaining ten classes with the exception of MorphedMonster are all too essential or too closely tied to engine feature so they should remain native.
* completely scriptified DehackedPickup and FakeInventory.
* scriptified all remaining virtual functions of Inventory, so that its inheritance is now 100% script-side.
* scriptified CallTryPickup and most of the code called by that.
- fixed: Passing local variables by reference did not work in the VM.
- ensure that actor defaults contain a valid virtual table and class pointer so that they can actually use virtual and class-dependent method functions. This is needed for retrieving script variables from them.
- added some helpers to set scripted member variables through the native property parser.
Unfortunately some classes, e.g. PowerMorph, MorphProjectile and the powerup contain some that cannot be handled through the 'property' definition on the script side so they need to be done from the native side.
- improved the class pointer to string cast to print the actual type it describes and not the class pointer's own type.
- fixed: The 'is' operator created non-working code when checking the inheritance of a class pointer, it only worked for objects.
This revealed an interesting bug: When the berserk fadout formula was changed in 2005 the result was essentially broken, resulting in values around 7000 - it only worked by happenstance because the lower 8 bits of the resulting values just happened to work to some degree and never overflowed. But the resulting fade was far too weak and a slightly different handling of the color composition code for the VM made it break down entirely.
This restores the pre-2005 formula but weakened intensity which now comes a lot closer to how it is supposed to look.
- support transient object member variables for information that does not need to be put in a savegame.
- fixed: special initialization of objects needs to pass the proper defaults along, otherwise the parent classes will use their own, inappropriate one.
gl/data/gl_setup.cpp:430:11: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
gl/data/gl_setup.cpp:527:19: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
gl/data/gl_setup.cpp:542:19: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
nodebuild.cpp:1056:63: warning: format specifies type 'ptrdiff_t' (aka 'long') but the argument has type 'int' [-Wformat]
p_glnodes.cpp:379:50: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
p_saveg.cpp:381:18: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
p_scroll.cpp:532:11: warning: comparison of integers of different signs: 'int' and 'unsigned int' [-Wsign-compare]
p_setup.cpp:2304:43: warning: format specifies type 'int' but the argument has type 'unsigned long' [-Wformat]
p_setup.cpp:2302:12: warning: comparison of integers of different signs: 'int' and 'unsigned long' [-Wsign-compare]
scripting/codegeneration/codegen.cpp:8488:20: warning: comparison of integers of different signs: 'int' and 'size_t' (aka 'unsigned long') [-Wsign-compare]
scripting/codegeneration/codegen.cpp:8606:15: warning: comparison of integers of different signs: 'int' and 'size_t' (aka 'unsigned long') [-Wsign-compare]
No need to maintain these clunky meta class for one single property. The overhead the mere existence of this class creates is far more than 100 spawned ammo items would cost.
There is no need to serialize AAmmo::DropAmount, this value has no meaning on an already spawned item.
There are a few which require explicit native construction or destruction that need to be exported to the VM, e.g. FCheckPosition.
The VM cannot handle this directly, it needs two special functions to be attached to handle such elements.
- created script exports for all relevant functions with all integral types.
- created script side definitions for the underlying data types.
- added a void pointer type so that the prototype for the pointer array can use a generic type every pointer can be assigned to.
- disabled the Build map loader after finding out that it has been completely broken and nonfunctional for a long time. Since this has no real value it will probably removed entirely in an upcoming commit.
- made ModifyDamage calls iterative instead of recursive. With going through the VM they'd be too costly otherwise.
- small optimization: Detect empty VM functions right when entering the VM and shortcut them. This is to reduce the overhead of virtual placeholders, which in a few cases (e.g. CanCollideWith and ModifyDamage) can be called quite frequently.
If the calling code wants to recycle this it will have to pass a container variable to AActor::UnlinkFromWorld and AActor::LinkToWorld.
This was changed because keeping such data in a global variable is dangerous for a set of functions that can be called from a script.
Note that the scripted versions do not yet support saving of the touching_sectorlist.
Instead of trying to fix Simplify, which seems to be a lost cause, the ring list now gets unraveled into an array which is immune from this type of problem.
This can see some heavy use in iterators where saving several hundreds of function calls can be achieved. In these cases, using a function to do the job will become a significant time waster.
* It will now use #include, just like most other definition formats and can be mixed with regular definitions. However, due to how the Lemon-generated parser works this will not recursively pull in all files, but store them in a list and process them sequentially. Functionally this shouldn't make a difference, because ZScript is mostly order-independent - the only thing where order is important is native classes, but these are completely internal to zdoom.pk3 where proper order is observed.
Note that this only applies if both are in the same block. Just like in C++, it is perfectly legal to have the same variable name in two different nested scopes.
The original implementation just printed a mostly information-free message and then went on as if nothing has happened, making it ridiculously easy to write broken code and release it. Changed it to:
* Any VMAbortException will now terminate the game session and go back to the console.
* It will also print a VM stack trace with all open functions, including source file and line numbers pointing to the problem spots. For this the relevant information had to be added to the VMScriptFunction class.
An interesting effect here was that just throwing the exception object increased the VM's Exec function's stack size from 900 bytes to 70kb, because the compiler allocates a separate local buffer for every single instance of the exception object.
The obvious solution was to put this part into a subfunction so that it won't pollute the Exec function's own stack frame. Interesting side effect of this: Exec's stack requirement went down from 900 bytes to 600 bytes. This is still on the high side but already a lot better.
- fixed PARAM_ACTION_PROLOGUE to assign correct types to the implicit pointers. It gave the actual class to the wrong one, which until now did not matter because all functions were using 'Actor', regardless of actual class association.
- fixed the definition of IceChunk and removed some redundant code here. Since A_FreezeDeathChunks already calls SetState, which in turn calls the state's action function, there is no need to call it again explicitly.
- throw a useful exception when a VM abort occurs, the simple enum was incapable of reporting anything more than the barest minimum, which at least for array index out of bounds errors was insufficient.
The current exception mechanism is still insufficient. It really has to report a proper crash location and print a stack trace to the maximum extent possible. Instead it just prints a message and happily goes on. This is not a good solution.
Although this already helps a lot with the messed up code generated for comparisons it's not really a solution for this - it still needs a proper implementation to generate efficient code.
- fixed: When replacing a tentative class, the pointers in the morph objects were not replaced. Instead of adding more ReplaceClassRef methods I chose to integrate this part into the PointerSubstitution mechanism and delete ReplaceClassRef entirely. The code had some oversights anyway that would have caused problems, now that non-actors can be created.
It is utterly pointless to require every function that wants to make a VM call to allocate a new stack first. The allocation overhead doubles the time to set up the call.
With one stack, previously allocated memory can be reused. The only important thing is, if this ever gets used in a multithreaded environment to have the stack being declared as thread_local, although for ZDoom this is of no consequence.
- eliminated all cases where native code was calling other native code through the VM interface. After scriptifying the game code, only 5 places were left which were quickly eliminated. This was mostly to ensure that the native VM function parameters do not need to be propagated further than absolutely necessary.
- moved health items to their own file.
- scriptified ScoreItem and MapRevealer whose entire functionality was a small TryPickup method.
- fixed: bit fields in global variables were not correctly written.
This should conclude the inventory cleanup. It is now possible again to find things in there.
- Changed the glass shards so that they do not have to override FloorBounceMissile. It was the only place where this was virtually overridden and provided little usefulness.
- made 'out' variables work.
- fixed virtual call handling for HandlePickup.
- added a String class to allow attaching methods to the builtin string type. This works by checking if the left side of the member accessor is a string and just replacing the tyoe in this one place, all the rest is automatic.
- merged the FrontBlock searcher for the Bloodscourge into RoughMonsterSearch. This also fixes the bug that the searcher was not initialized properly for the MageBoss.
A few notes:
* this accesses the lines array in sector_t which effectively is a pointer to an array of pointers - a type the parser can not represent. The compiler has no problems with it, so for now it is defined internally.
* array sizes were limited to 65536 entries because the 'bound' instruction only existed as an immediate version with no provisions for larger values. For the static map arrays 65536 is not sufficient so now there are alternative instructions for these cases.
* despite the above, at the moment there is no proper bounds checking for arrays that have no fixed size. To do this, a lot more work is needed. The type system as-is is not prepared for such a scenario.
- fixed: Assignment from a readonly to a read-allowed pointer must be an error.
- made GetDefaultByType a builtin so that it can do proper type assignment to the result, which for a function would be problematic in this case, even if automatic type deduction was implemented. Since this returns the class defaults which are not a real object, the result cannot be subjected to a type cast.
- error out if a type cast of a readonly pointer is attempted.
- fixed: FxBooleanNot could clobber a local variable because it used the source register to manipulate the result.
- fixed issues with the refactoring of the recent commits. This one starts again.
- added builtins for TextureID.
Note about builtins: Currently they are just hacked into the compiler backend. They really should be made part of the respective types to keep matters clean and allow more widespread use of builtins to create more efficient code.
Two reasons for this:
1. if this has to be routed through the VM each recursion will cost 1000 bytes of stack space which simply is not good.
2. having the virtual function only care about the item itself but not the entire inventory chain is a lot less error prone for scripting.
Since the scripting interface needs a separate caller function anyway this seemed like a good time to change it. The same will be done for the other chained inventory handlers as well.
- allow switch/case with names.
- fixed break jump target handling for switch/case. This only worked when the break was in the outermost compound statement, those in inner ones were missed.
Interesting tidbit: The damage calculation in P_MinotaurSlam had been incorrect for the Heretic version since the friendly Hexen Dark Servant was added, but nobody ever noticed in 14 years...
- Took the opportunity and fixed the logic for the Skull Rod's rain spawner. The old code which was part of the 3D floor submission was unable to work with portals at all. The new approach no longer tries to hide the dead projectile in the ceiling, it leaves it where it is and changes a few flags, so that its z-position can be used as reference to get the actual ceiling. This works for line portals, but for sector portals still requires some changes to sector_t::NextHighestCeilingAt to work, but at least this can be made to work unlike the old code.
- added names for the player-related translations to A_SetTranslation.
- fixed: Failure to resolve a function argument was checked for, too late.
- made the parameter for A_SetTranslation a name instead of a string, because it is more efficient. We do not need full strings here.
Needless to say, this is simply too volatile and would require constant active maintenance, not to mention a huge amount of work up front to get going.
It also hid a nasty problem with the Destroy method. Due to the way the garbage collector works, Destroy cannot be exposed to scripts as-is. It may be called from scripts but it may not be overridden from scripts because the garbage collector can call this function after all data needed for calling a scripted override has already been destroyed because if that data is also being collected there is no guarantee that proper order of destruction is observed. So for now Destroy is just a normal native method to scripted classes
- scriptified all Effect functions of Fastprojectile's children
- implemented access to class meta data.
- added a VM instruction to retrieve the class metadata, to eliminate the overhead of the function call that would otherwise be needed.
- made GetClass() a builtin so that it can use the new instruction
Important note about this commit: Scriptifying CFlameMissile::Effect revealed a problem with the virtual function interface: In order to work, this needs to be explicitly enabled for each single native class that may be used as a base for a scripted class. Needless to say, this will end up way too much work, as there are over 100 native classes, excluding those which will be scriptified. But in order to fix the problem this partially broken state needs to be committed first.
- fixed: FxAssignSelf did not the correct number of registers for vector operations.
- fixed a few asserts in vector2 instructions.
- turned the virtual AActor::HitFloor method into a flag MF7_SMASHABLE. The only use of this function was to kill Hexen's pottery when they hit the floor, and this looks like something that can be exposed to modders less clumsily.
This prevented any kind of error check on them.
Unfortunately, due to backwards compatibility needs, on DECORATE the missing class may not be fatal so a workaround had to be added to clear those bogus pointers later if they are discovered to be broken.
For ZScript, though, this will result in a compile error, which was the intention behind this change.
- added new VARF_Transient flag so that the decision whether to serialize a field does not depend solely on its native status. It may actually make a lot of sense to use the auto-serializer for native fields, too, as this would eliminate a lot of maintenance code.
- defined (u)int8/16 as aliases to the byte and short types (Can't we not just get rid of this naming convention already...?)
- exporting the fields of Actor revealed a few name clashes between them and some global types, so Actor.Sector was renamed to CurSector and Actor.Inventory was renamed to Actor.Inv.
- refactored state bitfield members into a flag word because the address of a bitfield cannot be taken, making such variables inaccessible to scripts.
- actually use PNativeStruct for representing native structs defined in a script.
- fixed flag CVAR access. As it turned out, OP_LBIT is a bit messy to set up properly when accessing integers that may or may not be big endian, so it now uses a shift and bit masking to do its work.
- used the SpawnPlayerMissile call in A_FireBFG to test named arguments.
- fixed: The code generator had no good safeguards for exceeding the maximum amount of registers.
All there was was a handful of pitiful asserts which in production code do nothing at all but generate broken output.
Even worse, the VM was hardwired to at most 255 constants per type per function by storing the constant count in a byte! This has been extended to 65535, but since many instructions only have a byte available for the constant index, a workaround had to be added to do a two-instruction setup if larger indices are needed.
- added new VM instructions to access the constant tables with a variable index.
- refactored VMFunctionBuilder's constant tables so that they are not limited to one entry per value. While this works fine for single values, it makes it impossible to store constant arrays in here.
- implemented method calls from struct instances.
- optimized disassembly of VM call instructions to print the function's name at the end where it is more visible and does not need to be truncated. Also use the printable name for script functions here.
- implemented multiple-return-value assignment. Due to some grammar conflicts the originally intended Lua-inspired syntax of 'a, b = Function()' could not be done, so it's '[a, b] = Function()'
- made APlayerPawn::PlayAttacking(2) virtual script functions so that mods have better control over player animations. Note that these have no native base so they skip the templated interface for managing virtual functions.
- added a new type 'NativeStruct'. This will be used for types that cannot be instantiated, and is also needed to cleanly handle many internal types that only can exist as reference.
- changed Dehacked weapon function lookup to check the symbol table instead of directly referencing the VM functions. Once scriptified these pointers will no longer be available.
- removed all special ATAGs from the VM. While well intentioned any pointer tagged with them is basically unusable because it'd trigger asserts all over the place.
- scriptified A_Punch for testing pass-by-reference parameters and stack variables.
- gave OP_CONCAT some sane semantics. The way this was defined, by specifying the source operands as a range of registers instead of a pair like everything else made it completely useless for the task at hand.
- changed formatting for floats to %.5f which for normal output in a game makes more sense. For special cases there should be a special formatting function for ints and floats that can do more specialized conversions.
- make the pointer to string cast a bit more useful by using the actual object's type rather than 'Object' which can be a great asset when debugging.
- fixed a few bad asserts.
- fixed code generation for using local variables as array index. This must use a different register for the array element offset because the original register may not be overwritten.
- instead add a list of SpecialInits to VMScriptFunction so this can be done transparently when setting up and popping the stack frame. The only drawback is that this requires permanent allocation of stack objects for the entire lifetime of a function but this is a relatively small tradeoff for significantly reduced maintenance work throughout.
- removed most #include "vm.h", because nearly all files already pull this in through dobject.h.
- fixed several places in the code generator that did not consider locked registers for local variables: array indices, abs and floating point builtin functions.
- added some debug aids to the bounds opcode. Just triggering an exception here which loses all relevant info is perfectly useless in a debug situation.
- added a DActorIterator class.
- fixed: It was not possible to have functions of the same name in two different classes because the name they were searched for was not qualified by the class. Changed so that the class name is included now, but to avoid renaming several hundreds of functions all at once, if the search fails, it will repeat with 'Actor' as class name.
This commit contains preparations for scriptifying Hexen's Dragon, but that doesn't work yet so it's not included.
- preserve a state's source line information for the postprocessing phase so that the checker can output more useful information.
- added missing check for weapon psprites to DPSprite::SetState.
- added state usage specifiers to Actor and Inventory. The states in these classes must be set to full access so that any existing mod can link to them.
This appears to be the only case where an actor was set to a state owned by a completely unrelated actor which would present some problems with state owner checking in AActor::SetState, so let's better get rid of it ASAP.
I believe the only reason this wasn't changed when all actors were exported 8 years ago was that old binary DEHSUPP lump.
This could cause problems with functions that take states as parameters but use them to set them internally instead of passing them through the A_Jump interface back to the caller, like A_Chase or A_LookEx.
This required some quite significant refactoring because the entire state resolution logic had been baked into the compiler which turned out to be a major maintenance problem.
Fixed this by adding a new builtin type 'statelabel'. This is an opaque identifier representing a state, with the actual data either directly encoded into the number for single label state or an index into a state information table.
The state resolution is now the task of the called function as it should always have remained. Note, that this required giving back the 'action' qualifier to most state jumping functions.
- refactored most A_Jump checkers to a two stage setup with a pure checker that returns a boolean and a scripted A_Jump wrapper, for some simpler checks the checker function was entirely omitted and calculated inline in the A_Jump function. It is strongly recommended to use the boolean checkers unless using an inline function invocation in a state as they lead to vastly clearer code and offer more flexibility.
- let Min() and Max() use the OP_MIN and OP_MAX opcodes. Although these were present, these function were implemented using some grossly inefficient branching tests.
- the DECORATE 'state' cast kludge will now actually call ResolveState because a state label is not a state and needs conversion.
Now even in DECORATE it is possible to report most cases in which user variables are accessed from non-item states as an error. This will report all states which can be traced by a direct link from a special state label. It will not find states that only get used via A_Jump etc.
- since we now look for fields in the base class first the restriction check is no longer needed as it was for a different mode of looking up the identifier.
- In DECORATE we do not need to bother with non-action functions. Non-action functions cannot be defined from DECORATE so there's no need to check if we are in one.
- make the warning for unsafe access a debug message because this can affect legitimate code.
- fixed creation of direct function invocations on a state line. In order to receive the implicit arguments this needs to be wrapped into a compound statement so that the local variable getter works.
The proper setup for such classes was only done in CreateDerivedClass, but not in FindClassTentative itself. This extends CreateDerivedClass to allow it to create a class without fully initializing it.
Syntax-wise I chose to make it as strict as possible to reduce the chance of errors: Virtual base functions must be declared with the 'virtual' keyword, and overrides in child classes with the 'override' keyword. This way any mismatch in parameters that otherwise would cause silent failure will outright produce a compile error.
- made 'DamageMultiply' an actor property and moved the initialization of ConversationRoot to the property handler for the compiler to get this stuff out of the type classes.
- consolidate default initialization into one function which performs all the required setup. The original implementation did this when adding the fields but that cannot work because at that time no defaults have been created yet.
- fixed: When deriving a class the child class's defaults also must initialize the copied parent fields with special initialization. This part was completely missing.
- removed DECORATE code for parsing native classes because it's no longer needed.
- Since the number of small allocations here is extremely high this will help a lot to prevent fragmentation and since most nodes are collected up front and this is done when no large resources are being loaded it won't cause heap spikes.
let Emit methods delete FxExpression arrays when they are done.
- For some reason the deletion process does not work 100%, there are always some nodes left behind and so far I haven't found them. This ensures that these arrays do not live any longer than needed.
- fixed: The state cast hack for DECORATE could not properly create state constants.
Instead they were passed to FxRuntimeStateIndex without resolving them to something constant. This adds proper handling of constant indices within that class.
* use the function build list instead of the function to pass the info. The function is permanent so not the best place for compile-time info.
* pass along the current state index which is needed to calculate the target state.
- made some tests about calling script code from native functions.
* scriptified A_SkullAttack to have something to test
* changed the A_SkullAttack call in A_PainShootSkull.
* use a macro to declare the function pointer. Using local static variable init directly results in hideous code for the need of being thread-safe (which, even if the engine was made multithreaded is not needed here.)
* Importsnt node here: Apparently passing an actor pointer to the VMValue constructor results in the void * version being called, not the DObject * version.
- fixed: The state index comparison against 0 was broken.
- fixed: Resolving codegen nodes must set the strictness flag per function so that ZSCRIPT and DECORATE are done properly.
This bypasses a declaration in the script in favor of a simpler implementation. In order to work it is always necessary to have an offset table to map the variables to, but doing it fully on the native side only requires adding the type to the declaration.
- fixed: divisions wasted one register for each operation due to a double allocation.
- changed math operations to use less registers. There was a well-intended change to allocate the destination first, but the better approach is to first allocate the operands and free then before allocating the destination register.
- added support for global variables to the code generator - not the compiler, though. For the handful of entries this is needed for it may just as well be done manually. So far FLevelLocals level is the only one being exported.
- fixed: The VM disassembler truncated 64 bit pointers to 15 digits because the output buffer was too small.
- resolve entire FxSequences instead of aborting on the first failed entry. This allows to output all errors at once.
* Revert "Modify CMPJMP to produce more compact code (as far as VC++ is concerned, anyway)"
This reverts commit 6ff973a06b.
This modification did not work and broke the comparisons. Actually this had three problems:
* the asserts checked the wrong instruction
* the mask was not applied to regular comparisons.
* incrementing PC before testing does not work because 'test' references the PC.
- exported thinker iterator and drop item chain to scripting. Unlike its native counterpart the script-side iterator is wrapped into a DObject to allow proper handling for memory management.
- fixed: The VMFunctionBuilder only distinguished between member and action functions but failed on static ones.
- fixed: FxAssign did not add all needed type casts. Except for purely numeric types it will now wrap the expression in an FxTypeCast. Numeric handling remains unchanged for both performance reasons and not altering semantics for DECORATE.
- exported all internal flags as variables to scripting. They still cannot be used in an actor definition.
- make ATAG_STATE the same as ATAG_GENERIC. Since state pointers exist as actual variables they can take both values which on occasion can trigger some asserts.
- gave PClass a bExported flag, so that scripts cannot see purely internal classes. Especially the types like PInt can cause problems.
Todo: we need readonly references to safely expose the actor defaults. Right now some badly behaving code could overwrite them.
- fixed: FxMinusSign trashed local variables that were used with negation.
- fixed: FxConditional only handled ints and floats, but not pointers and strings.
- fixed: A 'no states in non-actors' error was triggered, even for classes without any states.
